阅读说明：本技术 路线生成系统、路线生成方法和路线生成程序 (Route generation system, route generation method, and route generation program ) 是由 村山修 景祎 于 2019-03-18 设计创作，主要内容包括：行驶状况提取部(120)从蓄积有表示交通事故场景的状况的交通事故场景信息的交通事故数据库中,提取与所述交通事故场景对应的车辆的行驶状况。回避路线生成部(130)根据行驶状况,生成回避交通事故场景的多个回避路线。有效性决定部(140)针对多个回避路线分别决定表示有效性的值作为有效评价值。有效路线选择部(150)根据多个回避路线各自的有效评价值,从多个回避路线中选择最有效的回避路线作为有效路线。有效信息构建部(160)将把交通事故场景与有效路线对应起来的有效路线信息(172)存储于存储部(170)。(A travel situation extraction unit (120) extracts the travel situation of a vehicle corresponding to a traffic accident scene from a traffic accident database in which traffic accident scene information indicating the situation of the traffic accident scene is stored. An avoidance line generation unit (130) generates a plurality of avoidance lines for avoiding a traffic accident scene, based on the traveling condition. A validity determination unit (140) determines values indicating the validity as validity evaluation values for each of the avoidance lines. An effective route selection unit (150) selects the most effective avoidance line from the avoidance lines as an effective line on the basis of the effective evaluation values of the avoidance lines. An effective information construction unit (160) stores effective route information (172) in a storage unit (170), the effective route information associating a traffic accident scene with an effective route.)

1. A route generation system, the route generation system having:

a travel situation extraction unit that extracts a travel situation of a vehicle corresponding to a traffic accident scene from a traffic accident database in which traffic accident scene information indicating a situation of the traffic accident scene is accumulated;

an avoidance line generation unit that generates a plurality of avoidance lines for avoiding the traffic accident scene, based on the travel situation;

a validity determination unit that determines values indicating validity as validity evaluation values for each of the avoidance lines;

an effective route selection unit that selects, as an effective route, an avoidance route that is most effective from the plurality of avoidance routes, based on the effective evaluation values of the respective avoidance routes; and

and an effective information construction unit that stores effective route information in which the traffic accident scene and the effective route are associated with each other in a storage unit.

2. The route generation system according to claim 1,

the effectiveness determination unit determines an effective evaluation value for each of the avoidance lines using an effectiveness evaluation model indicating a correspondence between the effective evaluation value and a feature quantity indicating a feature of the avoidance line.

3. The route generation system according to claim 2,

the route generation system has model generation means for generating the effectiveness evaluation model,

the model generation device comprises:

a learning data storage unit that stores a learning data set in which each of the avoidance lines is associated with a preset subjective evaluation value that is a value indicating effectiveness;

a feature value extraction unit that extracts a feature value vector indicating a feature value for evaluating the effectiveness of each of the avoidance lines from the learning dataset; and

and an evaluation model generation unit that generates the effectiveness evaluation model using the feature vector of each of the avoidance lines.

4. The route generation system according to claim 2 or 3,

the effectiveness determination unit calculates a final score indicating a probability of obtaining an effective evaluation value for each of the plurality of avoidance lines using the effectiveness evaluation model and the feature vector of each of the plurality of avoidance lines, and determines an effective evaluation value having a highest final score among the plurality of effective evaluation values as an effective evaluation value for the avoidance line.

5. The route generation system according to claim 4,

the validity determination unit calculates a feature score of the feature amount for each of the plurality of valid evaluation values, and calculates a sum of the feature scores as the final score.

6. The route generation system according to any one of claims 1 to 5,

the avoidance line generating unit generates the plurality of avoidance lines using the traveling condition including the speed and direction of the vehicle corresponding to the traffic accident scene.

7. The route generation system according to any one of claims 1 to 6,

the effective information constructing unit stores the effective route information in the storage unit in the form of a knowledge base.

