Sensor evaluation system, sensor evaluation device, and vehicle

Abstract

The present invention relates to a sensor evaluation system, a sensor evaluation device, and a vehicle. The sensor evaluation system includes: an infrastructure device provided outside the vehicle; and an information processing device. The infrastructure device includes: an infrastructure sensor that senses the environment around the infrastructure device; and an infrastructure information processing device that performs information processing. The infrastructure information processing device calculates first feature point position information based on the detection information of the infrastructure sensor, and the first feature point position information represents information related to the position of a feature point of the environment around the infrastructure device. The information processing device calculates second feature point position information based on the detection information of the sensor, and the second feature point position information represents information related to the position of a feature point of the environment around the vehicle. Then, the information processing device evaluates the function of the sensor based on the difference between the first feature point position information and the second feature point position information related to the same feature point.

Description

Technical Field

[0001] The present invention relates to a sensor evaluation system, a sensor evaluation device, and a vehicle for evaluating sensors that sense the environment around a vehicle equipped in the vehicle. Background Art

[0002] When promoting an intelligent transport system (ITS), various technologies have been proposed that make flexible use not only of information from sensors equipped in a vehicle but also of information from devices installed outside the vehicle.

[0003] Japanese Patent Laid-Open No. 2006-236094 discloses an obstacle recognition system that coordinates between a vehicle and infrastructure to detect obstacles on the road. The obstacle recognition system is simple and inexpensive, and the recognition accuracy and recognition speed are improved. The obstacle recognition system includes: a roadside imaging unit provided on the roadside for imaging the road; and a roadside transmission unit provided on the roadside for transmitting roadside image data captured by the roadside imaging unit together with characteristic information to the vehicle. In addition, the obstacle recognition system includes: a vehicle-side imaging unit mounted on the vehicle for imaging the road; and a storage unit mounted on the vehicle for storing and holding the vehicle-side image data captured by the vehicle imaging unit in association with the imaging time. Further, the obstacle recognition system includes: a correction unit mounted on the vehicle for correcting the roadside image data transmitted by the roadside transmission unit based on the characteristic information and the driving state of the own vehicle; and an identification unit for extracting vehicle-side image data captured at the same time as the corrected roadside image data obtained by the correction unit from the storage unit, and detecting and identifying obstacles based on the extracted vehicle-side image data and the corrected roadside image data.

[0004] Accurately estimating the position of the own vehicle relative to the environment around the vehicle (preceding vehicle, lane, obstacle, etc.) is important for appropriately performing various controls of the vehicle such as autonomous driving control and collision mitigation braking. In particular, in autonomous driving control, it greatly affects the performance of the self-position estimation function.

[0005] The estimation of the position of the own vehicle relative to the environment around the vehicle is performed based on the sensing information of sensors equipped in the vehicle and sensing the environment around the vehicle. Therefore, if the calibration of the sensors becomes insufficient due to factors such as deviation of the mounting position or the function of the sensors deteriorates due to aging deterioration, etc., the accuracy of the estimation of the position of the own vehicle may decline, and furthermore, the control performance may decline.

[0006] However, there have been the following problems in the past: Regarding the deterioration of the functions of such sensors and further the deterioration of the control performance, it is almost impossible to notice unless control is executed and manifested as the behavior of the vehicle.

[0007] This is because it is impossible to judge the correctness of the conventional sensing information of the sensors equipped on the vehicle by the vehicle itself, and control is executed assuming that the sensing information is correct. Especially in autonomous driving control, control is executed assuming that the position of the vehicle on the map estimated by the self-position estimation function is correct. Summary of the Invention

[0008] The present invention provides a sensor evaluation system, a sensor evaluation device, and a vehicle that can continuously evaluate the functions of sensors equipped on a vehicle and sensing the environment around the vehicle without observing the behavior of the vehicle by flexibly using the information of a device provided outside the vehicle.

[0009] A sensor evaluation system according to one aspect of the present invention is a system for evaluating sensors equipped on a vehicle and sensing the environment around the vehicle, and the system includes: an infrastructure device provided outside the vehicle; and an information processing device.

