阅读说明：本技术 将自主行驶功能扩展到新地域 (Extending autonomous driving functionality to new territories ) 是由 A·C·雷什卡 C·麦格雷戈 于 2020-03-19 设计创作，主要内容包括：自主车辆使用准确和详细的地图进行导航。将自主车辆的功能扩展到新的(例如未示于地图中的)地域可以包括：确定新地域的可行驶表面区段,并将各区段与来自已制图地域的各区段或各区段类别进行比较。与来自已制图地域的各区段相似的新地域的各区段可以被识别为是潜在可导航的。自主车辆可以经由被指示为可导航的那些区段来行进通过新地域。此外,在行进通过新地域期间,可以使用自主车辆上的传感器收集数据以制图地域的附加部分和/或确认新地域中的行驶能力。自主车辆的功能可以基于各区段彼此的相似程度而受到限制。(Autonomous vehicles navigate using accurate and detailed maps. Extending the functionality of the autonomous vehicle to a new zone (e.g., not shown in the map) may include: the navigable surface segments of the new zone are determined and compared to the segments or segment categories from the mapped zone. Segments of the new zone that are similar to segments from the mapped zone may be identified as potentially navigable. The autonomous vehicle may travel through the new zone via those zones indicated as navigable. Further, during travel through the new territory, sensors on the autonomous vehicle may be used to collect data to map additional portions of the territory and/or to confirm travel capabilities in the new territory. The functionality of the autonomous vehicle may be limited based on how similar the segments are to one another.)

1. A system, comprising:

one or more processors; and

one or more non-transitory computer-readable media storing instructions executable by the one or more processors, wherein the instructions program the one or more processors to perform acts comprising:

receiving first map data comprising information about a first travelable surface in a first geographic region and one or more traffic features associated with the first travelable surface, the one or more traffic features comprising one or more of a number of lanes, a lane type, lane geometry, or traffic control information;

determining a plurality of first segments of a first travelable surface based at least in part on the one or more traffic characteristics;

determining, for a first segment of the plurality of first segments and based at least in part on the one or more traffic characteristics, a plurality of first segment parameters;

receiving second map data associated with a second geographic zone navigable by the autonomous vehicle, the second map data including information about a plurality of second segments of a second travelable surface in the second geographic zone;

determining a similarity metric indicative of a similarity between a first zone and one or more second zones of the plurality of second zones based, at least in part, on the plurality of first zone parameters and second map data; and is

Determining that the autonomous vehicle is able to navigate the first segment based at least in part on the similarity metric.

2. The system of claim 1, wherein the similarity metric is based at least in part on the one or more traffic characteristics associated with the first segment and one or more second traffic characteristics associated with the second segment, and

wherein the acts further comprise controlling the autonomous vehicle to traverse the second segment based at least in part on the similarity metric.

3. The system of claim 1 or 2, wherein:

the plurality of second segments includes a plurality of representative segments, and

determining a similarity metric includes comparing the one or more traffic characteristics with one or more second traffic characteristics associated with a second segment,

the acts further include:

receiving travel data associated with an autonomous vehicle traveling on a second segment; and

controlling the autonomous vehicle to traverse the first segment based at least in part on the travel data.

4. The system of claim 3, wherein:

the first section comprises a road section having a plurality of lanes,

determining a similarity metric is based at least in part on comparing a subset of the plurality of lanes to lane combinations in the second map data, and

determining that the autonomous vehicle is able to navigate the first section includes determining that the autonomous vehicle is able to navigate a portion of the first section that includes the subset of the plurality of lanes.

5. A computer-implemented method, comprising:

receiving first map data comprising information about a first travelable surface in a first geographic region;

determining a plurality of first segments of a first travelable surface, a first segment of the plurality of first segments being associated with one or more first parameters;

receiving second map data associated with a second geographic zone, the second map data including information about a plurality of second segments of a second travelable surface navigable by the autonomous vehicle, a second segment of the plurality of second segments associated with one or more second parameters;

determining a similarity metric indicative of a similarity between the first section and the second section based at least in part on the one or more first parameters and one or more second parameters;

determining zones of a first drivable surface navigable by the autonomous vehicle based, at least in part, on the similarity metric; and

and controlling the autonomous vehicle to run on each region.

6. The computer-implemented method of claim 5, wherein the one or more first parameters include at least one of a number of lanes, a lane type, a width, a length, a grade, a curvature, or traffic control information associated with the first segment.

7. The computer-implemented method of claim 5 or 6, further comprising:

determining second regions of the first travelable surface associated with regions to be avoided by the autonomous vehicle based at least in part on the similarity measure being below a threshold similarity; and

generating updated map data comprising a first travelable surface for the respective zones and the respective second zones.

8. The computer-implemented method of any of claims 5 to 7, wherein:

the plurality of second zones include a subset of all zones of the second map data indicating representative zones of the second map data, and

determining similarity is based at least in part on comparing respective values of the one or more first section parameters to the one or more of the second parameters.

9. The computer-implemented method of claim 8, further comprising:

determining that a maximum similarity metric between the first section and any one of the plurality of second sections is less than a threshold similarity; and

at least one of updating the plurality of second sections or creating a new representative section based at least in part on the maximum similarity metric being less than a threshold similarity.

10. The computer-implemented method of claim 8 or 9, further comprising:

determining one or more limits for controlling the second vehicle based at least in part on the similarity metric, the one or more limits including at least one of a maximum speed limit or a subset of the maneuvers.

11. A computer-implemented method as in any of claims 5 to 10, wherein determining the plurality of first segments of a first travelable surface comprises: clustering portions of a first travelable surface based at least in part on the one or more first parameters, including: at least one of a speed limit, a width of at least a portion of the lanes, a number of lanes, an untravelable zone, an inclination, a curvature, or a type of allowed travel on or adjacent to a lane in the road section.

12. The computer-implemented method of any of claims 5 to 11, wherein the first section comprises an intersection and the one or more first parameters comprise at least one of: the number of road sections connected at an intersection, the angle between road sections connected at an intersection, traffic control information at the end point of road sections at an intersection, or information on a proxy intersection at an intersection.

13. The computer-implemented method of any of claims 5-12, wherein one of the plurality of travelable surface sections comprises a road section having a plurality of lanes, and a first of the plurality of travelable surface sections comprises a subset of the plurality of lanes, and

determining the respective regions of the first travelable surface includes determining that the autonomous vehicle is able to navigate the subset of the plurality of lanes and that the autonomous vehicle is not able to navigate the respective lanes of the plurality of lanes other than the subset.

14. The computer-implemented method of any of claims 5 to 13, further comprising:

receiving travel data associated with a second segment; and

a driving maneuver is associated with the first segment based at least in part on the similarity score and the driving data, the driving maneuver including at least one of a maximum speed limit or an identification of one or more maneuvers associated with the driving.

15. One or more non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform the method of any one of claims 5-14.

Background

Vehicles are increasingly being used to supplement or replace manual functions with automatic controls. Semi-autonomous vehicles may provide assistance to the driver for certain functions, such as collision avoidance and braking, while fully autonomous vehicles may reduce all passengers to passive participants since all passengers may be directed to the destination. Often, such enhanced automation requires detailed knowledge of the vehicle environment (e.g., in the form of a detailed map). However, for unpatterned terrain (e.g., areas where autonomous vehicles have not been subjected to autonomous driving tests), acquiring details and understanding of the environment can be very expensive and time consuming, often requiring thousands or more hours and/or thousands or more miles.

Drawings

Fig. 1 illustrates an example of extending autonomous driving functionality to a new territory in accordance with aspects of the present application.

FIG. 2 includes a text and visual flow diagram to illustrate an exemplary method for extending autonomous driving functionality to a new territory.

FIG. 3 includes a text and visual flow diagram to illustrate another exemplary method for extending autonomous driving functionality to a new territory.

Fig. 4 is a block diagram illustrating an exemplary vehicle and remote computing system for extending autonomous driving functionality to a new geographic area, in accordance with aspects of the present application.

Fig. 5 is a flow diagram representing one or more processes for extending autonomous travel functionality to a new zone and traveling the new zone in accordance with aspects of the present application.

Fig. 6 is a flow diagram representing one or more processes for classifying road segments for extending autonomous driving functionality to a new territory in accordance with aspects of the present application.

Detailed Description

The following detailed description is directed to systems and processes for controlling an autonomous vehicle in an environment that includes a new environment in which the autonomous vehicle has not previously traveled. Unlike conventional automobiles, some autonomous vehicles may not have onboard navigation controls that are easily manually controlled, such as steering wheels, transmission controls, acceleration controls, and/or brake controls. Accordingly, such autonomous vehicles may require detailed maps of regions that may include details about the surfaces that may be traveled (e.g., range, grade, road surface type, etc.), details about lane and street configurations, details about traffic regulations, and so forth. Generating such complex maps may require a large amount of data about the territory, and acquiring such data typically requires a large number of manual trips through the territory. Furthermore, even if complex maps are generated, preparing for autonomous vehicles may require additional training of vehicles in the newly mapped regions. The techniques described herein may reduce the time and/or effort required to extend autonomous driving to a new area. For example, the present specification provides systems and methods for evaluating new (e.g., previously unpatterned for autonomous travel) geographic zones and for controlling autonomous vehicles in such zones. The techniques described herein may also update and/or generate new map data for previously unverified territories. Such map data may be used to facilitate control of autonomous vehicles in those previously unpatterned areas so that the vehicles may obtain sensor data regarding those areas to further update the map data.

In some examples, the techniques described herein may evaluate a new zone by comparing available map data for the new zone with map data of zones in which the vehicle has been operating autonomously and/or additional vehicle data, e.g., for potential navigation by the autonomous vehicle. For example, an autonomous vehicle may be interested in traveling at certain locations, such as a city, a portion of a city, etc. (e.g., this may be based on the ability to exhibit certain maneuvers, the ability to safely operate, the ability to recognize various scenarios that may occur in this portion, etc.) and proficiency may be based at least in part on the availability of detailed map data. For example, such map data may be acquired, generated, supplemented and/or updated through careful investigation of the venue, including using sensor data captured by vehicle-mounted sensors as those vehicles traverse the travelable surface in the area. For example, detailed map data of navigable (by autonomous vehicle) terrain may include information about the physical extent and arrangement of drivable surfaces. Such information may include the length and width of streets, the layout of intersections, ground hardness levels, the location and arrangement of buildings, fire hydrants or other fixtures, and other elements. The map data may also include information about features and conditions that affect travel on the travelable surface. For example, map data may include information about speed limits, crosswalks, traffic lights, street signs, etc., and the effect of such elements on travel in the environment. In at least some examples, such map data may further include weather, climate, and/or other environmental features (average lighting, average precipitation, average temperature, any snow or ice, etc.) that may autonomously affect travel in such regions.

