Advanced boarding of passengers in autonomous vehicles
阅读说明:本技术 自动驾驶车辆中乘客的提前上车 (Advanced boarding of passengers in autonomous vehicles ) 是由 P.内梅克 R-R.休伯特 J.S.赫巴赫 M.L.陈 M.爱泼斯坦 S.潘迪特 J.W 于 2018-05-09 设计创作,主要内容包括:该技术涉及在车辆100到达接载位置(由标记770表示)之前主动寻找分派的乘客。例如,接收到识别接载位置的信息和用于认证所分派的乘客的客户端设备信息。从车辆的感知系统(172)接收传感器数据,识别车辆环境中的目标。当车辆在接载位置的预定距离内(由距离条772表示)时,尝试使用客户端设备信息来认证客户端设备(420、430)。当客户端设备已经被认证时,传感器数据被用于确定行人是否在车辆的第一阈值距离(D1)内。当行人(650、652、654、656)被确定为处于车辆的第一阈值距离之内时,车辆会在到达接载位置之前停止,以等待车辆的第一阈值距离之内的行人进入车辆。(This technique involves actively seeking the dispatched passenger before vehicle 100 arrives at the pickup location (represented by reference number 770). For example, information identifying the pickup location and client device information for authenticating the assigned passenger are received. Sensor data is received from a sensing system (172) of the vehicle identifying a target in the vehicle environment. When the vehicle is within a predetermined distance of the pickup location (represented by distance bar 772), an attempt is made to authenticate the client device (420, 430) using the client device information. When the client device has been authenticated, the sensor data is used to determine whether the pedestrian is within a first threshold distance (D1) of the vehicle. When a pedestrian (650, 652, 654, 656) is determined to be within the first threshold distance of the vehicle, the vehicle may stop before reaching the pickup location to wait for a pedestrian within the first threshold distance of the vehicle to enter the vehicle.)
1. A method of actively finding a dispatched passenger prior to arrival of a vehicle at a pickup location, the method comprising:
receiving, by one or more processors, information identifying the pickup location and client device information for authenticating the dispatched passenger;
receiving, by the one or more processors, sensor data from a perception system of the vehicle that identifies a target in an environment of the vehicle;
attempting, by the one or more processors, to authenticate a client device using the client device information when the vehicle is within a predetermined distance of the pickup location;
when the client device has been authenticated, the one or more processors use the sensor data to determine whether a pedestrian is within a first threshold distance of the vehicle;
when it is determined that a pedestrian is within the first threshold distance of the vehicle, stopping, by the one or more processors, the vehicle before reaching the pickup location to wait for the pedestrian within the first threshold distance of the vehicle to enter a vehicle; and
after the pedestrian enters the vehicle, maneuvering, by the one or more processors, the vehicle to a destination with the pedestrian as an occupant of the vehicle.
2. The method of claim 1, further comprising:
identifying a first ring around the vehicle corresponding to the first threshold distance;
identifying a second loop around the vehicle corresponding to a second threshold distance by shrinking the first loop to a size of the second loop after a predetermined period of time;
determining whether a pedestrian is within the second threshold distance of the vehicle using the sensor data; and
continuing to wait for a pedestrian within the first threshold distance of the vehicle to enter the vehicle when it is determined that the pedestrian is within the second threshold distance of the vehicle and at a stop.
3. The method of claim 2, further comprising: moving the vehicle toward the pickup position without a passenger when it is determined that a pedestrian is not within the second threshold distance of the vehicle.
4. The method of claim 1, further comprising:
reducing the magnitude of the first threshold distance to one or more smaller threshold distances;
determining that there are no pedestrians within the one or more smaller threshold distances;
after reducing the magnitude of the first threshold distance to one or more smaller threshold distances and determining that no pedestrian is within the one or more smaller threshold distances, determining whether the pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data; and
after determining whether the pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data, when it is determined that the pedestrian or a different pedestrian is within the first threshold distance, stopping the vehicle again to wait for the pedestrian or a different pedestrian within the first threshold distance of the vehicle to enter the vehicle.
5. The method of claim 4, wherein determining that the pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data is performed only after the vehicle has traveled a minimum distance since initially using the sensor data to determine whether the pedestrian is within the first threshold distance of the vehicle.
6. The method of claim 1, wherein determining whether a pedestrian is within a first threshold distance of the vehicle using the sensor data is performed only when the vehicle is stopped or traveling below a predetermined maximum speed limit.
7. The method of claim 1, wherein determining whether a pedestrian is within a first threshold distance of the vehicle using the sensor data is performed only when the vehicle is in a particular lane of a road.