[0010] The infrastructure device includes: an infrastructure sensor for sensing the environment around the infrastructure device; and an infrastructure information processing device for performing information processing. The infrastructure information processing device executes: a first feature point detection process for detecting one or more feature points representing characteristic parts of the environment around the infrastructure device based on the detection information of the infrastructure sensor; and a first feature point position calculation process for calculating first feature point position information representing information related to the positions of the one or more feature points detected by the first feature point detection process.

[0011] The information processing device executes: a second feature point detection process for detecting one or more feature points of the environment around the vehicle based on the detection information of the sensor; a second feature point position calculation process for calculating second feature point position information representing information related to the positions of the one or more feature points detected by the second feature point detection process; a difference calculation process for calculating the difference between the first feature point position information and the second feature point position information related to the same one or more feature points; and a function evaluation process for evaluating the function of the sensor based on the magnitude of the difference.

[0012] It may be that the first feature point position information includes information on the first sensing time, where the information on the first sensing time represents the time when the infrastructure sensor detects the position information of the feature point related to the first feature point position information, and the second feature point position information includes information on the second sensing time, where the information on the second sensing time represents the time when the sensor detects the position information of the feature point related to the second feature point position information. And it may be that in the differential calculation process, the information processing device calculates the difference between the first feature point position information and the second feature point position information where the first sensing time is equal to the second sensing time.

[0013] It may be that in the function evaluation process, when the difference becomes equal to or greater than a specified threshold, the information processing device evaluates the function of the sensor as abnormal. In addition, it may be that assuming the vehicle is a vehicle that can receive remote support through the operation of a remote support device located at a separated distance, when the information processing device evaluates the function of the sensor as abnormal, it requests remote support from the remote support device.

[0014] It may be that the infrastructure information processing device stores first map information including the map and the position information of several feature points on the map. And it may be that in the first feature point position calculation process, the infrastructure information processing device calculates the distance between the feature point whose position information is stored in the first map information and the feature point whose position information is not stored in the first map information among the feature points detected by the first feature point detection process as the first feature point position information. In addition, it may be that the information processing device stores second map information including the map and the position information of several feature points on the map. And it may be that in the second feature point position calculation process, the information processing device calculates the distance between the feature point whose position information is stored in the second map information and the feature point whose position information is not stored in the second map information among the feature points detected by the second feature point detection process as the second feature point position information.

[0015] It may be that the sensor evaluation system further includes a storage device that stores map information including the map and the position information of several feature points on the map. And it may be that the infrastructure information processing device calculates the position on the map of the feature point whose position information is not stored in the map information based on the position information on the map of the feature point whose position information is stored in the map information among the feature points detected by the first feature point detection process as the first feature point position information. In addition, it may be that the information processing device calculates the position on the map of the feature point whose position information is not stored in the map information based on the position information on the map of the feature point whose position information is stored in the map information among the feature points detected by the second feature point detection process as the second feature point position information.

[0016] A sensor evaluation device according to one aspect of the present invention evaluates a sensor that is provided in a vehicle and senses the environment around the vehicle. The sensor evaluation device includes: a memory that stores a program; and a processor coupled to the memory. When executing the program, the processor performs: a process of acquiring detection information from the sensor; a process of acquiring first feature point position information from an infrastructure device provided outside the vehicle, the first feature point position information representing information related to the positions of one or more feature points of a characteristic part representing the environment around the infrastructure device; a feature point detection process of detecting one or more feature points of the environment around the vehicle based on the detection information of the sensor; a second feature point position calculation process of calculating second feature point position information representing information related to the positions of the one or more feature points detected by the feature point detection process; a difference calculation process of calculating the difference between the first feature point position information and the second feature point position information related to the same one or more feature points; and a function evaluation process of evaluating the function of the sensor based on the magnitude of the difference.

[0017] A vehicle according to one aspect of the present invention includes: a sensor that senses the environment around the vehicle; and an information processing device. The information processing device performs: a process of acquiring detection information from the sensor; a process of acquiring first feature point position information from an infrastructure device provided outside the vehicle, the first feature point position information representing information related to the positions of one or more feature points of a characteristic part representing the environment around the infrastructure device; a feature point detection process of detecting one or more feature points of the environment around the vehicle based on the detection information of the sensor; a second feature point position calculation process of calculating second feature point position information representing information related to the positions of the one or more feature points detected by the feature point detection process; a difference calculation process of calculating the difference between the first feature point position information and the second feature point position information related to the same one or more feature points; and a function evaluation process of evaluating the function of the sensor based on the magnitude of the difference.