As used herein, "verified map data," "data for mapped territories," and/or "data for verified territories" may refer to map data associated with a geographic area or venue in which an autonomous vehicle is verified or otherwise configured to operate (e.g., autonomously). Similarly, a "mapped territory" or an "authenticated territory" may refer to those geographic areas or locales. In contrast, "map data for a new territory," "unverified map data," and similar terms may refer to map data associated with a geographic area or locale in which the autonomous vehicle is unverified and/or insufficient to permit safe and/or legitimate operation of the autonomous vehicle based thereon. Similarly, "unpatterned territory," "new territory," and "unverified territory" may refer to such geographic areas or venues. Thus, map data may be used for "unpatterned", "new", or "unverified" territories, but such map data may not be sufficient for autonomous travel.

In some examples, the techniques described herein may parse map data (e.g., verified map data and/or map data of a new zone) into a plurality of segments of a travelable surface. For example, techniques of the present disclosure may generate sections from map data. In some cases, a segment may include: intersection sections (e.g., intersections, etc.) or connected road sections (e.g., the extent of the roads between intersections). The system described herein may also associate data with each of the individual sections. For example, the intersection zone may include an intersection type (e.g., a junction road, a "T" road, a roundabout, etc.); the number of roads that meet at the intersection; the relative locations of those roads (e.g., the angles between roads meeting at an intersection); information about traffic control signals at the intersection; and/or other features. The data associated with the road section being connected may include the number of lanes, the width of those lanes, the direction of travel of each of the lanes, the identity of the parking lane, the speed limit of the road section, and/or other characteristics.

In an example, the techniques described herein may compare segments of the travelable surface of the new zone (e.g., identified in the map data for the new zone) to segments of the travelable surface of the verified zone (e.g., identified in the verified map data). In embodiments, map data for the new territory may be obtained from a plurality of sources (including, but not limited to, third party vendors). In some examples, such map data may include 2D map data, 3D map data, satellite images, and/or other types of map data. In examples, any map data that provides information about travelable surfaces in a new territory may be used.

In some examples described herein, various metrics may be determined to define how well an autonomous vehicle is traveling in such existing zones. By way of non-limiting example, such metrics may include the number of occurrences over a period of time that require teleoperational intervention, passenger comfort (e.g., bump, shake, etc.), the ability of the vehicle to fully navigate the segment, the percentage of scenarios that the vehicle can safely overcome (e.g., if the segment includes an intersection with an unprotected left side, the vehicle cannot navigate), and so forth. Such metrics may be used with map data to indicate drivable segments and the capability levels associated therewith.

In some examples, the techniques described herein may group sections of traversable surfaces in a verified zone (e.g., according to a section classification or section paradigm). By way of non-limiting example, all intersection segments that meet a range of metrics or attributes may be grouped together (e.g., using k-means, evaluating weighted distances between segment parameters (e.g., euclidean distances), or otherwise clustering such segments based on individual segment parameters). For example, the autonomous vehicle may be expected to behave identically (and/or may have indeed acted) in each of the various intersections within the paradigm (stereotype). Similarly, the connected road segment paradigms may also be grouped according to one or more similarities. By way of non-limiting example, all two-lane road segments having a speed limit in the range of 10 miles per hour may be associated with the same paradigm. In some examples, the use of a paradigm may reduce the number of times similar segments are determined to be compared.

In some examples, the techniques described herein may compare sections of a new zone to sections (or canonical classifications) of verified zones. For example, the comparison may determine whether and to what extent each section in the new territory is similar to one or more navigable sections from the mapped territory. In some examples, the similarity may be expressed as a similarity measure, score, or other value. In some examples, segments in the new territory having a similarity metric, value, or score equal to or above a threshold may be identified as (potentially) navigable by the autonomous vehicle. In contrast, segments in the new territory having a similarity metric, score, or value below a threshold may be identified as not navigable by the autonomous vehicle. In some examples, various driving characteristics may be associated with each new zone. As a non-limiting example, full operation under all conditions may be associated with a zone, while single direction of travel, weather and/or environmental condition limited travel, etc. may be added to autonomously travel in such a zone. Such driving characteristics may be generally referred to as a strategy (e.g., a projected set of actions and/or limits on such actions, such as, but not limited to, maximum speed limits, avoidance of vehicles exceeding double stops, avoidance of vehicles exhibiting additional maneuvers, etc.).

The techniques described herein may also generate updated or new map data based on the indication of each drivable and each undrivable segment in the new region. For example, map data associated with various segments may be marked, annotated, or otherwise appended or modified to indicate that the segments may or may not be drivable (or otherwise have limited autonomous functionality). In some examples, the map data may be uploaded or otherwise made accessible to the autonomous vehicle. Techniques described herein may include controlling a vehicle to travel in a new region using updated or new map data.

Also, in examples, navigation of the vehicle may be evaluated as the autonomous vehicle is controlled according to the new map data. For example, the techniques may determine whether the vehicle is operating within predetermined criteria, which may be based at least in part on one or more of safety considerations, legitimacy considerations, passenger comfort considerations, and the like. In some examples, information that the vehicle is not operating properly in the section determined to be drivable or navigable may update the process of determining the similarity of the section to the verified section. For example, the techniques described herein may determine differences between properties of a section in a new region and a verified section or paradigm. In some examples, the differences may be used to update the paradigm and/or create a new paradigm.

The techniques discussed herein may improve the functionality of an autonomous vehicle in a number of ways. For example, the techniques discussed herein may facilitate travel in unpatterned areas and may map those unpatterned areas. Moreover, the techniques described herein may improve the functionality of a computer, for example, by reducing the amount of processing required to evaluate and/or map a new region. The techniques described herein may also navigate the area (e.g., via an autonomous vehicle) before generating and/or otherwise obtaining a detailed map of the entire territory. These and other improvements to the functionality of the autonomous vehicle, the computing system associated with the autonomous vehicle, and/or the user experience are discussed herein.

The techniques described herein may be implemented in a number of ways. Exemplary embodiments are provided below with reference to fig. 1-6. Although discussed in the context of an autonomous vehicle, the methods, apparatus, and systems described herein may be applied to a variety of systems and are not limited to autonomous vehicles. In another example, the techniques may be used in a robotic, aeronautical, and/or nautical context, for example, to extend the navigational coverage of the device.

Fig. 1 is a schematic diagram 100 of aspects of the present application. More specifically, fig. 1 shows an evaluation system 102 for determining whether an unverified (for autonomous navigation) area may be navigable and/or which portions of the unverified area are navigable even in the absence of detailed map data of the type and nature of the area that is typically required by the vehicle control system of the autonomous vehicle to navigate. As shown, the evaluation system 102 may be configured to receive or otherwise access map data for a new zone 104 (e.g., an "unverified zone"), and to receive or otherwise access verified map data 106. For example, a "new zone" may be a geographic area where operation of the autonomous vehicle has not been verified. For example, the new, unverified, or unpatterned zone may be a zone or zone adjacent to an already mapped zone (e.g., extending an existing travelable area or geofence for the autonomous vehicle), or may be a completely new geographic area (e.g., a different city). As further detailed herein, the map data for the new zone 104 may take any form of data from which aspects of the travelable surface in the new zone may be collected. For example, the map data for the new zone 104 may include any two-dimensional or three-dimensional map data that provides information about the new zone, including, but not limited to, map data available from third parties, such as satellite images, hand-drawn maps, and the like. In contrast, verified map data 106 may be map data for one or more geographic areas in which operation of the autonomous vehicle has been verified. The validated map data 106 may include detailed maps used by one or more autonomous vehicles for travel in an area, for example, using simultaneous localization and mapping (SLAM) technology. For example, the verified map data 106 may include a 3D mesh of an environment having a travelable surface and information regarding aspects of the travelable surface, including but not limited to range of the travelable surface, road marking information, traffic control information, and the like. In at least some examples, the map data for the new zone 104 and/or the verified map data 106 may include simulated map data. In such examples, approximate road parameters (grade, width, etc.) and/or actual road parameters (speed limit, number of lanes, etc.) may be used to generate such composite segments that approximate real-world road segments. For example, the simulated map data may be used to test aspects of the embodiments described herein and/or aspects of the autonomous vehicle. Thus, as used herein, a new zone and/or an unverified zone may refer to a geographic area for which detailed map data (e.g., sufficient to permit autonomous vehicle operation) is not available. In contrast, a verified zone and/or a mapped zone may refer to a geographic area where map data and/or travel data exists (e.g., verified map data 106), which may be used by an autonomous vehicle to navigate the area, e.g., autonomously.

The evaluation system 102 may use the map data for the new zone 104 and the verified map data 106 to generate an evaluation of the new zone. The evaluation 108 may identify non-drivable segments or regions 110 and drivable segments or regions 112 in the new territory. In some examples, evaluation system 102 may determine a similarity or similarity point of each road segment in the new territory to each road segment in the mapped territory to determine each non-drivable area 110 and each drivable area 112. For example, each non-drivable region 110 may be a region that the evaluation system 102 determines is unlikely (e.g., equal to or below a certain similarity metric threshold) to be navigable by an autonomous vehicle, while the drivable regions 112 may be regions that the evaluation system 102 determines is likely (e.g., equal to or above a certain similarity metric threshold) to be navigable by an autonomous vehicle.

More specifically, the evaluation system 102 can include one or more computing devices 114 configured to receive and/or otherwise access the map data for the new zone 104 and the verified map data 106. In general, and in some cases, the computing device(s) 114 may include various components to perform various processes and/or operations on the map data for the new zone 106 and the verified map data 108. For example, computing device(s) 114 may include a map partitioning component 116, a segment analysis component, and an evaluation/charting component 120.