8. The method of claim 1, wherein determining whether a pedestrian is within a first threshold distance of the vehicle using the sensor data is performed only when the vehicle is traveling on a road that meets a particular maximum speed limit.
9. The method of claim 1, further comprising: determining that the currently stopped vehicle is unsafe prior to stopping the vehicle and continuing to advance toward the pickup location.
10. The method of claim 9, further comprising: providing notification at the vehicle that an attempt to get on ahead is not possible.
11. The method of claim 1, wherein stopping the vehicle comprises stopping the vehicle on a current lane of the vehicle.
12. A system for actively finding a dispatched passenger prior to arrival of a vehicle at a pickup location, the system comprising one or more processors configured to:
receiving information identifying a pickup location and client device information for authenticating the assigned passenger;
receiving sensor data from a perception system of the vehicle identifying a target in an environment of the vehicle;
attempting to authenticate a client device using the client device information when the vehicle is within a predetermined distance of the pickup location;
when the client device has been authenticated, using the sensor data to determine whether a pedestrian is within a first threshold distance of the vehicle;
upon determining that a pedestrian is within the first threshold distance of the vehicle, stopping the vehicle to wait for the pedestrian within the first threshold distance of the vehicle to enter the vehicle before reaching the pickup location; and
after the pedestrian enters the vehicle, maneuvering the vehicle to a destination with the pedestrian as an occupant of the vehicle.
13. The system of claim 12, wherein the one or more processors are further configured to:
identifying a first ring around the vehicle corresponding to the first threshold distance;
identifying a second loop around the vehicle corresponding to a second threshold distance by shrinking the first loop to a size of the second loop after a predetermined period of time;
determining whether a pedestrian is within the second threshold distance of the vehicle using the sensor data; and
continuing to wait for a pedestrian within the first threshold distance of the vehicle to enter the vehicle when it is determined that the pedestrian is within the second threshold distance of the vehicle and at a stop.
14. The system of claim 13, wherein the one or more processors are further configured to: moving the vehicle toward the pickup position without a passenger when it is determined that a pedestrian is not within the second threshold distance of the vehicle.
15. The system of claim 12, wherein the one or more processors are further configured to:
reducing the magnitude of the first threshold distance to one or more smaller threshold distances;
determining that there are no pedestrians within the one or more smaller threshold distances;
after reducing the magnitude of the first threshold distance to one or more smaller threshold distances and determining that no pedestrian is within the one or more smaller threshold distances, determining whether the pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data; and
after determining whether the pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data, when it is determined that the pedestrian or a different pedestrian is within the first threshold distance, stopping the vehicle again to wait for the pedestrian or a different pedestrian within the first threshold distance of the vehicle to enter the vehicle.
16. The system of claim 15, wherein the one or more processors are further configured to: determining that the pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data only after the vehicle has traveled a minimum distance since initially using the sensor data to determine whether the pedestrian is within the first threshold distance of the vehicle.
17. The system of claim 12, wherein the one or more processors are further configured to: using the sensor data to determine whether a pedestrian is within the first threshold distance of the vehicle only when the vehicle is stopped or traveling within a predetermined maximum speed limit.
18. The system of claim 12, wherein the one or more processors are further configured to: determining whether a pedestrian is within the first threshold distance of the vehicle using the sensor data only when the vehicle is in a particular lane of a road.
19. The system of claim 12, wherein the one or more processors are further configured to: determining whether a pedestrian is within the first threshold distance of the vehicle using the sensor data only when the vehicle is traveling on a road that satisfies a particular maximum speed limit.
20. The system of claim 12, further comprising the vehicle.
Background
Autonomous vehicles, such as vehicles that do not require a human driver, may be used to assist in transporting passengers or items from one location to another. Such vehicles may operate in a fully autonomous mode, where the passenger may provide some initial input, such as a pickup or destination location, and the vehicle maneuvers itself to that location.
When a person (or user) wants to physically transport between two locations by vehicle, they can use any number of transport services. To date, these services typically involve a human driver who is given scheduling instructions to a certain location to pick up the user. In many cases, human drivers and users are able to arrange an accurate location for the user to get on the car. In addition, the driver and user can "wave the vehicle (flag down), use eye contact, talk to each other, or other signal to indicate that they are identified to each other, agreeing to a certain location before the vehicle reaches the exact location for pick-up. This is not easily achieved in the case of autonomous vehicles without a human driver.