8. A route generation method, wherein,

in a route generation system having a travel situation extraction unit, an avoidance route generation unit, a validity determination unit, a valid route selection unit, and a valid information construction unit,

the travel situation extraction unit extracts a travel situation of a vehicle corresponding to a traffic accident scene from a traffic accident database in which traffic accident scene information indicating a situation of the traffic accident scene is accumulated,

the avoidance line generating unit generates a plurality of avoidance lines for avoiding the traffic accident scene based on the traveling condition,

the effectiveness determination unit determines a value indicating effectiveness as an effectiveness evaluation value for each of the avoidance lines,

the effective route selecting unit selects the most effective avoidance line as an effective line from the avoidance lines based on the effective evaluation values of the avoidance lines,

the effective information constructing unit stores effective route information in which the traffic accident scene corresponds to the effective route in a storage unit.

9. A route generation program that causes a computer to execute:

a travel situation extraction process of extracting a travel situation of a vehicle corresponding to a traffic accident scene from a traffic accident database in which traffic accident scene information indicating a situation of the traffic accident scene is accumulated;

avoidance line generation processing for generating a plurality of avoidance lines for avoiding the traffic accident scene, based on the driving situation;

a validity determination process of determining a value indicating validity as a validity evaluation value for each of the avoidance lines;

effective route selection processing for selecting, as an effective route, an avoidance route that is most effective from the plurality of avoidance routes, based on the effective evaluation values of the respective avoidance routes; and

and an effective information construction process of storing effective route information in which the traffic accident scene and the effective route are associated with each other in a storage unit.

Technical Field

The invention relates to a route generation system, a route generation method, and a route generation program. And more particularly to a route generation system, a route generation method, and a route generation program for generating avoidance routes from a traffic accident database.

Background

The following techniques are disclosed in the prior art documents: the risk is predicted by a skilled experiential person and countermeasures are prepared in advance, thereby avoiding the risk.

Patent document 1 discloses the following technique: judging the danger degree according to the ship motion information, other ship motion information and various obstacle information; then, an avoidance route plan is generated based on the skilled knowledge base so that the future travel position of the ship is the optimal future travel position in accordance with the law knowledge base. In patent document 1, as a skilled knowledge base of an expert system, future sailing positions based on arbitrary acceleration, deceleration, and steering experienced by a skilled rider are stored. Furthermore, as the rule knowledge base, compliance items such as rules and regulations are stored.

Documents of the prior art

Patent document

Patent document 1: japanese laid-open patent publication No. 9-066894

Disclosure of Invention

Problems to be solved by the invention

Patent document 1 has the following problems: there is a limit to the knowledge of an individual skilled operator, and only a limited number of scenarios can be addressed. Further, there are problems as follows: it is difficult to build a knowledge base for coping extensively with all emergencies. Further, there are problems as follows: when a knowledge base based on personal knowledge is used, there are individual differences, and therefore, there are cases where optimal measures cannot be taken in emergency.

The purpose of the present invention is to realize safe and safe automatic driving and driving assistance by generating an appropriate avoidance line according to the traffic situation that actually occurs, and thereby avoiding an accident and reducing damage at the time of collision.

Means for solving the problems

The route generation device of the present invention includes:

a travel situation extraction unit that extracts a travel situation of a vehicle corresponding to a traffic accident scene from a traffic accident database in which traffic accident scene information indicating a situation of the traffic accident scene is accumulated;

an avoidance line generation unit that generates a plurality of avoidance lines for avoiding the traffic accident scene, based on the travel situation;

a validity determination unit that determines values indicating validity as validity evaluation values for each of the avoidance lines;

an effective route selection unit that selects, as an effective route, an avoidance route that is most effective from the plurality of avoidance routes, based on the effective evaluation values of the respective avoidance routes; and

and an effective information construction unit that stores effective route information in which the traffic accident scene and the effective route are associated with each other in a storage unit.

Effects of the invention

According to the route generation device of the present invention, it is possible to seek for avoiding an accident and reducing damage at the time of collision by generating an appropriate avoidance route as an effective route based on a traffic condition actually occurring, and to realize safe and safe automatic driving and driving assistance.

Drawings

Fig. 1 is a configuration example of a route generation system according to embodiment 1.

Fig. 2 is an example of generating a plurality of avoidance lines according to embodiment 1.

Fig. 3 shows an example of a validity evaluation model according to embodiment 1.

Fig. 4 is a configuration example of a model generation device that performs the model generation process of embodiment 1.

Fig. 5 is an example of a data set for learning for generating the effectiveness evaluation model according to embodiment 1.