The map partition component 116 may include functionality to identify segments of map data corresponding to discrete portions of the travelable surface in the geographic area. In some cases, the discrete portions may include road intersections and connected road sections. A road intersection may be a junction of two or more connected road sections, and a connected road section may be a portion of a road extending between two intersections. In other examples, the discrete portions may be sections of a road intersection and/or connected road sections. For example, the map-partitioning component 116 may identify individual lanes or groupings of two or more lanes in a road intersection or connected road section. Similarly, the map partition component 116 may include functionality to identify portions of an intersection (e.g., an intersection of two segments at a four-way intersection). It will be appreciated that the travelable surface may comprise a plurality of physical portions and the segments should not be limited to those explicitly discussed herein. In some examples, a map segment may define location, range, and semantic information associated with the map segment. As further described herein, the map partition component 116 can include functionality to identify sections in the map data for the new zone 104 and in the verified map data 106.

The map partition component 116 can implement a number of techniques to identify sections in the map data for the new zone 104 and/or the verified map data 106. By way of non-limiting example, when the map data includes image data, the map partition component 116 can implement feature recognition techniques (including, without limitation, edge detection techniques) to determine the extent of the drivable surface and/or lane markings on the drivable surface. In other examples, the map partition component 116 can implement partitioning and/or classification techniques on the map data, e.g., to identify portions of the sensor data associated with different classifications. In some examples, the drivable surface may be a classification identified using semantic partitioning. In at least some examples, various cluster-based algorithms may be used to identify certain lane segments (k-means, k-medoids, DBSCAN, etc.) based on any one or more of the various road parameters. Other techniques may also be used.

The map partition component 116 can also associate features or attributes (e.g., non-physical features/attributes) with the identified physical segments. For example, the map partition component 116 may also associate traffic control information with portions of the travelable surface associated with each segment. By way of non-limiting example, the traffic control information may include information about speed limits, about direction of travel, about traffic control devices (e.g., stop signs, yield signs, traffic lights). The additional traffic control information may include information on crosswalks, parking zones, no-parking zones, bicycle lanes, HOV information, and the like. In embodiments, map partition component 116 may identify physical travelable surface segments and associate any and all information that may affect safe and legitimate travel in such segments.

Further, in at least some examples, map partition component 116 may use previously recorded travel data from one or more vehicles associated with each segment to provide an autonomy level associated with each segment. As non-limiting examples, various zones may be identified as having full autonomy, limited autonomy (speed limits, maneuver limits, etc.), and not full autonomy in all environmental conditions.

The segment analysis component 118 can include functionality for comparing segments identified by the map partition component 116 and determining similarities therebetween. For example, the section analysis component 118 can identify that sections from the map data for the new zone 104 are similar or identical to sections from the verified map data 106. In some examples, the section analysis component 118 can compare individuals from sections of map data of the new zone 104 with individuals from sections of the verified map data 106 to find similarities. Such analysis may include determining a distance (e.g., a euclidean distance) between a feature vector associated with a segment and one or more clusters or segments identified in map partition component 116. According to a simple, non-limiting example, when the map segmentation component 116 identifies a relatively straight, two-lane, connected road segment with a 30 mph speed limit in the map data for the new zone 104, the segment analysis component 118 may look for a relatively straight, two-lane, connected road segment with a 30 mph speed limit in the verified map data 106. In this simple example, the measured properties (e.g., straightness of the road section, number of lanes, and speed limit) may be substantially the same. However, in other examples, the similarity may be less accurate. For example, the new territory may include a two-lane connected road segment with a curvature and a 40 mph speed limit. In the examples described herein, section analysis component 118 may determine a similarity metric that quantifies the similarity of a section in the new region and one or more sections in the verified region.

In some examples, the similarity metric may take into account multiple attributes and determine that several of those attributes are the same and/or similar between the compared sections. Some example attributes may include physical attributes or traffic control attributes. The physical attributes may include, for example, range (e.g., length, width), number of lanes, curvature, grade, road type (e.g., paved road, gravel road, dirt road, market road, unmarked road), presence or absence of shoulders, presence or absence of center isolation strips, and the like. For example, the traffic control attributes may include information about speed limits, direction of travel, traffic control elements (e.g., traffic lights, stop signs, yield signs), crosswalks, bike lanes, HOV rules, turn rules (e.g., inhibit left turns, inhibit right turns), and so forth. In some cases, the similarity measure or similarity score may be binary, such as 1 or 0, match or no match, drivable or undrivable. Of course, in other examples, such scores may be continuous and indicate a degree of similarity of one zone to another, and a lower level of similarity may be associated with disabling certain travel features (e.g., traveling beyond a certain speed limit, performing certain maneuvers (e.g., autonomously overriding a double parked vehicle, etc.)). In some examples, the similarity metric may indicate that there is no match when certain attributes are not the same in the two compared sections, regardless of whether other attributes are the same or similar. By way of non-limiting example, a two-lane section of a road having one lane dedicated to travel in one direction and another lane dedicated to travel in the opposite direction does not match a two-lane section of a road having two lanes dedicated to travel in the same direction. In contrast, other attributes may be different, but the sections may still be matched. By way of non-limiting example, two road segments that are actually the same may be matched except that the speed limit is 5 miles per hour difference. In this example, the similarity metric may take into account the degree of difference between the two attributes. For example, a 5 mph speed limit change may be more similar than a 20 mph speed limit change. In at least one example, the similarity metric can assign a value (e.g., a numerical value representative of such similarity), and the section analysis component 118 can generate a composite or other score that includes similar different attributes.

The segment analysis component 118 can also consider information about each segment paradigm. For example, the paradigm may define parameter values or ranges of parameter values, which define the road segment type. In some examples, the paradigm may be embodied by a representative section (e.g., a section that includes attributes/parameters/ranges associated with the paradigm). The segment analysis component 118 can determine various road segments that are similar to or include such parameters/ranges. For example, the paradigms may combine all segments that are similar enough that the predicted (or proven) autonomous vehicle behavior is similar. This similarity in behavior may be observed, for example, in hours and/or miles of travel over a road segment. In some cases, the segment analysis component may execute a machine learning model to group similar road segments into and/or update various paradigms. Thus, in examples of the present disclosure, the segment analysis component 118 can compare various road segments from the map data for the new zone 104 to the paradigm determined from and/or included in the verified map data.

The evaluation/charting component 120 can include functionality to evaluate the results of the segment analysis component 118 and update the map data for the new zone 104 to identify the non-drivable zones 110 and the drivable zones 112 (and/or limitations on driving capabilities). For example, evaluation/charting component 120 may determine, e.g., from the similarity metrics determined by section analysis component 118, that sections of the new zone are substantially similar to sections for which the autonomous vehicle has verified (e.g., from verified map data 106) for autonomous travel. In at least some examples, similar levels of autonomy corresponding to the original zone may be associated with such new zones. In examples, sections of map data for the new zone 104 that have no analogs in the verified map data 106 may be marked or otherwise indicated as not being verified for travel. Similarly, sections of map data for the new zone 104 for which there is a similarity in the verified map data 106 may be marked or otherwise indicated as verified (or potentially verified) for travel. As described above, in at least some examples, a limited amount of autonomy (or other limitation on functionality) may be associated with the one or more segments based on the determined similar segments. In some cases, the evaluation/mapping component 120 may generate new map data and provide updated (or new) map data to the autonomous vehicle, for example, by supplementing the map data for the new zone 104 with an indication of the drivable zone 112, for example, to equip the autonomous vehicle to operate in the drivable zone 112, such as to exclude the undrivable zone 116.

The evaluation/charting component 120 may also determine an overall score or coverage for the new zone. For example, once each of the sections in the map data for the new zone 104 are marked as non-drivable or drivable, the evaluation/mapping component 120 can determine the overall coverage of the new zone. In some examples, the overall score or coverage may correspond to a percentage of the drivable area in the new zone indicated as containing the drivable zone 112. Other assessment and/or scoring metrics may also be used. For example, the evaluation/charting component 120 can consider additional information about the new zone. For example, in the context of a ride service, it may be desirable to determine whether the drivable zone 112 includes a pick-up/drop-off location. The evaluation/mapping component 120 may determine a percentage of each section of the map data for the new zone 104 that also includes exposed curves, lanes, parking lots, or other attributes that may facilitate getting on/off the vehicle. In an additional example, the evaluation/charting component 120 can determine a metric that identifies how many new zones are connected. For example, determining drivable zones 112 in a new region to allow travel through a majority of the zones may have a relatively higher score than determining drivable zones 112 are far from each other (e.g., have little or no connected roads).

Thus, according to the techniques described herein, the vehicle control system will behave in the new zone how it can be determined using the map data for the verified zone and the map data for the new zone. Even though the map data for the new zone 104 may not be as detailed or specific as the verified map data 106, the evaluation system 102 may provide information regarding the potential for autonomous vehicles operating in the zone. In some embodiments, evaluation system 102 may help determine whether to provide autonomous vehicle service to a new zone and/or which new zone or zones to enter.

FIG. 2 depicts a pictorial flow diagram of an exemplary process 200 for identifying and/or associating autonomous features with travelable surfaces in a new territory. For example, process 200 may be implemented by one or more components of evaluation system 102, although process 200 is not limited to use with evaluation system 102 and evaluation system 102 may perform other operations and/or processes in addition to process 200.

At operation 202, the process 200 may include receiving map data for a new zone. For example, the new region may be an extension of the mapped region or a completely new region, e.g., corresponding to a new city or place. An example 204 of the join operation 202 includes a graphical representation of a map 206 of the new territory. For example, the map 206 may represent map data for the new zone 104, as discussed above in connection with fig. 1. In examples of the present disclosure, the map data may be any number of data types including representative information of a place. In particular, the techniques described herein may use any map data from which information regarding travelable surfaces in a new territory may be determined. In some examples, such maps 204 may be created based on collected data from one or more sensors of the vehicle collected during the mapping. In some examples, the map 204 may be determined based on simulated (synthetic) data generated by a user or a combination of simulated portions and sensor data. In any such example, additional parameters corresponding to roads may be associated with various portions of the map (speed limit, stop sign, traffic signal, lane indicator, crosswalk, etc.).

At operation 208, the process 200 may include identifying travelable surface segments in the new territory. In the example in connection with operation 208, multiple zones of the map 206 are specified. For example, four intersections 210(1), 210(2), 210(3), 210(4) (collectively, intersections 210) and four connected road sections 212(1), 212(2), 212(3), 212(4) (collectively, connected roads 212) (each additional road section is shown, but not labeled) are listed in the figure. In this example, each of the intersections 210 is an intersection of two or more connected roads 212, and each of the connected roads 212 extends between two of the intersections 210. More specifically, the first intersection 210(1) and the third intersection 210(3) may be "T" -shaped intersections at which three of the respective connected road sections meet. In the specific example of the first junction 210(1), the three junction sections include a first connected road 212(1), a fourth connected road 2212(4), and an unmarked connected road section that is a continuation of the first connected road 212 (1). The second intersection 210(2) and the fourth intersection 210(4) may be different types of intersections. For example, the second intersection 210(2) is shown as a four-way intersection, and the fourth intersection 210(4) is shown as a roundabout or a roundabout. Of course, these are only examples of intersections.