Disclosure of Invention
One aspect of the present disclosure provides a method of actively finding an assigned (assigned) passenger before the vehicle arrives at the pickup location. The method comprises the following steps: receiving, by one or more processors, information identifying a pickup location and client device information for authenticating the assigned passenger; receiving, by one or more processors from a perception system of a vehicle, sensor data identifying a target in a vehicle environment; when the vehicle is within a predetermined distance of the pickup location, the one or more processors attempt to authenticate the client device using the client device information; when the client device has been authenticated by the one or more processors, determining whether the pedestrian is within a first threshold distance of the vehicle using the sensor data; when it is determined that the pedestrian is within the first threshold distance of the vehicle, stopping, by the one or more processors, the vehicle before reaching the pickup location to wait for a pedestrian within the first threshold distance of the vehicle to enter the vehicle; and after the pedestrian enters the vehicle, maneuvering the vehicle to a destination by the one or more processors with the pedestrian as an occupant of the vehicle.
In one example, the method further comprises: identifying a first ring around the vehicle corresponding to a first threshold distance; identifying a second loop around the vehicle corresponding to a second threshold distance by shrinking the first loop to a size of the second loop after a predetermined period of time; determining whether the pedestrian is within a second threshold distance of the vehicle using the sensor data; and when it is determined that the pedestrian is within the second threshold distance of the vehicle and at the stop, continuing to wait for the pedestrian within the first threshold distance of the vehicle to enter the vehicle. In another example, the method further comprises: when it is determined that the pedestrian is not within the second threshold distance of the vehicle, the vehicle is moved to the pickup position without a passenger. In another example, the method further comprises reducing the size of the first threshold distance to one or more smaller threshold distances; determining that there are no pedestrians within one or more smaller threshold distances; after reducing the magnitude of the first threshold distance to one or more smaller threshold distances and determining that no pedestrian is within the one or more smaller threshold distances, determining whether a pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data; and after determining whether the pedestrian or the different pedestrian is within the first threshold distance of the vehicle using the sensor data, when it is determined that the pedestrian or the different pedestrian is within the first threshold distance, stopping the vehicle again to wait for the pedestrian or the different pedestrian within the first threshold distance of the vehicle to enter the vehicle. In this example, since the sensor data is initially used to determine whether the pedestrian is within the first threshold distance of the vehicle, the use of the sensor data to determine that the pedestrian or a different pedestrian is within the first threshold distance of the vehicle is performed only after the vehicle has traveled the minimum distance. In another example, determining whether the pedestrian is within the first threshold distance of the vehicle using the sensor data is performed only when the vehicle is stopped or traveling below a predetermined maximum speed limit. In another example, using the sensor data to determine whether the pedestrian is within the first threshold distance of the vehicle is performed only when the vehicle is in a particular lane of the road. In another example, using the sensor data to determine whether the pedestrian is within the first threshold distance of the vehicle is performed only when the vehicle is traveling on a road that meets a particular maximum speed limit. In another example, the method further comprises: before stopping the vehicle, it is determined that the currently stopped vehicle is unsafe and proceeds toward the pick-up location. In this example, the method further includes providing a notification at the vehicle that an early pick-up is not possible. In another example, stopping the vehicle includes stopping the vehicle in a current lane of the vehicle.
Another aspect of the present disclosure provides a system for actively finding a dispatched passenger prior to arrival of a vehicle at a pickup location, the system comprising one or more processors. The one or more processors are configured to: receiving information identifying a pickup location and client device information for authenticating the assigned passenger; receiving sensor data from a perception system of a vehicle to identify a target in a vehicle environment; attempting to authenticate the client device using the client device information when the vehicle is within a predetermined distance of the pickup location; when the client device has been authenticated, determining whether the pedestrian is within a first threshold distance of the vehicle using the sensor data; stopping the vehicle to wait for a pedestrian within the first threshold distance of the vehicle to enter the vehicle before reaching the pickup location when it is determined that the pedestrian is within the first threshold distance of the vehicle; and maneuvering the vehicle to a destination with the pedestrian as an occupant of the vehicle after the pedestrian enters the vehicle.