Fig. 6 is an example of extracting feature vectors from each avoidance line in embodiment 1.

Fig. 7 is a flowchart illustrating the model generation process according to embodiment 1.

Fig. 8 is a diagram for explaining an example of a traffic accident scene according to embodiment 1.

Fig. 9 is a flowchart for explaining the route generation processing according to embodiment 1.

Fig. 10 is an example of generating a plurality of avoidance lines according to embodiment 1.

Fig. 11 is an example of the feature vector V of the avoidance line 1 according to embodiment 1.

Fig. 12 is a diagram showing the final score S of each effective evaluation value for each avoidance line in embodiment 1.

Fig. 13 is a calculation formula of the final score S according to embodiment 1.

Fig. 14 is a configuration example of a route generation device according to a modification of embodiment 1.

Detailed Description

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding portions are denoted by the same reference numerals. In the description of the embodiments, the description of the same or corresponding portions is omitted or simplified as appropriate.

Embodiment mode 1

Description of the structure

Fig. 1 is a configuration example of a route generation system 500 according to the present embodiment.

The route generation system 500 includes a route generation device 100 and a model generation device 200. Here, the route generation device 100 and the model generation device 200 are shown as different devices, but the route generation device 100 and the model generation device 200 may be one device. The model generation device 200 may be mounted on the route generation device 100.

The route generation device 100 is a computer. The route generation device 100 has a processor 910, and has other hardware such as a memory 921, an auxiliary storage device 922, an input interface 930, an output interface 940, and a communication device 950. The processor 910 is connected to other hardware via signal lines, and controls these other hardware.

The route generation device 100 includes, as functional elements, an accident data acquisition unit 110, a travel situation extraction unit 120, an avoidance line generation unit 130, a validity determination unit 140, a valid route selection unit 150, a valid information construction unit 160, and a storage unit 170. The storage unit 170 stores a validity evaluation model 171 and valid route information 172.

The functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the valid information construction unit 160 are implemented by software. The storage unit 170 is provided in the memory 921 or the auxiliary storage device 922.

The processor 910 is a device that executes a route generation program. The route generation program is a program for realizing the functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance route generation unit 130, the validity determination unit 140, the valid route selection unit 150, and the valid information construction unit 160.

The processor 910 is an Integrated Circuit (IC) that performs arithmetic processing. Specific examples of Processor 910 include a CPU (Central Processing Unit), a DSP (Digital Signal Processor), and a GPU (Graphics Processing Unit).

The memory 921 is a storage device that temporarily stores data. An example of the Memory 921 is a Static Random Access Memory (SRAM) or a Dynamic Random Access Memory (DRAM).

The auxiliary storage 922 is a storage device that stores data. A specific example of the auxiliary storage device 922 is an HDD. The auxiliary storage 922 may be a removable storage medium such as an SD (registered trademark) memory card, CF, NAND flash memory, a flexible disk, an optical disk, a compact disk, a Floppy (registered trademark) disk, or a DVD. In addition, the HDD is an abbreviation of Hard Disk Drive. SD (registered trademark) is an abbreviation of Secure Digital. CF is an abbreviation of CompactFlash (compact flash (registered trademark)). DVD is an abbreviation of Digital Versatile Disk.

The input interface 930 is a port to which an input device such as a mouse, a keyboard, or a touch panel is connected. Specifically, the input interface 930 is a USB (Universal Serial Bus) terminal. The input interface 930 may be a port connected to a LAN (Local Area Network).

The output interface 940 is a port to which a cable of an output device such as a display is connected. Specifically, the output Interface 940 is a USB terminal or an HDMI (High Definition Multimedia Interface) terminal. Specifically, the Display is an LCD (Liquid Crystal Display).

The communication device 950 has a receiver and a transmitter. The communication device 950 is connected to a communication network such as a LAN, the internet, or a telephone line in a wireless manner. Specifically, the communication device 950 is a communication chip or NIC (Network Interface Card).

The route generation program is read into the processor 910 and executed by the processor 910. The memory 921 stores not only a route generation program but also an OS (Operating System). The processor 910 executes the route generation program while executing the OS. The route generation program and the OS may also be stored in the secondary storage 922. The route generation program and the OS stored in the secondary storage device 922 are loaded into the memory 921 and executed by the processor 910. In addition, part or all of the route generation program may be incorporated in the OS.