It may be appreciated that the map 206 may provide only a small portion of a (larger) area. Thus, in some embodiments, there may be hundreds and potentially thousands or more junctions 210 and a similar number of connected roads 212. Further, each intersection may have its own characteristics and/or each connected link 212 may have its own characteristics. By way of non-limiting example, while the first intersection 210(1) and the third intersection 210(3) may appear the same, in practice they may be completely different. For example, the traffic control at each of the intersections 210(1), 210(3) may be different. In one example, the intersection 210(1) may be a three-way waypoint and the connected road segment 212(3) may be a main road, such that vehicles traveling straight through the intersection 210(3) from the connected road segment 212(3) may not have a waypoint, while traffic on the fourth connected road 212(4) must stop before entering the intersection embodied by the intersection 210 (4). Of course, these are only examples. In other examples, one or more intersections may include other features, such as crosswalks, turning lanes, yield signs, traffic lights, and the like. In addition, the fourth connected road section 212(4) may be a one-way road in either direction. Similarly, other factors may also affect traffic control at each intersection 210. For example, the speed limit on the road comprising the respective connected road sections 212 may vary for each of the individual junctions 210. Each intersection 210 may also have additional variations. By way of non-limiting example, while each connected roadway 212 is generally illustrated as being disposed at a right angle with respect to adjacent sections, the angle between the sections of roadway may vary. Further, the individuals in each junction 210 may include crosswalks, bike paths, physical obstacles, or other features that may affect navigation therethrough. Other examples are provided herein.

Similarly, although the referenced connected road segments 212 appear substantially the same in the map 206, each may have unique features or attributes. By way of non-limiting example, each section may have a different curvature, a different slope, a different speed limit, a different number of lanes or lane configurations (e.g., width, length, or other range, different directions of travel of the lanes, etc.). Also in examples, each road segment 212 may include a parking lane, crosswalk, no-parking zone, central isolation zone or other physical barrier, bridge, tunnel, adjacent sidewalk, speed bump, bike lane, or other feature. Other examples are provided herein.

At operation 214, the process 200 may include comparing the segments to mapped segments and/or paradigms from the reference map data. For example, each of the segments identified by operation 208 (e.g., each of the intersections 210 and each of the connected road segments 212) may be compared to cartographic data from the navigable terrain, such as the verified map data 106 discussed above. The first intersection 210(1) from the map 206 and associated section data 216 is presented in conjunction with the illustration of operation 214. By way of non-limiting example, section data 216 may include any features, attributes, and/or variations that may be quantified for intersection 210(1), including but not limited to the cases discussed above. The comparison may be based on determining a similarity measure, as defined herein.

An example charted intersection 218 is also shown in connection with the example of operation 214. For example, the mapped intersection 218 may include intersection segments identified in a navigable area (such as in the validated map data 106). In some examples, each intersection in the verified map data 106 may be stored separately with associated data, much like the section data 216. In this example, a first mapped intersection 218(1) shows a four-way intersection and associated data, a second mapped intersection 218(2) shows a first three-way intersection and accompanying intersection data, and a third mapped intersection 218(3) shows a second three-way intersection and accompanying intersection data. It should be noted that the various mapped intersections 218 may be from the territory that the autonomous vehicle is verified to travel, and thus it may be known to some extent that the autonomous vehicle may navigate each of the various mapped intersections 218. Thus, if it is determined at operation 214 that intersection 210(1) is the same as one of the mapped intersections 218, it may be reasonably expected that the autonomous vehicle may navigate intersection 210(1) in the new territory. In the example of join operation 214, junction 210(1) may be determined to be similar and/or identical to third charted junction 218 (3). As described above, this may be done based on determining a similarity score or metric associated with one or more features/parameters of the two sections.

In some embodiments, operation 214 may compare intersection 210(1) to each mapped intersection 218. However, in other embodiments, each of the various mapped intersections 218 may represent a class or set of similar intersections. For example, the autonomous vehicle may use the same functionality to navigate a four-way intersection through a main street and two minor streets, such as whether the minor street has a stop sign, a yield sign, and/or a flashing red light and/or whatever speed limit is on the main street. In such cases, the mapped intersection 218 may comprise an intersection or intersection paradigm. For example, the intersection paradigm may include one or more ranges or types of criteria within a single parameter. In the example just given, a main street and a four-way intersection of two minor streets can be grouped together, regardless of the traffic control devices used at the intersections of the minor streets with the main street (or the traffic control devices are devices on a list of similar devices) and/or regardless of the speed limit on the main street (or if the speed limit is within a certain range of the speed limits). In some examples, using the paradigm may reduce computational load, for example, by comparing individual sections in a new region to fewer mapped sections (e.g., mapped intersections 218). For example, using a paradigm or classification can reduce thousands of junctions to tens of paradigms. In at least some examples where a clustering algorithm is used to cluster sections, the mean, median, or mode of each cluster may be used as an indication of the characteristics of the section, with the variance indicating how close each cluster is.

At operation 220, the process 200 may include identifying each travelable surface in the new territory. For example, the process 200 may include identifying (e.g., based on the comparison(s) at operation 214) those sections in the map 206 for which analogs are present in the verified map data. In some examples, those zones where the analog does exist may be indicated as (potential) drivable zones 222 (e.g., may be navigable by an autonomous vehicle) and those zones where the analog does not exist may be indicated as non-drivable zones 224 (e.g., may not be navigable by an autonomous vehicle). In the example of the combining operation 220, the second connection road 212(2), the third connection road 212(3), and the fourth intersection 210(4) are indicated as non-drivable zones 224, while all remaining segments are indicated as including drivable zones 222. In at least some examples, such determinations may not be so binary (e.g., drivable versus undrivable). In such examples, various operations (e.g., maximum speed limits, actions that may be performed, times and/or conditions in which autonomous travel is allowed, etc.) may be limited based at least in part on the similarity metric. As a non-limiting example, similar driving characteristics may be tolerated if the newly mapped road segment is within 95% of the characteristics of the existing segments. In another case, if the new section has a 0.8 similarity measure of the paradigm road section, a maximum speed limit may be set that is a fraction of the posted speed limit for the new section.

In some embodiments, the autonomous vehicle may be provided with updated map data identifying the drivable terrain 222 so that the autonomous vehicle may operate in the drivable terrain 222 subject to any restrictions imposed. In some examples, the autonomous vehicle may operate with exclusion of information regarding, for example, avoidance of the no-travel zone 224. However, in still other examples, the indication that the zone or segment is not drivable may prompt additional processing. For example, fig. 3 shows an example of determining whether segments indicated as non-drivable by process 200 or some other process are still navigable by the autonomous vehicle. More specifically, fig. 3 depicts a pictorial flow diagram of an exemplary process 300 for identifying travelable subsections (e.g., subsets of lanes) in a new territory. For example, the process 300 may be implemented by one or more components of the evaluation system 102, although the process 300 is not limited to use with the evaluation system 102 and the evaluation system 102 may perform operations and/or processes in addition to the process 300 or in place of the process 300.

At operation 302, the process 300 may include receiving information about a road segment. For example, a road segment may be an intersection segment or a connected road segment, such as discussed herein. In some examples, the road segment may be a segment that is marked or otherwise identified as non-drivable, e.g., according to process 200 described above or some other process. In such examples, the segment may be non-drivable in that it does not clearly correspond to each verified map segment, as described above in connection with fig. 2. In example 304 of join operation 302, the road segments are embodied as connected road segments 306. For example, the connected road segment 306 may be one of the connected roads 212 discussed above in connection with the process 200. More specifically, the connected road segments 306 may include a plurality of lanes including a first lane 308, a second lane 310, a third lane 312, a fourth lane 314, a fifth lane 316, a sixth lane 318, and a parking lane including a plurality of parking spaces 320. Still more particularly, a centerline may separate the third lane 312 from the fourth lane 314, and lane markings 324 may be provided between the other lanes. For example, the first lane 308 may be a turn lane (e.g., via which traffic may turn right from the road segment 306), and the second lane 310 and the third lane 312 may be for traffic traveling in the same direction, e.g., toward right to left in the illustration. In contrast, the fourth lane 314, the fifth lane 316, and the sixth lane 318 may facilitate travel in opposite directions, such as from left to right in the orientation illustrated. Parking spaces 320 may be arranged for parallel parking adjacent to sixth lane 318.

At operation 326, the process 300 may include identifying a subset of the lanes in the road section. For example, in an example in connection with operation 326, process 300 may identify a subset 328 of lanes that includes fifth lane 316, sixth lane 318, and that includes parking space 320. In various embodiments, a minimum of three lanes may be selected, for example, because safe travel of the autonomous vehicle may require knowledge of objects, obstacles, and/or the like in the lane in which the vehicle is traveling and in both adjacent lanes. In other examples, more or fewer lanes may be included in the subset. In a non-limiting example, only sixth lane 318 and the lane including parking space 320 may be used as a subset of the lanes. Further, while road segment 306 is a contiguous road segment, similar techniques may be used to identify segments or portions of junction segments. By way of non-limiting example, a meeting of two connected road segments at an intersection may be identified by operation 326 (e.g., excluding other connected roads that meet at the intersection). In at least some examples where at least one lane of a section is similar to another section, those lanes will be preferred over other lanes for travel and/or additional restrictions may be imposed on such other lanes. Various other subsets or groups of features within a segment are also contemplated, including various other features and various attributes described herein.

At operation 330, the process 300 may include comparing the identified lanes to verified lane combinations and/or paradigms. Operation 330 may be similar to operation 214 described above. For example, each of the lane combinations, including lane combination 328, may be compared to mapped data from a navigable terrain, such as map data 106 discussed above. The lane combination 328 and associated lane data 332 from the lane section 306 are demonstrated in conjunction with the illustration of operation 330. By way of non-limiting example, lane data 332 may include any features, attributes, and/or variants quantifiable for lane combinations 328, including but not limited to a direction of travel for each lane, each range of each lane, lane marking information, speed limit information, orientation of parking spaces 320, and/or other data associated with lane combinations 328.