[0018] According to the sensor evaluation system, the sensor evaluation device, and the vehicle of the present invention, for the same one or more feature points detected both around the vehicle and around the infrastructure device, the difference between the first feature point position information and the second feature point position information is calculated. Then, the function of the sensor is evaluated based on the magnitude of the difference. Thus, the function evaluation of the sensor can be continuously performed without observing the behavior of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Hereinafter, with reference to the drawings, the features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described, where like reference numerals represent like elements, wherein:

[0020] Figure 1 It is a conceptual diagram for explaining the outline of the sensor evaluation system of the present embodiment.

[0021] Figure 2 It is a conceptual diagram for explaining the outline of the evaluation of the sensor by the sensor evaluation system of the present embodiment.

[0022] Figure 3 It is a block diagram showing a configuration example of the sensor evaluation system of the first embodiment.

[0023] Figure 4 It is a flowchart showing the processing executed by the infrastructure information processing device of the present embodiment.

[0024] Figure 5 It is a conceptual diagram for explaining the first feature point position calculation processing executed by the infrastructure information processing device of the first embodiment.

[0025] Figure 6 It is a flowchart showing the processing executed by the function evaluation unit of the present embodiment.

[0026] Figure 7 It is a conceptual diagram for explaining the second feature point position calculation processing executed by the function evaluation unit of the first embodiment.

[0027] Figure 8 It is a block diagram showing a configuration example of the sensor evaluation system of the second modification of the first embodiment.

[0028] Figure 9 It is a block diagram showing a configuration example of the sensor evaluation system of the second embodiment.

[0029] Figure 10 It is a conceptual diagram for explaining the first feature point position calculation processing executed by the infrastructure information processing device of the second embodiment.

[0030] Figure 11 It is a conceptual diagram for explaining the second feature point position calculation processing executed by the function evaluation unit of the second embodiment.

[0031] Figure 12 It is a block diagram showing a configuration example of the sensor evaluation system of the third embodiment. Detailed implementation mode

[0032] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the embodiments described below, when numerical values such as the number, quantity, amount, and range of each element are mentioned, the present invention is not limited to the mentioned numerical values, unless otherwise explicitly stated or clearly determined to be that numerical value in principle. In addition, with respect to the configurations and the like described in the embodiments below, they are not necessarily essential in the present invention, unless otherwise explicitly stated or clearly determined to be that configuration or the like in principle. It should be noted that in each figure, the same or corresponding parts are denoted by the same reference numerals, and the repeated description is appropriately simplified or omitted.

[0033] 1. Summary

[0034] Figure 1 It is a conceptual diagram for explaining the summary of the sensor evaluation system 10 of the present embodiment. Figure 1 It shows a situation where the vehicle 2 is traveling on a road. The vehicle 2 can be an autonomous vehicle that performs autonomous driving through autonomous driving control. The vehicle 2 is equipped with an autonomous sensor 200, which is a sensor that senses the environment around the vehicle 2 to obtain information related to the control of the vehicle 2. The autonomous sensor 200 is, for example, a radar, a camera, a LIDAR (Light Detection And Ranging), etc. Alternatively, the autonomous sensor 200 may be composed of multiple sensors and a processing device for sensor fusion. The environment around the vehicle 2 sensed by the autonomous sensor 200 is, for example, a billboard SGN, a stopped vehicle SVC, a pedestrian PDT, a white line WL, etc. However, it is not limited to objects, and it can also be a wall, a road surface, a shape, a color, a brightness value, etc.

[0035] The sensor evaluation system 10 evaluates the autonomous sensor 200. The sensor evaluation system 10 includes: an infrastructure device 1 provided outside the vehicle 2; and an information processing device 210 that executes processing related to the evaluation of the autonomous sensor 200. The information processing device 210 is configured to be able to transmit information with the autonomous sensor 200 and the infrastructure device 1. In Figure 1 this case, the information processing device 210 is equipped in the vehicle 2. In this case, for example, the vehicle 2 and the infrastructure device 1 are equipped with communication devices, and the vehicle 2 and the infrastructure device 1 transmit information by performing communication realized by wireless. And the information processing device 210 is electrically connected or wirelessly connected to the autonomous sensor 200 and the communication device, so as to transmit information with the autonomous sensor 200 and the infrastructure device 1. The information processing device 210 equipped in the vehicle 2 is typically an ECU (Electronic Control Unit) having a memory and a processor.