The route generation apparatus 100 may also have a plurality of processors instead of the processor 910. The plurality of processors share the execution of the route generation program. Like the processor 910, each processor is a device that executes a route generation program.

The data, information, signal values, and variable values utilized, processed, or output by the route generation program are stored in memory 921, registers within secondary storage 922, or processor 910, or flash memory.

The "units" of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the valid information construction unit 160 may be rewritten into "processing", "procedure", or "procedure". Further, "processing" of the accident data acquisition processing, the travel situation extraction processing, the avoidance line generation processing, the validity determination processing, the valid line selection processing, and the valid information construction processing may be rewritten to "program", "program product", or "computer-readable storage medium having a program recorded thereon".

The route generation program causes the computer to execute each process, each procedure, or each step by rewriting the "section" of each section to the "process", "procedure", or "step". Further, the route generation method is a method performed by the route generation device 100 executing a route generation program.

The route generation program may be provided by being stored in a computer-readable recording medium. Further, the route generation program may also be provided as a program product.

Summary of the function

The accident data acquisition unit 110 acquires traffic accident scene information from a conventional large traffic accident database. Traffic accident scene information indicating the situation of a traffic accident scene is accumulated in the traffic accident database.

The driving situation extraction unit 120 extracts the driving situation of the vehicle corresponding to the traffic accident scene. The running condition of the vehicle includes the speed and direction of the vehicle. Specifically, the traveling situation extraction unit 120 extracts traveling situations such as the position, speed, road surface situation, road shape, traveling direction, position, speed, and traveling direction of the oncoming vehicle from the traffic accident scene information acquired by the accident data acquisition unit 110. The vehicle here is referred to as a host vehicle.

The avoidance line generating unit 130 generates a plurality of avoidance lines 40 for avoiding a traffic accident scene according to the traveling condition. The avoidance line is a line of an accident that is considered to be avoided, such as an accident of damaging an object, a personal accident, a rear-end collision, or a secondary disaster. The avoidance line generating unit 130 generates the plurality of avoidance lines 40 based on the elements of the traveling condition extracted by the traveling condition extracting unit 120.

Specifically, the distance of the emergency braking is first determined according to the speed and direction of the vehicle. For example, if the vehicle is traveling at a speed of 40km/h, the braking distance is 7.9m when the road surface condition is dry (the friction coefficient is 0.8).

Fig. 2 is a diagram showing an example of generating a plurality of avoidance lines 40 according to the present embodiment.

As shown in fig. 2, when the left direction is negative and the right direction is positive and the direction change is applied, a plurality of paths generated by changing the direction change by 2 degrees each time in the range of-40 degrees to +40 degrees are shown. For simplicity of explanation, the dynamic characteristics such as a slip when the vehicle is turning sharply are omitted. For the sake of simplicity of explanation, the amount of change in angle at the time of route generation is assumed to be constant. However, the angle may also be changed during the actual braking.

The effectiveness determination unit 140 determines a value indicating effectiveness as an effectiveness evaluation value for each of the avoidance lines 40. The effectiveness determination unit 140 evaluates effectiveness for each of the avoidance lines 40 generated by the avoidance line generation unit 130, and determines an effectiveness evaluation value for each of the avoidance lines 40. The validity determination unit 140 is also referred to as a validity route evaluation unit.

In the present embodiment, as a method of evaluating the effective route, for example, a maximum entropy method can be used. When evaluating the avoidance rules, subjective evaluation is performed by a statistical method based on the avoidance line generated in advance. From the pair of the avoidance line and the subjective evaluation value, it is determined to what degree the evaluation values corresponding to the evaluation feature quantities of the avoidance line are similar. The data pairs are collated by manual work in advance. For simplicity of description, subjective evaluation will be described with 5-stage evaluation of 1 to 5. For example, a plurality of evaluators constituted by experts respectively score a plurality of avoidance lines by 1 to 5. If the score is high, it means that the appropriateness of the route is high.

The effectiveness determination unit 140 determines the effectiveness evaluation value of each of the plurality of avoidance lines by using an effectiveness evaluation model 171, the effectiveness evaluation model 171 indicating the correspondence between the effectiveness evaluation value P and a feature quantity T indicating the feature of the avoidance line.