The example in connection with operation 330 also shows verified lane data 334. For example, the verified lane data may include information about lane combinations identified in the navigable area (such as in the verified map data 106). In some examples, each lane combination in the verified map data 106 may be stored independently with associated data (e.g., lane data 332). Operation 330 may compare the lane data 332 associated with the lane combination 328 in the new region with each mapped lane combination (e.g., in verified lane data 334) to determine a correspondence between the lane combination 328 and the lane combination for which autonomy is verified. However, in other embodiments, the verified lane data 334 may include information for each category or each group of each similar lane combination. For example, an autonomous vehicle may use the same functionality to navigate at a speed limit of 65 miles per hour on the two rightmost lanes of a five-lane highway, just as it does at a speed limit of 70 miles per hour on the two rightmost lanes of an eight-lane highway. Thus, for example, the verified lane data 334 may include a lane paradigm including a two-lane configuration with a speed limit ranging from 65 miles per hour to 70 miles per hour. Of course, this is merely a simple example, and additional and/or different attributes, ranges, and/or the like may be used to develop the various paradigms. In some examples, using paradigms may reduce computational load, for example, by comparing lane combinations in a new territory to fewer mapped combinations. In at least some examples, such verified lane data 334 may include verified (actual, successful) travel behaviors of one or more autonomous vehicles. Such driving behavior and similarity scores may be used to impose one or more restrictions on driving in the newly indicated zones.

At operation 336, the process 300 may include identifying a subset of the total lanes as navigable. For example, the process 300 may include determining whether analogs are present in the verified map data for each combination of lanes (including combination 328). In some examples, those zones where the similar does exist may be indicated as drivable zones 338 (e.g., potentially navigable by an autonomous vehicle) and those zones where the similar does not exist may be indicated as non-drivable zones 340 (e.g., non-navigable by an autonomous vehicle). In the example of the combination operation 336, the fifth lane 316, the sixth lane 318, and the lane including the parking space 320 are indicated as a drivable zone 338, while the first lane 308, the second lane 310, the third lane 312, and the fourth lane 314 are indicated as a non-drivable zone 340.

For example, as described above, the road segment 306 may have been identified as a non-navigable road segment, for example because the map data for the navigable terrain may not include a road segment with the same or sufficiently similar characteristics that cause the road segment 306 to be identified as similar to a verified navigable segment. As a result of process 300, the techniques described herein may identify additional portions of a new (e.g., unverified) zone on which the autonomous vehicle may operate. In examples, the autonomous vehicle may be provided with information about the travelable zone 338 and may therefore only traverse the road segment 306 in that zone. In the illustrated example, the autonomous vehicle may travel only in fifth lane 316 or sixth lane 318, and/or may park or otherwise park in one of parking spaces 320. Thus, in the context of a ride service, a vehicle may provide (e.g., at parking space 320) an entry and/or exit service, even though verification of travel on all lanes of road segment 306 may not be possible. Further, as the vehicle navigates through the travelable zone 338, it may generate sensor data, for example using sensors mounted on the vehicle, to allow mapping of the non-travelable zone 340.

Fig. 4 depicts a block diagram of an exemplary system 400 for implementing the techniques described herein. In at least one example, the system 400 may include a vehicle 402, which is an autonomous vehicle, such as the autonomous vehicle described herein. The vehicle 402 may include a vehicle computing device 404, one or more sensor systems 406, one or more transmitters 408, one or more communication connections 410, at least one direct connection 412, and one or more drive modules 414.

The vehicle computing device 404 may include one or more processors 416 and memory 418 communicatively coupled with the one or more processors 416. In the example shown, vehicle 402 is an autonomous vehicle; however, the vehicle 402 may be any other type of vehicle, or any other system having at least an image capture device (e.g., a smartphone with a camera). In the example shown, the memory 418 of the vehicle computing device 404 stores a positioning component 420, a perception component 422, a planning component 424, one or more system controllers 426, and one or more maps 428. While depicted in fig. 4 as residing in memory 418 for purposes of illustration, it is contemplated that positioning component 420, perception component 422, planning component 424, the one or more system controllers 426 and the one or more maps 428 may additionally or alternatively be open access (e.g., stored remotely) to vehicle 402.

In at least one example, the positioning component 420 can include functionality to receive data from the sensor system(s) 406 to determine the location of the vehicle 402. For example, the location component 420 can include and/or request/receive a map of the environment, and can continuously determine the location of the autonomous vehicle within the map. In some cases, the location component 420 may utilize SLAM (simultaneous location and illustration) or CLAMS (simultaneous calibration, location and mapping) to receive image data, LIDAR data, radar data, IMU data, GPS data, wheel encoder data, etc., in turn, accurately determine the location of the autonomous vehicle. In some cases, the positioning component 420 may provide data to various components of the vehicle 402 to determine an initial position of the autonomous vehicle to generate a candidate trajectory as discussed herein.

In some cases, perception component 422 may include functionality to perform object detection, partitioning, and/or classification. In some examples, the perception component 422 may provide processed sensor data that indicates the presence of and/or classifies entities in the vicinity of the vehicle 402 as entity types (e.g., cars, pedestrians, riders, animals, buildings, trees, pavements, curbs, sidewalks, unknowns, etc.). In additional and/or alternative examples, sensing component 422 can provide processed sensor data indicative of one or more characteristics associated with the detected entity and/or the environment in which the entity is located. In some examples, the characteristics associated with the entity may include, but are not limited to, an x-position (global position), a y-position (global position), a z-position (global position), an orientation, an entity type (e.g., classification), a speed of the entity, a range (size) of the entity, and the like. The characteristics associated with the environment may include, but are not limited to, the presence of another entity in the environment, the status of another entity in the environment, the time of day, the day of the week, the season, weather conditions, indications of light and dark conditions, and the like.

In general, the planning component 424 may determine a path to be followed by the vehicle 402 to traverse the environment. For example, the planning component 424 can determine various routes, trajectories, and various levels of detail. For example, the planning component 424 may determine a route to travel from a first location (e.g., a current location) to a second location (e.g., a target location). For ease of discussion, the route may be a series of landmarks for traveling between two locations. By way of non-limiting example, each landmark includes a street, an intersection, Global Positioning System (GPS) coordinates, and the like. Additionally, the planning component 424 may generate instructions for guiding the autonomous vehicle along at least a portion of a route from the first location to the second location. In at least one example, the planning component 424 may determine how to direct an autonomous vehicle from a first landmark in a series of landmarks to a second landmark in the series of landmarks. In some examples, the instructions may be a trace or a portion of a trace. In some examples, multiple traces may be generated substantially simultaneously (e.g., within a technical tolerance) in accordance with a fallback level technique.

The system controller(s) 426 may be configured to control steering, propulsion, braking, safety, transmitters, communications, and other systems of the vehicle 402. The system controller(s) 426 may communicate with and/or control corresponding systems of the drive module(s) 414 and/or various other components of the vehicle 402.

The map(s) 428 may be used by the vehicle 402 to navigate within the environment. For purposes of discussion, a map may be any number of data structures modeled in two, three, or N dimensions that are capable of providing information about an environment, such as, but not limited to, a topology (such as an intersection), a street, a mountain, a road, terrain, and a general environment. In some cases, the map may include, but is not limited to: texture information (e.g., color information (e.g., RGB color information, Lab color information, HSV/HSL color information), etc.), intensity information (e.g., LIDAR information, RADAR information, etc.); spatial information (e.g., image data projected onto a mesh, independent "bins" (e.g., polygons associated with independent colors and/or intensities)), reflectivity information (e.g., specular reflectivity information, retroreflectivity information, BRDF information, BSSRDF information, etc.). In one example, the map may include a three-dimensional mesh generated using the techniques discussed herein. In some cases, the map may be stored in a tile format, such that individual tiles of the map represent discrete portions of the environment and may be loaded into working memory as needed. In at least one example, the map(s) 428 may include at least one map (e.g., an image and/or a grid) generated in accordance with the techniques discussed herein. For example, the map(s) 428 may include information regarding travel zones in the new environment, where the travel zones are determined according to the techniques described herein. In some cases, the map(s) 428 may include information only about drivable zones, while other embodiments may include map data for the entire zone, including map data for each undrivable zone. In some examples, the vehicle 402 may be controlled based at least in part on the map(s) 428. That is, the map 428 may be used in conjunction with the positioning component 420, the perception component 422, and/or the planning component 424 to determine the location of the vehicle 402, identify objects in the environment, and/or generate routes and/or trajectories to navigate within the environment. Further, data from sensor system(s) 406 and/or other data may be used to augment, supplement, and/or generate map data included in map(s) 428.

In some examples, map(s) 428 may be stored on remote computing device(s) (e.g., computing device(s) 432) that are accessible via network(s) 430. In some examples, multiple ones of map(s) 428 may be stored based on characteristics (e.g., type of entity, time of day, day of week, season of year, etc.), for example. Storing multiple maps 428 may have similar storage requirements, but increases the speed at which data in the heat map may be accessed.

In some cases, aspects of some or all of the components discussed herein may include models, algorithms, and/or machine learning algorithms. For example, in some cases, components in memory 418 (and/or memory 436 discussed below) may be implemented as a neural network.

As described herein, an exemplary neural network is a biologically motivated algorithm that passes input data through a series of connected layers to produce an output. Each layer in the neural network may also include another neural network, or may include any number of layers (whether convolutional or not). As can be appreciated in the context of the present disclosure, neural networks may utilize machine learning, which may refer to a broad class of algorithms in which outputs are generated based on learned parameters.