[0036] However, the information processing device 210 may also be provided outside the vehicle 2. For example, the information processing device 210 may also be a server configured on the network to which the vehicle 2 and the infrastructure device 1 are connected. In this case, the information processing device 210 transmits and receives information to and from the vehicle 2 and the infrastructure device 1 via the network.

[0037] The infrastructure device 1 includes: an infrastructure sensor 100 that senses the environment around the infrastructure device 1; and an infrastructure information processing device 110 that performs information processing. The infrastructure device 1 is arranged such that the range of the environment sensed by the infrastructure sensor 100 sometimes overlaps with the range of the environment sensed by the autonomous sensor 200. For example, as Figure 1 shown, the infrastructure device 1 is provided beside the road on which the vehicle 2 travels.

[0038] The environment around the infrastructure device 1 sensed by the infrastructure sensor 100 is the same as the environment around the infrastructure device 1 sensed by the autonomous sensor 200. However, the range of the environment sensed by the infrastructure sensor 100 and the range of the environment sensed by the autonomous sensor 200 may also be different. For example, as Figure 1 shown in the top view, in the range of the environment (dashed-dotted line) sensed by the autonomous sensor 200, there are included a billboard SGN and a stopped vehicle SVC but not a pedestrian PDT. On the other hand, in the range of the environment (dotted line) sensed by the infrastructure sensor 100, all the billboards SGN, the stopped vehicle SVC, and the pedestrian PDT are included. Thus, it is sufficient that the ranges of the environments sensed by the infrastructure sensor 100 and the autonomous sensor 200 overlap, and they do not need to be the same.

[0039] The infrastructure information processing device 110 acquires detection information from the infrastructure sensor 100 and calculates first feature point position information based on the detection information, where the first feature point position information represents information related to the positions of one or more feature points of the environment around the infrastructure device 1. A feature point is a point representing a characteristic part, for example, a point that is a corner of a billboard SGN, a point that is a break in a white line WL, etc. In addition to this, a feature point may also be a boundary of a color, a brightness value, etc.

[0040] The information processing device 210 acquires detection information from the autonomous sensor 200, and calculates second feature point position information based on the detection information, where the second feature point position information represents information related to the positions of one or more feature points of the environment around the vehicle 2. Then, it acquires first feature point position information from the infrastructure device 1, and evaluates the function of the autonomous sensor 200 based on the magnitude of the difference between the first feature point position information and the second feature point position information related to the same one or more feature points. For example, when any of the calculated differences exceeds a specified threshold, it is evaluated that the function of the autonomous sensor 200 is abnormal. This is because when the difference exceeds the specified threshold, it can be considered that there is a difference between the sensing information of the infrastructure sensor 100 and the sensing information of the autonomous sensor 200. Alternatively, a graded evaluation can also be performed according to the degree of the calculated difference magnitude.

[0041] Figure 2 is a conceptual diagram for explaining the outline of the evaluation of the autonomous sensor 200 by the sensor evaluation system 10. In Figure 2 , FP represents a feature point. It should be noted that in order to distinguish multiple feature points, additional numbers are attached to the reference numerals in the drawings.

[0042] In Figure 2 , the infrastructure information processing device 110 detects four feature points, namely FP1, FP2, FP3, and FP4, from the environment sensed by the infrastructure sensor 100. Therefore, the infrastructure information processing device 110 detects first feature point position information related to the feature points FP1, FP2, FP3, and FP4. The first feature point position information is, for example, information related to the distances between the feature point FP1 and the feature points FP2, FP3, and FP4.

[0043] In Figure 2 , the information processing device 210 detects three feature points, namely FP1, FP2, and FP3, from the environment sensed by the autonomous sensor 200. Therefore, the information processing device 210 detects second feature point position information related to the feature points FP1, FP2, and FP3. The second feature point position information is, for example, information related to the distances between the feature point FP1 and the feature points FP2 and FP3.