Fig. 3 is a diagram illustrating an example of the effectiveness evaluation model 171 according to the present embodiment.

The storage unit 170 stores a validity evaluation model 171 used by the validity determination unit 140. The validity evaluation model 171 is generated in advance by the model generation device 200 described later and stored in the storage unit 170. As shown in fig. 3, the effectiveness evaluation model 171 describes the strength of the relationship between each feature quantity T of the avoidance line and the effectiveness evaluation value P as a feature quantity score Si. Here, the number of characteristic amounts for evaluation is 7 (i ═ 7) as described below.

(1) Whether or not to collide with the vehicle

(2) Whether or not to collide with the road shape

(3) Whether or not to collide with a person

(4) Whether or not it is a collision from the front

(5) Whether or not it is a collision from the side

(6) Whether or not to enter the opposite lane

(7) Whether or not to run in reverse

The effective route selection unit 150 selects the most effective avoidance line from the plurality of avoidance lines 40 as an effective line based on the effective evaluation value of each of the plurality of avoidance lines 40. The effective route selection unit 150 selects the avoidance route having the highest effective evaluation value determined by the effectiveness determination unit 140 as an effective route, and outputs the effective route to the effective information construction unit 160.

The effective information constructing unit 160 stores effective route information 172, which associates the traffic accident scene with the effective route, in the storage unit 170. The valid route information 172 is also referred to as a knowledge base. The effective information constructing section 160 is also referred to as a knowledge base constructing section.

Description of actions

The operation of the route generation system 500 according to the present embodiment will be described.

First, a model generation process for generating the effectiveness evaluation model 171 will be described. The route generation system 500 includes a model generation device 200 that generates a validity evaluation model 171.

Fig. 4 is a configuration example of a model generation device 200 that performs the model generation process according to the present embodiment.

The model generation device 200 includes, as functional elements, a learning data storage unit 210, a feature extraction unit 220, and an evaluation model generation unit 230. The model generation device 200 is a computer, and has the same hardware configuration as the route generation device 100. The model generation device 200 generates a validity evaluation model 171. The generated validity evaluation model 171 is stored in the storage unit 170 of the route generation device 100.

The learning data storage unit 210 stores a learning data set 211 in which each of the plurality of avoidance lines 40 is associated with a preset subjective evaluation value. The subjective evaluation value is a value that is set in advance by an expert and indicates the effectiveness of the avoidance line. The learning data storage unit 210 stores pairs of the avoidance line and a subjective evaluation value indicating the effectiveness of the avoidance line as a learning data set 211.

Fig. 5 is a diagram showing an example of the learning dataset 211 for generating the effectiveness evaluation model according to the present embodiment.

In the model generation device 200, the learning data storage unit 210 assigns a route number to each of the avoidance lines 40 in fig. 2, and stores a pair of each avoidance line and a corresponding effective evaluation value (subjective evaluation value) as a learning data set 211 in a storage device. The effective evaluation value (subjective evaluation value) is given by an expert in advance through 5 stages of subjective evaluation.

The feature extraction unit 220 extracts a feature vector V representing a feature for evaluating the effectiveness of each of the avoidance lines 40 from the learning dataset 211. Specifically, the feature extraction unit 220 extracts the feature vector V from each of the avoidance lines 40 in fig. 2. The extracted feature vector V is used to generate a validity evaluation model 171 indicating the correspondence between the feature vector and the validity evaluation value. For example, the one-hot model is used for the extraction of the feature quantity. As a specific example, when a collision with a vehicle occurs even when an avoidance line is used, the feature amount of the collision with the vehicle in the avoidance line is 1. Here, the feature vector V is generated from each avoidance line by such a method.

Fig. 6 is a diagram showing an example in which the feature vector V is extracted from each avoidance line in the present embodiment. Pairs of the feature vector V and the effective evaluation value (subjective evaluation value) are associated with each avoidance line.

Fig. 7 is a flowchart for explaining the model generation processing according to the present embodiment.