Although discussed in the context of a neural network, any type of machine learning may be used in accordance with the present disclosure. For example, the machine learning algorithms may include, but are not limited to, regression algorithms (e.g., Ordinary Least Squares Regression (OLSR), linear regression, logarithmic regression, stepwise regression, Multivariate Adaptive Regression Splines (MARS), local estimation scatter plot smoothing (lous), example-based algorithms (e.g., ridge regression, Least Absolute Shrinkage and Selection Operator (LASSO), elastic nets, Least Angle Regression (LARS)), decision tree algorithms (e.g., classification and regression trees (CART), iterative dichotomy 3(ID3), chi-square automated interaction detection (CHAID), decision trees, conditional decision trees), bayesian algorithms (e.g., naive, gaussian naive, polynomial naive bayes, average univariate estimator (AODE), bayes belief network (BNN), bayes networks), clustering algorithms (e.g., k-means, k-median, Expectation Maximization (EM) Hierarchical clustering), association rule learning algorithms (e.g., perceptron, backpropagation, jump networks, Radial Basis Function Networks (RBFN)), deep learning algorithms (e.g., Deep Boltzmann Machine (DBM), Deep Belief Networks (DBN), Convolutional Neural Networks (CNN), stacked autoencoders), dimension reduction algorithms (e.g., Principal Component Analysis (PCA), Principal Component Regression (PCR), Partial Least Squares Regression (PLSR), samming mapping, multidimensional scaling (MDS), projection pursuit, Linear Discriminant Analysis (LDA), Mixture Discriminant Analysis (MDA), Quadratic Discriminant Analysis (QDA), Flexible Discriminant Analysis (FDA)), integration algorithms (e.g., enhancement, bootstrap aggregation (MDA), AdaBoost, stacked generalization (mixture), gradient enhancer (GBM), gradient enhancement regression tree (GBRT), random forests), SVM (support vector machines), supervised learning, Unsupervised learning, semi-supervised learning, etc.

Other examples of architectures include neural networks such as ResNet70, ResNet101, VGG, DenseNet, PointNet, and the like.

In at least one example, sensor system(s) 406 may include LIDAR sensors, radar sensors, ultrasonic transducers, sonar sensors, position sensors (e.g., GPS, compass, etc.), inertial sensors (e.g., Inertial Measurement Unit (IMU), accelerometer, magnetometer, gyroscope, etc.), cameras (e.g., RGB cameras, IR cameras, intensity cameras, depth cameras, time-of-flight cameras, etc.), microphones, wheel encoders, environmental sensors (e.g., temperature sensors, humidity sensors, etc.). Sensor system(s) 406 may include multiple instances of each of these or other types of sensors. For example, the LIDAR sensors may include individual LIDAR sensors located at corners, front, rear, sides, and/or top of the vehicle 402. As another example, the camera sensor may include a plurality of cameras placed at various locations around the exterior and/or interior of the vehicle 402. The sensor system(s) 406 may provide input to the vehicle computing device 404. Additionally or alternatively, sensor system(s) 406 may transmit sensor data to the one or more computing devices via the one or more networks 430 at a particular frequency, after a predetermined period of time has elapsed, in near real-time, or the like.

The emitter(s) 408 may be configured to emit light and/or sound. The transmitter 408 in this example includes an internal audio and visual transmitter to communicate with the occupants of the vehicle 402. By way of example and not limitation, the internal transmitters may include speakers, lights, signs, display screens, touch screens, tactile transmitters (e.g., vibration and/or force feedback), mechanical actuators (e.g., seatbelt pretensioners, seat positioners, headrest positioners, etc.), and the like. The transmitter 408 in this example also includes an external transmitter. By way of example and not limitation, each external transmitter in this example may include a signal light for emitting a direction of travel signal or other indicator of vehicle action (e.g., indicator light, sign, array of lights, etc.) and/or one or more audio transmitters (e.g., speakers, array of speakers, horn, etc.) for audible communication with pedestrians or other nearby vehicles, one or more of which may include acoustic beam steering technology.

Communication connection(s) 410 may permit communication between vehicle 402 and one or more other local or remote computing devices, such as computing device 432. For example, the communication connection(s) 410 may facilitate communication with other local computing device(s) on the vehicle 402 and/or the drive module(s) 414. Also, communication connection(s) 410 may allow the vehicle to communicate with other nearby computing device(s) (e.g., other nearby vehicles, traffic lights, etc.). The communication connection(s) 410 also allow the vehicle 402 to communicate with remotely operated computing devices or other remote services.

Communication connection(s) 410 may include a physical and/or logical interface to connect the vehicle computing device 404 to another computing device or network, such as network(s) 430. For example, communication connection(s) 410 may permit Wi-Fi based communications (such as via frequencies defined by the IEEE 402.11 standard), short range wireless frequencies (such as) Cellular communication (e.g., 2G, 4G LTE, 5G, etc.), or any suitable wired or wireless communication protocol that allows the respective computing device to interface with the other computing device(s).

The drive module(s) 414 may include a number of vehicle systems including a high voltage battery, an electric machine that propels the vehicle, an inverter that converts direct current from the battery to alternating current for use by other vehicle systems, a steering system including a steering motor and a steering rack (which may be electric), a braking system including hydraulic or electric actuators, suspension systems including hydraulic and/or pneumatic components, stability control systems for distributing braking force to mitigate traction loss and maintain control, HVAC systems, lighting systems (e.g., illuminating exterior environments such as headlights/taillights to illuminate the vehicle), and one or more other systems (e.g., cooling systems, security systems, on-board charging systems, other electrical components such as DC/DC converters, high voltage connectors, high voltage cables, charging systems, charging ports, etc.). Additionally, the drive module(s) 414 may include a drive module controller that may receive and pre-process data from the sensor system(s) and control the operation of various vehicle systems. In some examples, the drive module controller may include one or more processors and memory communicatively coupled with the one or more processors. The memory may store one or more modules to perform various functions of the driver module(s) 414. In addition, the driver module(s) 414 also include one or more communication connections that enable the respective driver module to communicate with one or more other local or remote computing devices.

In some examples, the vehicle 402 may have a single drive module 414. In at least one example, if the vehicle 402 has multiple drive modules 414, the respective drive modules 414 may be positioned on opposite ends (e.g., front and rear, etc.) of the vehicle 402. In at least one example, the drive module(s) 414 can include one or more sensor systems to detect conditions of the drive module(s) 414 and/or the surroundings of the vehicle 402. By way of example and not limitation, the sensor system(s) may include one or more wheel encoders (e.g., rotary encoders) to sense rotation of the wheels of the drive module, inertial sensors (e.g., inertial measurement units, accelerometers, gyroscopes, magnetometers, etc.) to measure orientation and acceleration of the drive module, cameras or other image sensors, ultrasonic sensors to acoustically detect objects in the surroundings of the drive module, LIDAR sensors, radar sensors, etc. Some sensors, such as wheel encoders, may be unique to the drive module(s) 414. In some cases, the sensor system(s) on the drive module(s) 414 may overlap or supplement corresponding systems of the vehicle 402 (e.g., the sensor system(s) 406).

In at least one example, positioning component 420, perception component 422, and/or planning component 424 can process the sensor data as described above and can send their respective outputs to computing device(s) 432 over network(s) 430. In at least one example, positioning component 420, perception component 422, and/or planning component 424 can send their respective outputs to the one or more computing devices 432 at a particular frequency, after a predetermined period of time has elapsed, in near real-time, and/or the like. In some examples, the vehicle 402 may send raw sensor data to the computing device(s) 432. In other examples, the vehicle 402 may send the processed sensor data and/or a representation of the sensor data to the computing device(s) 432. In some examples, the vehicle 402 may transmit the sensor data to the computing device(s) 432 at a particular frequency, after a predetermined period of time has elapsed, in near real-time, and/or the like. In some cases, the vehicle 402 may send the sensor data (raw or processed) to the computing device(s) 432 as one or more log files.

Computing device 432 may receive sensor data (raw or processed) and may generate and/or update maps based on the sensor data. For example, the computing device(s) 432 may compare map data of, for example, a new zone in which the autonomous vehicle 402 is not verified for travel to verified map data from, for example, regions in which the autonomous vehicle 402 is verified for travel or otherwise configured to identify sections of the new zone in which the autonomous vehicle 402 may travel. For example, the computing device(s) 432 may generate updated map data that includes the navigable area determined and provide or otherwise make available the updated map data to the vehicle 402. Thus, the vehicle may be controlled to identify new (e.g., not previously charted for navigation), obscured geographic regions in the image, and may generate a textured 3D map without shading.

In at least some examples, computing device(s) 432 can include one or more processors 432 and memory 436 communicatively coupled with the one or more processors 434. The computing device(s) 432 may also include first map data 438 and second map data 440. The second map data 440 may also include paradigm data 442. Also in the illustrated example, the memory 436 of the computing device(s) 432 stores the map partition component 116, the segment analysis component 118, and the evaluation/mapping component 120. In at least one example, computing device(s) 432 can include some or all of the functionality of evaluation system 102 of fig. 1.

The map partition component 116 is described above in connection with FIG. 1. In general, the map-partitioning component may include a feature recognition component and/or a machine learning algorithm trained to recognize segments of the travelable surface. In some cases, the map partition component 116 may receive map data (such as the first map data 438 and/or the second map data 440) and analyze the data to determine segments, such as connected road segments and junction segments where the connected road segments meet. Also in various embodiments, the map-partitioning component 116 can identify sub-regions of each section, including but not limited to lane-combinations, portions of intersections, and the like.

The segment analysis component 118 is described above in connection with fig. 1. In general, the segment analysis component 118 can compare map segments from map data of the new zone (e.g., the first map data 438) to map segments from map data of the verified zone (e.g., the second map data 440). For example, the segment analysis component 118 can determine whether segments identified in the map data of the new region are similar or similar to segments from verified map data. As described herein, because the autonomous vehicle may successfully travel in the geographic territory corresponding to the verified map data, the autonomous vehicle may also successfully travel over sections of the new territory having similar or identical features, attributes, and/or configurations.

In some cases, the section analysis component 118 can also compare various map sections from the map data of the new territory to various paradigms of various section classifications. For example, and as described herein, paradigms (such as paradigms data 442) may be used to group sections where a vehicle may operate (or have proven to operate) in the same or similar manner. Further, the section analysis component 118 can generate and/or update the paradigm data 442. For example, the process 600 described herein includes an example of generating a new paradigm based on operation of the autonomous vehicle 402 in a drivable terrain.

Evaluation/charting component 120 is described above in connection with fig. 1. In general, evaluation/charting component 120 can aggregate or otherwise evaluate the results of the processing (e.g., comparison) by section analysis component 118. For example, the evaluation/mapping component 120 may mark or otherwise identify some portions of the new zone as drivable portions and additional portions of the new zone as undrivable portions. The evaluation/charting component 120 may also provide a score or other metric that generally indicates the driveability of the new zone and/or sections of the zone. For example, the score metric may be based at least in part on a percentage of the new territory expected to be drivable based on a comparison of sections of the new territory and sections in the mapped territory.