[0044] The information processing device 210 acquires the first feature point position information calculated by the infrastructure information processing device 110. Then, it calculates the difference between the first feature point position information and the second feature point position information related to the feature points FP1, FP2, and FP3 detected by both the information processing device 210 and the infrastructure information processing device 110. On the other hand, since the information processing device 210 does not detect the feature point FP4, the first feature point position information related to the feature point FP4 is not considered.

[0045] The information processing device 210 evaluates the function of the autonomous sensor 200 based on the magnitude of the difference thus calculated.

[0046] 2. First Embodiment

[0047] 2-1. Configuration

[0048] Figure 3 FIG. is a block diagram showing a configuration example of the sensor evaluation system 10 according to the first embodiment. The sensor evaluation system 10 evaluates the autonomous sensor 200, which is a sensor provided in the vehicle 2 for sensing the environment around the vehicle 2.

[0049] The sensor evaluation system 10 includes an infrastructure device 1. The infrastructure device 1 includes: an infrastructure sensor 100 for sensing the environment around the infrastructure device 1; an infrastructure information processing device 110 for performing information processing; and a communication device 130.

[0050] The infrastructure sensor 100 detects information related to the environment around the infrastructure device 1 and outputs the detection information. The detection information is provided to include at least information indicating the positional relationship between the infrastructure device 1 and the environment around the infrastructure device 1. The infrastructure sensor 100 may be in any form as long as it can detect information indicating the positional relationship between the infrastructure device 1 and the environment around the infrastructure device 1. For example, it may be a pressure sensor that senses the pressure applied to the ground around the infrastructure device 1 and detects information indicating the positional relationship based on the sensed pressure. Similar to the autonomous sensor 200, the infrastructure sensor 100 may also be composed of a radar, a camera, a LIDAR, etc.

[0051] The infrastructure information processing device 110 is typically an electronic computer having a memory and a processor. The memory includes: a RAM (Random Access Memory) for temporarily storing data; and a ROM (ReadOnly Memory) for storing a control program executable by the processor and various data related to the control program. The processor reads the program from the memory and performs processing according to the program based on various data read from the memory. The infrastructure information processing device 110 performs processing according to the program based on the detection information obtained from the infrastructure sensor 100 and outputs the execution result. The processing performed by the infrastructure information processing device 110 includes the processing of calculating the first feature point position information described later. That is, the execution result includes the first feature point position information. Details of the processing performed by the infrastructure information processing device 110 will be described later.

[0052] Here, as data related to the processing performed by the processor, the memory of the infrastructure information processing device 110 stores first map information that is information of a map. The first map information can be either two-dimensional map information or three-dimensional map information. Moreover, the information of the positions on the map related to the first map information, which includes several feature points FP, is included in the first map information.

[0053] The communication device 130 is a device that communicates with the vehicle 2 to transmit and receive various information (communication information). The communication information transmitted by the communication device 130 at least includes information on the output result of the infrastructure information processing device 110. That is to say, it includes the first feature point position information. The communication related to the communication device 130 can be carried out in any form. For example, it can be carried out through the transmission and reception of radio waves, or the communication device 130 can be connected to a network and communicate the communication information via the network.

[0054] The vehicle 2 includes: an autonomous sensor 200 that senses the surroundings of the vehicle 2; an ECU 210 (information processing device); a vehicle state sensor 220 that detects the state of the vehicle 2; a communication device 230; and an actuator 240. The vehicle 2 performs various controls through the ECU 210. Various controls are, for example, autonomous driving control, collision mitigation braking, etc. The ECU 210 performs various processes related to the control of the vehicle 2 based on the information obtained from the autonomous sensor 200, the vehicle state sensor 220, and the communication device 230, and generates a control signal. Then, the actuator 240 realizes various controls of the vehicle 2 by operating according to the control signal.

[0055] The autonomous sensor 200 includes a radar 201, a camera 202, and a LIDAR 203. The autonomous sensor 200 can also include other sensors that sense the surroundings of the vehicle 2. Moreover, the autonomous sensor 200 can also include a processing device that performs sensor fusion on the detection information of these sensors. The autonomous sensor 200 detects information related to the environment around the vehicle 2 and outputs the detection information. For example, regarding the radar 201, the detection information is the information on the distance, angle, and speed of the target located in the irradiation direction of the radio wave with respect to the vehicle 2. Regarding the camera 202, the detection information is the information of the image data obtained by photographing the environment around the vehicle 2 and the information detected by analyzing the image data. Regarding the LIDAR 203, the detection information is the information of the point cloud data of the environment around the vehicle 2.