In step S1, the feature extraction unit 220 extracts a feature vector V indicating a feature for evaluating the effectiveness of the avoidance line from the learning dataset 211. Specifically, the feature extraction unit 220 extracts the feature vector V from each of the avoidance lines 40. The feature extraction unit 220 outputs the feature vector V to the evaluation model generation unit 230.

In step S2, the evaluation model generation unit 230 generates the effectiveness evaluation model 171 using the learning data set 211 and the feature vector V of each of the avoidance lines 40. Specifically, the evaluation model generation unit 230 generates the effectiveness evaluation model 171 using a pair of the feature vector V obtained from the feature extraction unit 220 and an effectiveness evaluation value (subjective evaluation value). For example, when a collision with a pedestrian occurs, it cannot be said that the collision is a good avoidance line, and therefore the evaluation value is low. Therefore, the weight of whether or not to collide with a pedestrian increases. The evaluation model generation unit 230 performs the same processing as described above on all the data included in the learning data set 211, and optimizes the weight for each feature amount item. Then, the evaluation model generation unit 230 calculates each validity evaluation value so as to optimize the weight for each feature quantity item, and finally generates a validity evaluation model 171 as shown in fig. 3.

Next, a route generation process of the route generation device 100 will be described.

Fig. 8 is a diagram illustrating an example of a traffic accident scene according to the present embodiment.

In fig. 8, the following traffic accident scenario is shown: when the own vehicle 301 passes through an intersection without a traffic signal at a speed of 40km/h from left to right, another vehicle 302 turning left suddenly appears from a blind spot, and the vehicle cannot avoid the collision.

Fig. 9 is a flowchart for explaining the route generation processing according to the present embodiment.

As in the traffic accident scene shown in fig. 8, the own vehicle 301 is traveling from left to right.

In step S11, the accident data acquisition unit 110 acquires a traffic accident scene from the traffic accident database. Then, the driving situation extraction unit 120 extracts the driving situation from the traffic accident scene.

Fig. 10 is a diagram showing an example of generating a plurality of avoidance lines according to the present embodiment.

In step S12, the avoidance line generating unit 130 generates the plurality of avoidance lines 40 when the sudden brake is temporarily depressed from the position of the own vehicle 301. Specifically, the avoidance line generating unit 130 generates the plurality of avoidance lines 40 as shown in fig. 10.

In step S13, the effectiveness determination unit 140 acquires the plurality of avoidance lines 40 and evaluates the effectiveness of each avoidance line. Specifically, the effectiveness determination unit 140 determines an effectiveness evaluation value for each of the avoidance lines 40.

The effectiveness determination unit 140 calculates final scores S indicating the level of probability of obtaining an effective evaluation value for each of the plurality of effective evaluation values for each of the plurality of avoidance lines by using the effectiveness evaluation model 171 and the feature vector V of each of the plurality of avoidance lines 40. Then, the effectiveness determination unit 140 determines, as the effective evaluation value P of the avoidance line, the effective evaluation value having the highest final score S among the plurality of effective evaluation values. In this case, the validity determination unit 140 calculates the feature score Si of the feature amount for each of the plurality of valid evaluation values, and calculates the sum of the feature scores Si as the final score S.

Fig. 11 is a diagram showing an example of the feature vector V of the avoidance line 1 according to the present embodiment.

The effectiveness determination unit 140 extracts the feature vector V for each avoidance line. For example, as shown in fig. 11, in the case of the example of the avoidance line 1, the feature value of the collision with the vehicle is 1 because the collision with the vehicle occurs. Further, the feature amount of the collision with the road shape or the pedestrian is 0. Further, since the vehicle is involved in a frontal collision, the feature amount of the frontal collision is 1. The feature amount of the avoidance line 1 is vectorized by such a modeling method.

Next, the effectiveness determination unit 140 obtains a feature quantity score Si indicating a score of the feature quantity for each effectiveness evaluation value for each avoidance line using the effectiveness evaluation model 171 stored in the storage unit 170. Then, the effectiveness determination unit 140 calculates the sum of the feature amount scores Si as the final score S for each effectiveness evaluation value for each avoidance line.

Fig. 12 is a diagram showing the final score S of each effective evaluation value for each avoidance line in the present embodiment. Fig. 13 is a calculation formula of the final score S according to the present embodiment.