In the example of fig. 4, the first map data 438 may include data about the new zone, such as map data for the new zone 104. As described above, the first map data 438 may be any of a number of types of 2D, 3D, or other map data from which the features of the drivable surface can be obtained. In examples, the first map data may include satellite images, sensor-generated data, or other map data. The second map data 440 may be the verified map data 106 and may be similar or identical to some or all of the map(s) 428 described above. For example, the second map data 440 may be detailed map data associated with a zone in which the vehicle 402 may operate autonomously. As noted herein, the second map data 440 may include 2D or 3D data and may be used by the positioning component 420, the planning component 424, and/or other components of the vehicle 402 to navigate within the mapped territory. As also shown in FIG. 4, the second map data 440 may also include or be associated with paradigm data 442. The pattern data 442 may include information, such as parameters, values, and/or ranges, that may be used to characterize the map data for each similar segment, as described herein. In this example, first map data 438 and second map data 440 are shown as being associated with computing device(s) 432. In various embodiments, one or both of the first map data 438 and the second map data 440 may be stored in the memory 436 and/or accessed by the computing device(s) 432. By way of non-limiting example, some or all of the second map data 440 may be stored on the vehicle 402, for example, as the map(s) 428.

The processor(s) 416 of the vehicle 402 and the processor(s) 434 of the computing device(s) 432 may be any suitable processor capable of executing instructions to process data and perform operations as described herein. By way of example, and not limitation, the processor(s) 416, 434 may include one or more Central Processing Units (CPUs), Graphics Processing Units (GPUs), or any other device or portion of a device that can process electronic data to convert that electronic data into other electronic data that can be stored in registers and/or memory. In some examples, integrated circuits (e.g., ASICs, etc.), gate arrays (e.g., FPGAs, etc.), and other hardware devices may also be considered processors, provided they are configured to implement the coded instructions.

Memory 418 and memory 436 are examples of non-transitory computer-readable media. Memory 418 and memory 436 may store an operating system and one or more software applications, instructions, programs, and/or data to implement the methods described herein and the functions that contribute to the different systems. In various embodiments, the memory may be implemented using any suitable memory technology, such as Static Random Access Memory (SRAM), synchronous dynamic ram (sdram), non-volatile/flash type memory, or any other type of memory capable of storing information. The architectures, systems, and elements described herein may comprise many other logical, procedural, and physical components, and those shown in the figures are merely examples of what is discussed herein.

It should be noted that although fig. 4 is illustrated as a distributed system, in alternative examples, various components of vehicle 402 may be associated with computing device(s) 432 and/or various components of computing device(s) 432 may be associated with vehicle 402. That is, vehicle 402 may perform one or more functions associated with computing device(s) 432, and vice versa.

Fig. 2, 3, 5 and 6 illustrate exemplary processes according to embodiments of the present disclosure. The processes are illustrated as logical flow diagrams, each operation of which represents a sequence of operations that can be implemented in hardware, software, or a combination thereof. In the context of software, the operations represent computer-executable instructions stored on one or more non-transitory computer-readable storage media that, when executed by one or more processors, perform the recited operations. Generally, computer-executable instructions include routines, programs, objects, components, data structures, etc. that perform particular functions or implement particular abstract data types. The order in which the operations are described is not intended to be construed as a limitation, and any number of the described operations can be combined in any order and/or in parallel to implement a flow.

Fig. 5 depicts an exemplary process 500 for determining travelable segments of a new zone and controlling an autonomous vehicle in accordance with map data generated based on the travelable segments. For example, some or all of process 500 may be performed by one or more components in fig. 4, as described herein. For example, some or all of process 500 may be performed by computing device(s) 432 and/or autonomous vehicle 402.

At operation 502, the process 500 may include receiving map data for an unverified zone. For example, the unverified zone may be a new geographic zone in which it has not been verified whether the autonomous vehicle may travel. In some embodiments, the new region may be an extension of an existing geographic zone, for example to extend a geofence or other virtual boundary associated with the travelable zone. In embodiments, the map data received at operation 502 may be any type of map data from which the characteristics and/or details of the drivable surface may be determined. In some examples, the map data may include two-dimensional information and/or three-dimensional information about the new zone.

At operation 504, the process 500 may include receiving map data for the verified zone. For example, an autonomous vehicle (such as vehicle 402) may require extensive and detailed map data to navigate autonomously in the environment. For example, such map data may include three-dimensional mesh data for objects in a mapped territory, including, but not limited to, travelable surfaces, obstacles, buildings, curbs, and the like. Further, such map data may include any information and/or data that may affect travel over a drivable surface. For example, the map data for the verified zone received at operation 504 may include speed limit information, traffic control information, and/or any other information needed to safely and/or legally travel through the verified zone. In at least some examples, previous autonomous operations on such zones may be associated with the one or more zones.

At operation 506, the process 500 may include determining sections of the travelable surface in the unverified area. For example, the map partition component 116 can identify portions of the drivable surface in the unverified area. By way of non-limiting example, such a section may include intersections of connected roads and/or connected roads extending between intersections. In some embodiments, the segments may be a subset of intersections and/or connected road segments. In the example of fig. 3 described herein, a section may include a set of traffic lanes in connected road sections or at an intersection. In some cases, map partition component 116 may perform image processing techniques to identify a drivable surface and/or segments thereof. In at least some examples, such zones may be determined based on one or more clustering algorithms.

At operation 508, the process may include, for each section, determining whether the section is similar to a section or paradigm from the verified territory. For example, the section analysis component 118 can compare sections from the new zone (as determined at operation 506) to mapped sections from the verified zone. For example, the section analysis component 118 determines a similarity score or other metric. Such a similarity score may be, for example, a distance (e.g., a weighted euclidean distance) between one or more features/parameters associated with the two segments. In some examples, section analysis component 118 may determine whether values associated with certain features or attributes from sections of the new territory are consistent with those features/attributes in sections of charted data, as described herein.

If it is determined at operation 508 that the individual segment is similar to a segment or paradigm from the verified territory (e.g., the similarity score or metric meets or exceeds a threshold), at operation 510, the process 500 may identify the segment as potentially drivable. For example, as described herein, map data of a verified zone may include sections of a drivable zone that an autonomous vehicle may easily navigate. In at least some examples, the level of autonomy associated therewith may be based at least in part on the similarity score. As described in detail above, as the similarity scores differ, one or more functions of autonomy may be constrained relative to the new domain. Thus, when the section analysis component 118 determines that the sections in the new and verified regions are substantially similar, the process 500 may determine that the autonomous vehicle is also likely to successfully operate on similar sections of the new region.

In contrast, if it is determined at operation 508 that the individual segment is not similar to a segment or paradigm from a verified territory (e.g., the similarity score does not meet or exceed the threshold score), at operation 512, the process 500 may include identifying the segment as potentially undrivable. Thus, operations 508, 510, 512 may be used to determine whether segments in the new territory are drivable or non-drivable. In embodiments, the extent of the entire travelable surface may be resolved into discrete segments, each segment being identified as travelable or non-travelable.

At operation 514, the process 500 may include generating new (e.g., unverified zones) map data to include drivable and undrivable zones. For example, the map data of the unverified territory received at operation 502 may be supplemented to include a flag, identification, or other indication associated with each road segment indicating whether such segment is drivable or non-drivable, and/or any other limitation.

At operation 516, the process 500 may include receiving travel data for the verified zone. For example, since a segment in the unverified area is indicated as drivable based on the similarity of the segment to a segment in the verified area, driving data for controlling vehicles in the segment in the verified area may be helpful in navigating the similar, unverified segment. In some examples, the travel data may be limited to only those sections from the verified territory that are considered similar to sections in the unverified territory, although in other embodiments, more or less travel data may be received.

At operation 518, the process 500 may include generating a driving maneuver. For example, a driving strategy may be a set of rules, controls, limits, or other data that permits or otherwise informs of autonomous driving by a vehicle (such as vehicle 402). The driving strategy may be executed by the vehicle as it navigates over each potential drivable segment in the new territory. Thus, for example, a driving strategy may include information that assists the vehicle in traveling over potential driving zones and avoids non-driving zones. In some cases, a driving strategy may include all functions, controls, etc. for autonomously navigating in the verified zone. In other embodiments, the driving strategy may include information only about controlling autonomous vehicles in those zones that match zones in the new territory. Further, the driving strategy may be based at least in part on the degree of similarity. For example, for a section in a new territory that is deemed similar to but not exactly matching a section in a verified territory, the driving strategy may include information about limited functionality for traversing the section in the new territory. For example, a driving strategy may limit the speed limit at which the vehicle may traverse the pass through zone, may keep the vehicle from performing certain actions, and so on.

At operation 520, the process 500 may include controlling the vehicle to travel in a travelable region of the unverified region based on the travel strategy. For example, the map data generated at operation 514 may be uploaded to or made available to an autonomous vehicle (such as autonomous vehicle 402), and the autonomous vehicle 402 may use the map data associated with the drivable terrain to travel in the new terrain. The autonomous vehicle 402 may also use the travel data received at operation 516. As described herein, autonomous vehicles may be expected to operate in zones that are indicated as being drivable. Further, in some embodiments, the vehicle may collect sensor data as the vehicle travels through a new region in the drivable segment (e.g., using one or more sensor systems disposed on or otherwise associated with the vehicle), and such sensor data may be used to supplement the map (e.g., by obtaining information about regions indicated as non-drivable and/or obtaining additional, more detailed information about regions indicated as drivable). As also described herein, the functionality of the vehicle 402 may be limited (e.g., based on the similarity score and/or other parameters) in various sections of the unverified zone.

Fig. 6 shows an exemplary process 600 of verifying the driveability of a new zone and generating a paradigm or other segment classification. In some examples, process 600 may be performed, at least in part, by autonomous vehicle 402. However, some or all of the process 600 may be performed by other vehicles under the system, and the autonomous vehicle 402 is not limited to implementing the process 600.

At operation 602, the process 600 includes receiving information regarding potential travel sections of an unverified zone. For example, as just described with reference to fig. 5, the techniques described herein may be used to determine segments of a road or other travelable surface on which an autonomous vehicle is expected to operate (e.g., because the segments of the new territory and the paradigm or set associated with the verified map segment are substantially the same or consistent). In some cases, the map data for the unverified zone may be updated to identify those potential travelable segments.

At operation 604, the process 600 may include controlling the vehicle along each potential travel segment using controls associated with the verified zone. For example, an autonomous vehicle (such as vehicle 402) may include functions, rules, and/or other control logic that allow the vehicle to navigate verified (e.g., mapped) sections. In embodiments, these controls may be implemented at operation 604 to control the vehicle in zones determined to be similar to those from the mapped territory.