In one example, the one or more processors are further configured to: identifying a first ring around the vehicle corresponding to a first threshold distance; identifying a second loop around the vehicle corresponding to a second threshold distance by shrinking the first loop to a size of the second loop after a predetermined period of time; determining whether the pedestrian is within a second threshold distance of the vehicle using the sensor data; and when it is determined that the pedestrian is within the second threshold distance of the vehicle and at the stop, continuing to wait for the pedestrian within the first threshold distance of the vehicle to enter the vehicle. In this example, the one or more processors are further configured to: when it is determined that the pedestrian is not within the second threshold distance of the vehicle, the vehicle is moved to the pickup position without a passenger. In another example, the one or more processors are further configured to: reducing the size of the first threshold distance to one or more smaller threshold distances; determining that there are no pedestrians within one or more smaller threshold distances; after reducing the magnitude of the first threshold distance to one or more smaller threshold distances and determining that no pedestrian is within the one or more smaller threshold distances, determining whether a pedestrian or a different pedestrian is within the first threshold distance of the vehicle using the sensor data; and after determining whether the pedestrian or the different pedestrian is within the first threshold distance of the vehicle using the sensor data, when it is determined that the pedestrian or the different pedestrian is within the first threshold distance, stopping the vehicle again to wait for the pedestrian or the different pedestrian within the first threshold distance of the vehicle to enter the vehicle. In this example, since the sensor data is initially used to determine whether the pedestrian is within the first threshold distance of the vehicle, the one or more processors are configured to use the sensor data to determine that the pedestrian or a different pedestrian is within the first threshold distance of the vehicle only after the vehicle has traveled the minimum distance. In another example, the one or more processors are configured to use the sensor data to determine whether the pedestrian is within the first threshold distance of the vehicle only when the vehicle is stopped or traveling within a predetermined maximum speed limit. In another example, the one or more processors are configured to perform using the sensor data to determine whether the pedestrian is within the first threshold distance of the vehicle only when the vehicle is in a particular lane of the road. In another example, the one or more processors are configured to use the sensor data to determine whether the pedestrian is within the first threshold distance of the vehicle only when the vehicle is traveling on a road that meets a particular maximum speed limit. In another example, the system further comprises a vehicle.
Drawings
FIG. 1 is a functional diagram of an example vehicle, according to aspects of the present disclosure.
Fig. 2 is an example representation of detailed map information in accordance with aspects of the present disclosure.
Fig. 3A-3D are example exterior views of a vehicle according to aspects of the present disclosure.
Fig. 4 is an example schematic diagram of a system according to aspects of the present disclosure.
Fig. 5 is an example functional diagram of a system according to aspects of the present disclosure.
Fig. 6 is a view of a portion of a roadway according to aspects of the present disclosure.
Fig. 7 is an example of sensor data and other information for a road segment in accordance with aspects of the present disclosure.
Fig. 8 is another example of sensor data and other information for road segments in accordance with aspects of the present disclosure.
Fig. 9 is another example of sensor data and other information for road segments in accordance with aspects of the present disclosure.
Fig. 10 is an example diagram of data in accordance with aspects of the present disclosure.
Fig. 11 is a flow chart in accordance with aspects of the present disclosure.
Detailed Description
Overview
Aspects of the present technology relate to carrying passengers in a vehicle without a human driver (e.g., an autonomous vehicle). This can be challenging due to variations in environmental conditions, lack of a human driver, and uncertainty as to how long the vehicle may need to wait (or be able to wait) for a passenger. Additionally, recognizing that a particular vehicle may be approaching a predetermined pickup location to pick up his or her person, it may be desirable to enter the vehicle as quickly as possible, rather than waiting for both the vehicle and the person to arrive at the pickup location. In another example, a person waits at the curb and observes that the vehicle is slowing down or stopping for other reasons than initiating pick-up (e.g., avoiding or yielding to another target). In these cases, the person may think that the vehicle has stopped for him or her and moved towards the vehicle, while in fact the vehicle is intended to continue driving towards the pick-up location and to allow the person to enter the vehicle at that location. This, in turn, may cause the vehicle to slow down or yield and further make it less likely that the car will advance to the pick-up point. This can frustrate users who do not know where or whether to enter the car now, and can also disturb the traffic around vehicles that may not be able to start traveling.
To facilitate faster connection between the vehicle and a person (or passenger) waiting (or assigned) to the vehicle for travel to a destination, the vehicle's computing device may operate the vehicle to actively seek that person to get onto the vehicle in the morning. The active seek logic may begin once the vehicle is within a predetermined distance of time or space from the pickup location (such as a period of time or distance before or after the computing device of the vehicle should begin seeking parking and/or a place to park), once the passenger and/or the passenger's client device have been authenticated by the vehicle, or a combination of both. Additionally or alternatively, the initiation of the logic may be linked to a set of predetermined requirements. For example, the computing device may only launch logic once all or part of the requirements are met.
At the same time, the sensing system of the vehicle may identify the object as a person and/or pedestrian from sensor data collected from the vehicle environment. The perception system may provide this information to a computing device of the vehicle. Once the predetermined distance has been reached, the computing device may in turn begin looking for a pedestrian within a short distance of the vehicle, which may be reasonably close to the vehicle, and thus the passenger assigned to the vehicle. In other words, the computing device may use information from the perception system to look for a pedestrian within a first predetermined distance of the vehicle corresponding to the walking distance in time.