As shown in fig. 12, the effectiveness determination unit 140 determines the feature quantity score Si for each effective evaluation value of the avoidance line 1 using the feature quantity vector V of the avoidance line 1 and the effectiveness evaluation model 171. Then, the effectiveness determination unit 140 calculates the final score S for each effective evaluation value of the avoidance line 1 using the calculation expression of fig. 13. The final score S indicates the likelihood of the corresponding effective evaluation value, i.e., the level of probability that the corresponding effective evaluation value can be obtained.

Si is the score of the i-th feature of the avoidance line to be calculated as the final score S. The final score S is the sum of the feature score Si of the avoidance line to be calculated. Where i is a natural number, and i is the number of feature quantities.

The effectiveness determination unit 140 determines the effective evaluation value with the highest final score S as the effective evaluation value of the avoidance line. The effectiveness determining unit 140 outputs the effectiveness evaluation value of the determined avoidance line to the effectiveness line selecting unit 150.

In the example of fig. 12, the effectiveness determination unit 140 determines the effective evaluation value 5 having the highest final score S (0.6) as the effective evaluation value of the avoidance line 1.

The effective route selection unit 150 selects the most effective avoidance line among the avoidance lines 40 as the effective line Rb based on the effective evaluation values of the respective avoidance lines 40.

Specifically, the effectiveness determination unit 140 determines the effectiveness evaluation value for each of the avoidance lines 1 to 9 in fig. 10. The effective route selecting unit 150 selects an avoidance line corresponding to the highest effective evaluation value among the effective evaluation values of the avoidance lines 1 to 9 as the effective route Rb, and outputs the selected avoidance line to the effective information constructing unit 160. There may also be a plurality of active routes Rb.

The effective information constructing unit 160 stores the effective route information 172 in the storage unit 170, the effective route information associating the traffic accident scene with the effective route Rb. At this time, the effective information constructing unit 160 stores the effective route information 172 in the storage unit 170 in the form of a knowledge base.

Specifically, the effective information constructing unit 160 converts the pair of the traffic accident scene and the effective route Rb into a form as knowledge, and stores the knowledge. For example, when the avoidance line of "fully steering the wheel to the right and fully depressing the brake" is the effective line Rb, the knowledge is set as "TrunRight: full, shake: full "such an abstract description.

Other structure

< modification 1>

Part of the functions of the route generation device 100 described in the present embodiment may be executed by another device. For example, a part of the functions of the route generation device 100 may be executed by a device such as an externally provided server.

< modification 2>

In the present embodiment, the functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the valid information construction unit 160 are realized by software. As a modification, the functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the valid information construction unit 160 may be realized by hardware.

Fig. 14 is a diagram showing the configuration of a route generation device 100 according to a modification of the present embodiment.

The route generation device 100 has an electronic circuit 909, a memory 921, an auxiliary storage 922, an input interface 930, and an output interface 940.

The electronic circuit 909 is a dedicated electronic circuit that realizes the functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the valid information construction unit 160.

Specifically, electronic circuit 909 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA, an ASIC, or an FPGA. GA is an abbreviation for Gate Array. ASIC is an abbreviation of Application Specific Integrated Circuit. FPGA is an abbreviation for Field-Programmable Gate Array.

The functions of the accident data acquisition unit 110, the traveling condition extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the valid information construction unit 160 may be implemented by 1 electronic circuit, or may be implemented by being distributed among a plurality of electronic circuits.

As another modification, some of the functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the validity information construction unit 160 may be realized by electronic circuits, and the remaining functions may be realized by software. As another modification, a part or all of the functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid line selection unit 150, and the valid information construction unit 160 may be implemented by firmware.

The processor and the electronic circuit, respectively, are also referred to as processing circuitry. That is, in the route generation device 100, the functions of the accident data acquisition unit 110, the travel situation extraction unit 120, the avoidance line generation unit 130, the validity determination unit 140, the valid route selection unit 150, and the valid information construction unit 160 are realized by processing lines.