At operation 606, the process 600 may include determining whether the vehicle performs acceptably for each segment. For example, although the techniques described herein may identify zones in which an autonomous vehicle is expected to operate successfully, the operation of the vehicle on those zones may still require testing, e.g., to confirm that the operation was in fact successful. In some embodiments, the vehicle may generate information as it attempts to travel over portions indicated as being drivable. In some examples, successful navigation in a segment may include traversing from an origin or origin to a destination while traversing the segment. In other examples, the acceptable operation may include a time component, such as determining that the vehicle is traveling from a departure point to a destination at a particular time. Acceptable performance may also take into account whether the vehicle is in compliance with other driving parameters, including but not limited to safety parameters such as keeping a minimum distance from objects in the environment, keeping the time interval between leading vehicles, keeping the time to collision at or below a threshold time, or other parameters. In still further examples, parameters affecting occupant comfort may be evaluated to determine whether the vehicle was successful. For example, the amount and severity of acceleration and deceleration, the severity of maneuvers or turns, etc., as the vehicle travels through the segment may also be considered. In those examples where additional restrictions are imposed on travel, such restrictions may be gradually removed based on one or more successful travels over such zones (e.g., by increasing the maximum speed limit, allowing more complex maneuvers, etc.).

If it is determined at operation 606 that the vehicle performs acceptably on individual segments, at operation 608, the process 600 may include verifying the driveability of the segments. For example, if the vehicle is performing according to one or more expectations or requirements, a section of a new territory identified using the techniques described herein may be indicated as a verified section.

Alternatively, if it is determined at operation 606 that the vehicle is not performing acceptably on an individual segment, at operation 610, the process 600 may include determining a difference between the segment and a verified segment or range to which the segment is compared. For example, and as discussed herein, some embodiments may determine that segments are similar to each other based on their consistency with the paradigm and/or despite some differences in attributes. Operation 610 may identify these differences.

At operation 612, the process 600 may include updating the pattern data/identifying a new pattern. For example, when the vehicle is not expected to be present in a section identified or considered to be a drivable section, it may be determined that the difference between the section and the section or paradigm considered to be similar is significant. Thus, in some cases, section analysis component 118 can update the paradigm, for example, such that values or attributes associated with sections in the new territory are no longer included in the paradigm. In other examples, other (e.g., previously unaccounted for) attributes may also be considered. When such a new or different attribute is identified, a new paradigm may be generated to include different ranges or values associated with the attribute.

Thus, the functionality of an autonomous vehicle may be more easily extended to new geographic areas in accordance with the techniques described herein. For example, sections of map data may be compared to determine similarities between new and charted regions. Also in embodiments, information regarding different features and attributes other than those attributable to the sections of the travelable surface may also be considered as detailed above. By way of non-limiting example, the techniques described herein may also include filtering the new zone to determine a baseline compatibility of the autonomous vehicle with the new zone. For example, verified map data for a first city in a first climate and with a first driving attribute may be easier than respective regions with similar climate and attributes.

Exemplary clauses

A: an exemplary computer-implemented method, comprising: receiving first map data comprising information about a first travelable surface in a first geographic region and one or more traffic features associated with the first travelable surface, the one or more traffic features including one or more of a number of lanes, a lane type, lane geometry, or traffic control information; determining a plurality of first segments of a first travelable surface based at least in part on the one or more traffic characteristics; determining, for a first segment of the plurality of first segments and based at least in part on the one or more traffic characteristics, a plurality of first segment parameters; receiving second map data associated with a second geographic zone navigable by the autonomous vehicle, the second map data including information about a plurality of second segments of a second travelable surface in the second geographic zone; determining a similarity metric indicative of a similarity between the first zone and one or more of the plurality of second zones based at least in part on the plurality of first zone parameters and the second map data; and determine that the autonomous vehicle may navigate the first segment based at least in part on the similarity metric.

B: the computer-implemented method of example a, wherein the similarity metric is based at least in part on the one or more traffic characteristics associated with the first segment and one or more second traffic characteristics associated with the second segment, and wherein the method further comprises controlling the autonomous vehicle to traverse the second segment based at least in part on the similarity metric.

C: the computer-implemented method of example a or example B, wherein: the plurality of second segments includes a plurality of representative segments, and determining the similarity metric includes comparing the one or more traffic characteristics with one or more second traffic characteristics associated with the second segments, the method further comprising: receiving travel data associated with an autonomous vehicle traveling on a second segment; and controlling the autonomous vehicle to traverse the first segment based at least in part on the travel data.

D: the computer-implemented method of any one of examples a to C, wherein: the first section includes a road section having a plurality of lanes, determining the similarity metric is based at least in part on comparing a subset of the plurality of lanes to lane combinations in the second map data, and determining that the autonomous vehicle can navigate the first section includes determining that the autonomous vehicle can navigate a portion of the first section that includes the subset of the plurality of lanes.

E: an exemplary system, comprising: one or more processors; and one or more non-transitory computer-readable media storing instructions executable by the one or more processors, wherein the instructions program the one or more processors to perform acts comprising: receiving first map data comprising information about a first travelable surface in a first geographic region; determining a plurality of first segments of the first travelable surface, a first segment of the plurality of first segments being associated with one or more first parameters; receiving second map data associated with a second geographic zone, the second map data including information about a plurality of second segments of a second travelable surface navigable by the autonomous vehicle, a second segment of the plurality of second segments associated with one or more second parameters; determining a similarity metric indicative of a similarity between the first section and the second section based at least in part on the one or more first parameters and the one or more second parameters; determining zones of a first drivable surface navigable by the autonomous vehicle based, at least in part, on the similarity metric; and controlling the autonomous vehicle to travel over the respective territory.

F: the system of example E, wherein the one or more first parameters include at least one of a number of lanes, a lane type, a width, a length, a grade, a curvature, or traffic control information associated with the first segment.

G: the system of example E or example F, each action further comprising: determining second regions of the first drivable surface associated with regions to be avoided with the autonomous vehicle based, at least in part, on the similarity measure being below a threshold similarity; and generating updated map data comprising the first travelable surface for each zone and each second zone.

H: the system of any one of examples E to G, wherein: the plurality of second segments includes a subset of all segments of the second map data indicating representative segments of the second map data, and determining the similarity is based at least in part on comparing respective values of the one or more first segment parameters to the one or more of the second parameters.

I: the system of any one of examples E to H, each action further comprising: determining that a maximum similarity metric between the first segment and any one of the plurality of second segments is less than a threshold similarity; and at least one of updating the plurality of second segments or creating a new representative segment based at least in part on the maximum similarity metric being less than the threshold similarity.

J: the system of any one of examples E to I, each act further comprising: one or more limits for controlling the second vehicle are determined based at least in part on the similarity metric, the one or more limits including a maximum speed limit.

K: the system of any one of examples E to J, the determining the plurality of first segments of the first travelable surface comprising clustering the portions of the first travelable surface based at least in part on the one or more first parameters, the one or more first parameters including at least one of a speed limit, a width of at least a portion of a lane, a number of lanes, an untravelable region, an inclination, a curvature, or a type of allowed travel on or adjacent to each lane in the road segment.

L: the system of any one of examples E to K, wherein the first section comprises an intersection, and the one or more first parameters comprise a number of road sections connected at the intersection, an angle between road sections connected at the intersection, traffic control information at an end point of road sections at the intersection, or information about an agent intersection (agent intersections) at the intersection.

M: the system of any of examples E-L, wherein one of the plurality of travelable surface sections comprises a road section having a plurality of lanes, and a first section of each section comprises a subset of the plurality of lanes, and determining the regions of the first travelable surface comprises determining that the autonomous vehicle can navigate the subset of the plurality of lanes and that the autonomous vehicle cannot navigate the lanes of the plurality of lanes other than the subset.

N: the system of any one of examples E to M, each action further comprising: receiving travel data associated with a second segment; and associating the travel strategy with the first segment based at least in part on the similarity score and the travel data.

O: the system of any one of examples E to N, wherein the driving strategy includes a maximum speed limit.

P: one or more exemplary non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform operations comprising: receiving first map data comprising first information about a first travelable surface in a first geographic region; determining a plurality of first segments of a first travelable surface based at least in part on the first information; receiving second map data associated with a second geographic zone, the second map data including second information about a plurality of second segments of a second travelable surface in the second geographic zone navigable by the autonomous vehicle; for a first section of the plurality of first sections, determining one or more first section parameters for the first section; determining a similarity measure indicative of a similarity between the first zone and a second zone of the plurality of second zones using the second map data and the one or more first zone parameters; and determining a driving strategy for the autonomous vehicle with respect to the first zone based on the similarity metric.

Q: the non-transitory computer-readable medium of example P, wherein: determining that the plurality of first segments cluster portions of the first travelable surface based at least in part on the first information, determining that the plurality of second segments cluster portions of the second travelable surface based at least in part on the second information, and determining the similarity metric includes comparing a first portion of the first information associated with the first segment to a second portion of the second information associated with the second segment.

R: the non-transitory computer-readable medium of example P or example Q, wherein the driving strategy comprises: avoidance of travel, travel under a maximum speed limit, or avoidance of performance of one or more actions.

S: the non-transitory computer-readable medium of any one of examples P to R, wherein: the plurality of second segments includes one or more representative segments, a representative segment of the one or more representative segments is associated with one or more parameters and values, and determining the similarity metric is based at least in part on comparing respective values of the first parameter to respective values of one or more second parameters associated with the second segment.

T: the non-transitory computer-readable medium of any one of examples P to S, the acts further comprising: travel data associated with navigating the vehicle in the first segment is received, wherein the travel strategy is further based at least in part on the travel data.

Conclusion

Although one or more examples of the techniques described herein have been described, various modifications, additions, permutations and equivalents thereof are intended to be included within the scope of the claims described herein.

In the description of the examples, reference is made to the accompanying drawings which form a part hereof, and in which is shown by way of illustration specific examples of the claimed subject matter. It is to be understood that other examples may be used and that changes or substitutions, such as structural changes, may be made. Such examples, changes, or variations are not necessarily departures from the intended scope of the claimed subject matter. Although the steps herein may be shown in a certain order, in some cases the order may be changed so that certain inputs may be provided at different times or in a different order without changing the functionality of the systems and methods described. The disclosed processes may also be performed in a different order. In addition, various computations need not be performed in the order disclosed herein, and other examples using alternative orders of computation may be readily implemented. In addition to reordering, these calculations can be decomposed into sub-calculations with the same result.