[0056] These detection information are provided to include at least information indicating the positional relationship between the vehicle 2 and the environment around the vehicle 2. For example, in the detection information of the camera 202, by analyzing the image data, information indicating the positional relationship between the environment around the vehicle 2 corresponding to a plurality of specific points on the image and the vehicle 2 is provided. In the detection information of the LIDAR 203, for each point on the point cloud data, information indicating the positional relationship with the vehicle 2 is provided. Alternatively, information indicating the positional relationship between the vehicle 2 and the environment around the vehicle 2 can also be provided by performing sensor fusion on the detection information of the radar 201, the camera 202, and the LIDAR 203.

[0057] The vehicle state sensor 220 detects information related to the state of the vehicle 2 and outputs the detection information. The vehicle state sensor 220 is, for example, a wheel speed sensor, a G sensor, a gyro sensor, etc. The wheel speed sensor detects the speed of the vehicle 2. The G sensor detects the acceleration of the vehicle 2. The gyro sensor detects the angular velocity or angular acceleration in the yaw direction, roll direction, and pitch direction of the vehicle 2.

[0058] The communication device 230 is a device that communicates with devices external to the vehicle 2 to transmit and receive various information (communication information). The communication device 230 is, for example, a device for vehicle-to-vehicle communication, road-to-vehicle communication, a device providing GPS (Global Positioning System) function, etc. The communication device 230 is configured to be able to communicate with the infrastructure device 1 at least via the communication device 130. In addition, among the communication information received by the communication device 230 from the infrastructure device 1 via the communication device 130 and output by the communication device 230 to the devices equipped in the vehicle 2, at least includes the first feature point position information. It should be noted that the communication related to the communication device 230 can be carried out in any form.

[0059] The ECU 210 includes a function evaluation unit 211 and a control unit 212. The function evaluation unit 211 performs a process of evaluating the function of the autonomous sensor 200. The function evaluation unit 211 transmits the evaluation result of the function of the autonomous sensor 200 to the control unit 212. Details of the process performed by the function evaluation unit 211 will be described later.

[0060] The control unit 212 performs processes related to various controls of the vehicle 2 and generates control signals. In addition, the control unit 212 performs processes corresponding to the evaluation results transmitted from the function evaluation unit 211. For example, when the evaluation result indicates that the function of the autonomous sensor 200 is abnormal, the operation of the control function based on the detection information of the autonomous sensor 200 is stopped. Or, when the evaluation result indicates that the function of the autonomous sensor 200 has deteriorated, the vehicle 2 operator is notified that the performance of the control function based on the detection information of the autonomous sensor 200 has deteriorated, or the state of the control function associated with the deterioration of the function of the autonomous sensor 200 is evaluated. The control functions based on the detection information of the autonomous sensor 200 are, for example, the self-position estimation function and the autonomous driving control.

[0061] Here, as data related to the processes performed by the function evaluation unit 211 and the control unit 212, the memory of the ECU 210 stores second map information that is information of a map. The second map information can be either two-dimensional map information or three-dimensional map information. In addition, the second map information can be either pre-provided information or information created by the control unit 212 through mapping. Moreover, the information on the positions on the map related to the second map information includes several feature points FP. The information on the positions of the feature points FP on the map can be either pre-provided information or information provided by the control unit 212 during mapping.

[0062] Note that the second map information stored in the memory of the ECU 210 and the first map information stored in the memory of the infrastructure information processing device 110 may not be the same map information. In addition, the feature points FP whose position information is included in the second map information and the feature points FP whose position information is included in the first map information may be different.