Description of effects of the present embodiment

In the route generation system according to the present embodiment, a necessary travel situation at the time of occurrence of a traffic accident is extracted in advance from a conventional large traffic accident database. The running condition includes information such as a collision speed, a collision target, and a road surface condition. The route generation system designs an emergency avoidance travel route (emergency avoidance route) based on the extracted information. In addition, a knowledge base is constructed by using the driving condition during the traffic accident and the designed emergency avoidance line as a pair. When a dangerous traffic scene actually occurs, the vehicle condition, the obstacle condition, or the surrounding traffic condition is input from the sensor device, and the optimal emergency avoidance line can be searched for in a knowledge base constructed in advance. The knowledge base outputs an emergency avoidance route of a traffic accident most similar to the current dangerous traffic scene, and emergency avoidance can be performed.

Therefore, according to the route generation system of the present embodiment, it is possible to generate an appropriate emergency avoidance route in accordance with a traffic situation that actually occurs. Therefore, it is possible to seek to avoid an accident or reduce damage at the time of collision, and realize safe and safe automatic driving and driving assistance.

In the route generation system of the present embodiment, a validity evaluation model is generated that records the weight of giving an evaluation value from each avoidance route. Then, an effectiveness evaluation model is used to determine an effectiveness evaluation value that is optimal for the avoidance line.

Therefore, according to the route generation system of the present embodiment, it is possible to select the most appropriate emergency avoidance route obtained from the actual traffic accident scene, and it is possible to realize safer and more secure automatic driving and driving assistance.

The route generation system according to the present embodiment includes: an effective route selection unit that selects an avoidance route having the highest effect based on the determined effective evaluation value; and a knowledge base construction unit that constructs a knowledge base by associating the traffic accident scene with the emergency avoidance line.

Therefore, according to the route generation system of the present embodiment, it is possible to select the most appropriate emergency avoidance route, and it is possible to realize safer and more secure automatic driving and driving assistance.

The route generation system according to the present embodiment includes a learning data storage unit that stores a learning data set prepared by pairing the avoidance route and the effective evaluation value. At this time, first, a learning data set is prepared by subjective evaluation. The route generation system further includes a feature extraction unit that extracts a feature for evaluating the effectiveness from the learning data. The route generation system further includes an evaluation model generation unit that generates a validity evaluation model from the learning data. The evaluation model generation unit learns in advance the evaluation models of the plurality of avoidance lines by a statistical method. Then, the route generation system generates a plurality of avoidance lines based on information such as the direction and speed of the vehicle, and outputs the avoidance line having the highest effective evaluation value during execution.

Therefore, according to the route generation system of the present embodiment, a more effective effectiveness evaluation model based on an actual traffic accident scene can be generated, and an optimum avoidance route can be output even in an actually occurring traffic accident scene.

In embodiment 1 above, the description has been given of the case where each unit of each device of the route generation system is an independent functional block. However, the configuration of each device of the route generation system may not be the configuration of the above-described embodiment. The functional blocks of each device of the route generation system may have any configuration as long as the functions described in the above embodiments can be realized. Further, each device of the route generation system may be a system configured by a plurality of devices instead of 1 device.

In addition, a plurality of portions in embodiment 1 may be combined. Alternatively, some of the embodiments may be implemented. Further, the present embodiments may be implemented in any combination as a whole or in part.

That is, in embodiment 1, the optional combinations of the respective embodiments, the modifications of any of the components of the respective embodiments, and the omission of any of the components of the respective embodiments can be realized.

The above embodiments are merely preferable examples in nature, and are not intended to limit the scope of the present invention, the scope of the application of the present invention, and the scope of the application of the present invention. The above embodiment can be variously modified as necessary.

Description of the reference symbols

40: a plurality of avoidance lines; 100: a route generation device; 110: an accident data acquisition unit; 120: a driving condition extraction unit; 130: an avoidance line generation unit; 140: a validity determination unit; 150: an effective route selection unit; 160: an effective information constructing section; 170: a storage unit; 171: an effectiveness evaluation model; 172: valid route information; 200: a model generation means; 210: a data storage unit for learning; 211: a learning dataset; 220: a feature value extraction unit; 230: an evaluation model generation unit; 301: a host vehicle; 302: other vehicles; 500: a route generation system; 909: an electronic circuit; 910: a processor; 921: a memory; 922: a secondary storage device; 930: an input interface; 940: an output interface; 950: a communication device; p: a valid evaluation value; t: a characteristic amount; si: a feature quantity score; v: a feature quantity vector; s: and (6) finally scoring.