If such a pedestrian is not found or identified, the vehicle may continue to reach the pickup location. Alternatively, if or when a pedestrian is identified within a predetermined distance of the vehicle, the vehicle may come to a complete stop (if not already stopped), unlock one or more doors of the vehicle, and allow the pedestrian to enter or board the vehicle at that location. Once the pick-up is complete, rather than continuing to the pickup location, the computing device may simply begin routing the vehicle to the destination. Of course, parking at any time or at any location must be a compromise in safety.
After a predetermined period of time, the logic is returned and the computing device will begin looking for a pedestrian or determining whether the same pedestrian is within a second, smaller predetermined distance of the vehicle. In other words, the loop begins to contract. Again, if such a pedestrian is not found or identified within the second predetermined distance, the vehicle may no longer wait (i.e., begin moving again) and continue to reach the pickup location. If there is a pedestrian within the second predetermined distance, a third predetermined distance may be used, and so on, until the last predetermined distance is reached, depending on the size of the initial distance and how long the pedestrian at average speed is expected to reach the vehicle.
In this regard, the computing device is able to find and identify a pedestrian that is actively heading in the boarding direction. The use of the ring makes it possible to check the proximity to the vehicle by means of ever decreasing distances, thus ensuring the travel schedule. Of course, other mechanisms besides shrink rings may be used to estimate whether a passenger is attempting to get on.
In the event that the pedestrian is within a loop but not moving toward the vehicle fast enough to stay within the loop or within a smaller loop when it is retracted (if a series of thresholds are used), the predetermined distance may be reset to the first predetermined distance as the vehicle continues to drive toward the pickup position. This will in effect provide another opportunity for a pedestrian who may or may not be assigned to the vehicle to arrive at the vehicle. Of course, to avoid the vehicle from constantly parking, the predetermined distance may be reset only after the vehicle has reached at least a minimum distance from the vehicle to first park.
The features described herein allow a vehicle without a driver to enable passengers to board the vehicle in advance (before the vehicle reaches the pickup location) in an efficient and reasonably safe manner. By actively seeking potential passengers, the computing device can allow passengers to enter the vehicle as quickly as possible, thereby improving the efficiency of the transportation system. This, in turn, reduces the likelihood of confusion for a passenger attempting to enter the vehicle and makes the passenger feel as if he or she is interacting with the vehicle as if he or she is interacting with the driver.
Example System
As shown in fig. 1, a
The
The one or
In one example, the
The
As an example, the
The
The
Fig. 2 is an example of map information 200 for a portion of a road 210. The map information 200 includes information identifying the shape, location, and other characteristics of various road features. In this example, the road 210 includes three lanes 212, 214, 216 bounded by a curb 220, lane lines 222, 224, 226, and a curb 228. Lanes 212 and 214 have the same traffic flow direction (eastward), while lane 216 has a different traffic flow (westward). Additionally, lane 212 is significantly wider than lane 214, for example, to allow the vehicle to park near curb 220. Although examples of map information include only some road features, such as curbs, lane lines, and lanes, given the nature of the road 210, the map information 200 may also identify various other road features, such as traffic lights, crosswalks, sidewalks, stop signs, yield signs, speed limit signs, road signs, and so forth. Although not shown, the detailed map information may also include information identifying speed limits and other legal traffic demands, as well as historical information identifying typical and historical traffic conditions at various dates and times.
Although the detailed map information is described herein as an image-based map, the map information need not be entirely image-based (e.g., a grid). For example, detailed map information may include one or more road maps or a graphical network of information, such as roads, lanes, intersections, and connections between these features. Each feature may be stored as graphical data and may be associated with information such as geographical location and whether to link to other related features, e.g. stop signs may link to roads and intersections etc. In some examples, the associated data may include a grid-based index of road maps to allow efficient lookup of certain road map features.
Fig. 3A-3D are examples of exterior views of the
One or
As shown in fig. 4, each
In one example, one or
Further, the
As shown in fig. 5, each
Although the
In some examples, the
The
The
The
As with the
Example method
In addition to the operations described above and illustrated in the figures, various operations will now be described. It should be understood that the following operations need not be performed in the exact order described below. Rather, various steps may be processed in a different order or concurrently, and steps may also be added or omitted.
In one aspect, a user may download an application for requesting a vehicle to a client computing device. For example,
The user may then use his or her client computing device to access the application and request the vehicle. As an example, a user, such as
These pickup and destination locations may be predefined (e.g., a particular area of a parking lot, etc.), or may simply be any location within the vehicle service area. As an example, the pickup location may default to the current location of the user's client computing device, or may be input by the user at the user's client device. For example, the user may enter an address or other location information, or select a location on a map to select a pickup location. Once the user selects one or more of the pickup and/or destination locations, the
Fig. 6 is an exemplary diagram of the
As the vehicle moves along the
As described above, to facilitate a faster connection between the vehicle and a person (or passenger) waiting (or assigned) to the vehicle for travel to a destination, the vehicle's computing device may operate the vehicle to actively seek that person to facilitate getting on ahead. The active seek logic may begin once the vehicle is within a predetermined distance in time or space from the pickup location, once the passenger or the passenger's client device has been authenticated by the
The authentication of the pedestrian may include, for example, determining whether the pedestrian in the vehicle environment is or may be an assigned passenger using one or more of facial recognition, gait detection, gesture detection, trajectory information, and the like. Face recognition or gait detection may be achieved by establishing a data timeout, for example by capturing images or video of the dispatched passenger through the perception systems of various vehicles on different trips, or by setup by the dispatched passenger providing the images or video as part of an application, as described above. The
Additionally or alternatively, the initiation of the logic may be linked to a set of predetermined requirements. For example, a computing device may initiate logic only after all or a subset of the following requirements are met:
the vehicle is within a predetermined distance from the pickup location and/or a predetermined distance assigned to a passenger of the vehicle, e.g., a location at which the computing device of the vehicle receives GPS information from the client device of the person.
The computing device has successfully verified the dispatched passenger and/or the client device of the dispatched passenger, as shown in any of the examples discussed above.
The vehicle has come to a complete stop or is traveling at an extremely low speed (such as 3mph or higher or lower) or near pedestrian walking speed.
The vehicle is not currently located in an area where parking or containment is prohibited (such as an intersection, a railroad crossing, and/or according to traffic regulations or signs in the area).
Vehicles travel on lanes suitable for side-to-side parking (pull over) (e.g., the rightmost lane of the right-hand driving country, the leftmost lane of the left-hand driving country, or lanes near the edge of a lane where parking may be justified on either side (such as one-way streets)) or off-lane on public roads (e.g., on parking lots or private roads). Of course, it is also contemplated to do so if it is appropriate to drop the passenger in certain areas that do not belong to the rightmost (or leftmost) lane.
The vehicle is currently on a road that is considered to be receptive to passengers (taking into account the safety of passengers, vehicles and other road users), such as a road with certain speed limits, e.g., less than 35 miles per hour, or more or less. If there is a parking lane, a higher speed road may be accepted. In this regard, an expressway without a parking lane may be unacceptable.
The vehicle is not in a complex operation, such as a multi-point turn.
The vehicle is not currently following a route instruction from a remote operator or dispatch system that requires the vehicle to stop before reaching the destination.
The computing device has not received instructions to pick up the user at a location other than the pickup location. For example, if the computing device has attempted a "fly-by" pickup, or a passenger pickup that is implemented before the vehicle reaches a predetermined pickup area for the vehicle to pickup passengers. The passenger may initiate the skip-through pickup by, for example, physically signaling the vehicle or using a client computing device to request the skip-through pickup. Likewise, if the passenger reports a location indicating that the passenger has been waiting at a particular location (e.g., roadside) and is therefore likely ready to depart, the vehicle may initiate a skip pickup. In this example, the vehicle will attempt to stop at that location and no logic will be required. Alternatively, the vehicle may ask the user if he or she is interested in real-time skip-loading, for example, by displaying a pop-up notification and the user may select an option to accept or decline the request to perform skip-loading. In some examples, the abbreviated take-over notification may be made using a visual representation of a relationship between the vehicle and a location of the passenger's client computing device displayed on the vehicle or the passenger's client computing device. Thus, in view of better or alternative loading locations or areas for the vehicle, the passenger, or both, the skip loading may be the result of the last minute changing to the passenger's loading location or area.
The computing device is not yet too close in time or distance to reach the destination, for example, where the computing device is already attempting to maneuver to stop and wait for a passenger. Of course, it may be faster or more convenient to allow the vehicle to stop at an angle and allow passengers to enter the vehicle if the stopping maneuver would take some additional time to park parallel to the curb.
Again, prior to initiating the logic, the computing device may determine one or more of the aforementioned requirements. This may also include determining whether a particular subset and/or combination of requirements has been met.
As described above, at the same time, the sensing system of the vehicle may identify the target as a person and/or pedestrian from sensor data collected from the vehicle environment. The perception system may provide this information to a computing device of the vehicle. Once the predetermined distance has been reached and any other necessary requirements have been met, the
If such a pedestrian is not found or identified, the vehicle may proceed to the pickup location. Alternatively, if or when one or more pedestrians are identified within a predetermined distance of the vehicle, the vehicle may come to a complete stop (if not already parked), unlock one or more doors of the vehicle, and allow the pedestrians to enter or board the vehicle at that location. Returning to fig. 8, the computing device may identify two pedestrians corresponding to the bounding
Of course, stopping the vehicle at any time or at any location must be balanced in terms of safety. For example, the computing device may override a stop (stop), such as when the vehicle is on or near a rail, an emergency vehicle is identified in the area (e.g., due to an alarm and/or flashing emergency light being detected by the sensing system), there are many other vehicles in motion within the area, the vehicle is not currently parked on the correct lane (left lane instead of right lane, etc.), the vehicle is attempting a particular maneuver (multi-point or at an intersection), whether the vehicle is in a no-parking zone, or there are other obstacles that result in an undesirable stop. In addition, the computing device may determine whether the pickup location is merely a good side of the parking or a close proximity to the parking space (i.e., parking side by side is better in the parking area than on the lane). In this case, the computing device may of course continue to the pickup location, but at the same time send (if there is a link between the computing device and the client device), display or sound a message to the pedestrian indicating that the vehicle will not stop. Although it is also possible, it is not useful to display such information on the client device, as pedestrians will pay more attention to the vehicle than the client device.
Returning to logic, after a predetermined period of time while stopping and waiting for a pedestrian to enter the vehicle, the computing device will begin looking for a pedestrian (or determining whether the same or a different pedestrian) within a second, smaller predetermined distance of the vehicle. This constriction may be a "continuous" function that reduces or diminishes the size of the ring over time, or simply replaces the ring with a smaller ring at a discrete distance after a predetermined period of time has been reached. In other words, the
Again, if a pedestrian is not found or identified within the loop when the loop is retracted or within a second predetermined distance if the loop is replaced with a smaller loop, the vehicle may no longer wait (i.e., begin moving again) and proceed to the pickup location. If there are pedestrians remaining in the loop or within the second predetermined distance as the loop continues to contract, the loop may contract further or a third predetermined distance may be used, and so on until the loop reaches the vehicle or meets the last predetermined distance depending on the size of the initial distance and the walking speed used. For example, fig. 10 depicts a series of
In this regard, the computing device is able to find and identify a pedestrian that is actively heading in the direction of boarding. The use of the ring makes it possible to check the proximity to the vehicle by means of ever decreasing distances, thus ensuring the travel schedule. Of course, other mechanisms besides shrink rings may be used to estimate whether a passenger is attempting to get on.
In some examples, the computing device may "invalidate" or otherwise ignore a particular pedestrian that continuously appears within a predetermined distance but does not actually attempt to enter the vehicle. This may occur when a pedestrian happens to walk down a sidewalk next to the vehicle, actually off the vehicle, or does not actually advance toward the vehicle to indicate that the pedestrian is actually about to get on the vehicle. The invalidation may also occur based on other signals that may indicate the intent of the pedestrian, such as by gaze detection (pedestrian looking toward vehicle), facial recognition or gait detection (if the computing device has comparison data assigned to the occupant of the vehicle), GPS location from the client device (i.e., if the user appears to be hundreds of meters or feet away), whether the pedestrian is not assigned to the occupant of the vehicle, whether information from the client device indicates that the occupant assigned to the vehicle is moving compared to the pedestrian being observed (changes in GPS coordinates or accelerometer or gyroscope information), gesture detection, trajectory information, whether the pedestrian is gesturing toward the vehicle with an arm or hand, lifting a device having a particular color (e.g., displaying a particular predetermined color on a device for recognition by the computing device of the vehicle), and so forth. In other words, if more than one pedestrian appears to be actively heading in the boarding direction as discussed above, one or more of these methods may be used to narrow the range of the pedestrian and identify the assigned passenger.
In the event that the pedestrian is within the loop but not moving toward the vehicle fast enough to stay within the loop or within a smaller loop when the loop is retracted (if a series of thresholds are used), the predetermined distance may be reset to the first predetermined distance as the vehicle continues to travel to the pickup location. This will in effect provide another opportunity for a pedestrian who may or may not be assigned to the vehicle to arrive at the vehicle. Of course, to avoid the vehicle from constantly parking, the predetermined distance may be reset only after the vehicle has reached at least a minimum distance from its first parking. For example, the minimum distance may be 5 meters or more or less that the vehicle travels forward after the vehicle stops and/or a pedestrian is identified within the first predetermined distance.
In some examples, prior to the vehicle authenticating the client device, the pedestrian may attempt to enter the slow-driving vehicle, for example, by pulling a door handle of the vehicle. In response, the computing device may simply stop and allow the pedestrian to enter the vehicle. Once in the vehicle, the computing device may continue to attempt to authenticate the client device before proceeding to the destination, or may alternatively begin moving toward the destination while continuing to authenticate in order to avoid remaining stopped on the road for too long. Similarly, if the authentication is unsuccessful, the operator may be assisted on site via speakers and/or video connections within the vehicle, prior to or while the vehicle is traveling, to communicate with the passenger and confirm his or her identity and desired destination. This functionality is particularly useful when the client device's battery is low or exhausted and a wireless link cannot be established with the vehicle's computing device.
If, once the vehicle has authenticated the client device and unlocked the door, the pedestrian opens the door but does not actually enter the vehicle, but merely closes the door without entering, the computing device may wait for the pedestrian to move a certain distance (e.g., a few feet or more or less from the vehicle) and then simply proceed to the pickup location. To ensure that no person is in the vehicle, in some cases, the computing device may use feedback from internal sensors to determine whether a passenger is in the vehicle. Additionally or alternatively, a live operator may view live video and/or audio feeds to confirm whether a passenger is in the vehicle.
In addition to stopping the vehicle to wait for passengers and unlocking one or more doors of the vehicle, the above-described features may also (additionally or alternatively) be used to determine whether to open one or more doors of the vehicle to allow pedestrian entry. Of course, if the vehicle is stopped, the doors will automatically unlock or open, which may pose a risk of unauthorized personnel entering the vehicle. In this regard, this behavior may be limited to only certain situations, such as a client computing device that has verified the dispatched passenger, location information generated by the client computing device (such as GPS location), or a perception system of the vehicle (such as sensor data generated by one or more lidar, camera, radar, or other sensors) indicating that the pedestrian is within a short distance of the vehicle (such as 10 meters or more or less), that the pedestrian has met some additional confirmation requirements (e.g., using face recognition, gait detection, gaze detection, and/or trajectory confirmation, as described above), and/or that unauthorized persons attempting to enter the vehicle can be more easily identified only during certain periods of time (such as daytime) or certain periods of time (such as 7 am to 6 pm). Furthermore, the vehicle may also be required to meet certain requirements, such as a full stop and the vehicle's gearbox having been shifted to a park position, before one or more doors are opened.
In addition, all doors of the vehicle, or only certain doors, that passengers may use to enter the vehicle may be opened. For example, the computing device may open a door of the vehicle proximate to a curb or a location of one or more pedestrians identified as being within a predetermined distance of the vehicle.
Opening one or more doors may even encourage passengers to sit in a particular row near an opened door. For example, where there are multiple rows of seats, there is an open door adjacent a particular row, for example the middle row when there are three rows of seats, which may actually encourage passengers to use that row, or may not be fully utilized.
Additionally, the doors may remain open until the passenger closes one or more of the doors, the passenger begins traveling (e.g., by pressing a start button in the vehicle), or after a predetermined period of time has elapsed. In the latter two cases, the
In the above example, the computing device is able to determine that the dispatched passenger is in the vicinity of the vehicle due to authentication of the client device of the dispatched passenger and use it as a signal to initiate the logic described above. Alternatively, if authentication has not been performed, GPS or other location information provided to the computing device, for example by sending from the client device to a server that relays the information to the client device, may be used to determine that the assigned passenger is in the vicinity of the vehicle, and thus, the location information may be sufficient for the computing device to initiate logic.
As computing devices approach the pickup location, they and/or the server computing device (to which the passenger is assigned and to which the vehicle is dispatched) may send information to the client device to display various notifications to the passenger assigned to the vehicle. This may include, for example, a notification of when the vehicle will arrive, and the like. However, once the computing device has initiated the logic described above, this information may or may not be sent to the client device, and some of the above notifications may not be displayed, in order to reduce the likelihood of the passenger being distracted by the notification when attempting to get on in advance, and to make the process of getting on in advance more difficult. For example, it may still be useful to provide notification that the doors of the vehicle have been unlocked once the vehicle is actually stopped, or in some cases shortly before.
Fig. 11 is a flow diagram 1100 that may be executed by one or more processors, such as one or
Unless otherwise specified, the foregoing alternative examples are not mutually exclusive and may be implemented in various combinations to achieve unique advantages. As these and other variations and combinations of the features discussed above can be utilized without departing from the subject matter defined by the claims, the foregoing description of the embodiments should be taken by way of illustration rather than by way of the subject matter defined by the claims. Furthermore, the provision of examples described herein, as well as the use of phrases such as "and" including "and the like, should not be construed to limit claimed subject matter to the particular examples; rather, these examples are intended to illustrate only one of many possible embodiments. Further, the same reference numbers in different drawings may identify the same or similar elements.
INDUSTRIAL APPLICABILITY
The techniques described herein enjoy wide industrial applicability, including, for example, in the autonomous field.
- 上一篇:一种医用注射器针头装配设备
- 下一篇:用于借助于增强现实辅助团队工作的系统