[0063] The function evaluation unit 211 and the control unit 212 can be respectively implemented as parts of the processes in the control program, or can be respectively implemented as separate processors. Or, the function evaluation unit 211 and the control unit 212 can also be respectively constituted by separate ECUs. Moreover, the control unit 212 can also be constituted by a separate ECU for each of multiple controls or a group of controls. In this case, the ECU 210 is constituted by multiple ECUs. At this time, each ECU is connected so as to be able to mutually transmit information to the extent that the information required during the execution of the process can be obtained. At least the ECU related to the function evaluation unit 211 is connected so as to be able to transmit information to each ECU constituting the control unit 212.

[0064] The actuator 240 operates according to the control signal provided from the ECU 210. The actuator 240 is composed of various actuators corresponding to functions. For example, the ECU 210 executes processing related to autonomous driving control, and control signals related to acceleration, deceleration, and steering are transmitted to the actuator 240. Then, the actuator 240 operates according to the control signal, thereby performing autonomous driving of the vehicle.

[0065] 2-2. Processing

[0066] 2-2-1. Infrastructure Information Processing Device

[0067] Figure 4 It is a flowchart showing the processing executed by the infrastructure information processing device 110. Figure 4 The processing shown can be repeatedly executed at a prescribed cycle, or the execution of the processing can be started according to specific conditions. For example, the execution of the processing can be started with the vehicle 2 approaching the infrastructure device 1 to a certain extent as a condition. In this case, the start of the execution of the processing can be judged based on the detection information of the infrastructure sensor 100 and the communication information between the vehicle 2 and the infrastructure device 1.

[0068] In step S100 (first feature point detection processing), the infrastructure information processing device 110 detects one or more feature points FP of the environment around the infrastructure device 1 based on the detection information of the infrastructure sensor 100. Here, among the detected feature points FP, the feature points FP whose position information is included in the first map information (hereinafter, also referred to as "first registered feature points") and the feature points FP whose position information is not included in the first map information (hereinafter, also referred to as "first unregistered feature points") are included.

[0069] Here, the detection method of the feature point FP can vary according to the form of the infrastructure sensor 100 and the content of the detection information. For example, when the detection information is provided as image data, the feature point FP is detected by image analysis. When the detection information is provided as point cloud data, the feature point FP is detected by shape recognition. Or, it can also be that these detections are performed in the infrastructure sensor 100, and the feature point FP is provided as the detection information.

[0070] In addition, it is desirable that the position of the detected feature point FP on the map hardly changes with time. For example, it is desirable that the feature point FP is related to a fixed landmark (billboard, pillar, etc.) or a stationary landmark (stopped vehicle, etc.). This is to reduce the error associated with the change in position over time in the differential calculation processing described later.

[0071] After step S100, the processing proceeds to step S110.

[0072] In step S110 (first feature point position calculation process), the infrastructure information processing device 110 calculates information related to the positions of one or more feature points FP detected in step S100 (first feature point position information). Figure 5 It is a conceptual diagram for explaining the first feature point position calculation process. Figure 5 It shows a case where the infrastructure information processing device 110 detects four feature points, namely FP1, FP2, FP3, and FP4, in step S100. Here, the feature point FP1 is the first unregistered feature point, and the feature points FP2, FP3, and FP4 are the first registered feature points. In step S110, the infrastructure information processing device 110 calculates the distances between the feature point FP1, which is the first unregistered feature point, and the feature points FP2, FP3, and FP4, which are the first registered feature points, as the first feature point position information.

[0073] More specifically, as follows. The infrastructure information processing device 110 respectively provides position vectors a1, a2, a3, and a4 based on the detection information of the infrastructure sensor 100 with the position of the infrastructure device 1 as the base point for the detected feature points FP1, FP2, FP3, and FP4. Then, the differences between the position vectors, namely a2 - a1, a3 - a1, and a4 - a1, are used as the first feature point position information.

[0074] 2 - 2 - 2. Function evaluation unit

[0075] Figure 6 It is a flowchart showing the processing performed by the function evaluation unit 211. Figure 6 The shown processing can be repeatedly executed at a specified cycle or the execution of the processing can be started according to specific conditions.

[0076] In step S200 (second feature point detection process), the function evaluation unit 211 detects one or more feature points FP in the environment around the vehicle 2 based on the detection information of the autonomous sensor 200. Here, among the detected feature points FP, there are feature points FP whose position information is included in the second map information (hereinafter, also referred to as "second registered feature points") and feature points FP whose position information is not included in the second map information (hereinafter, also referred to as "second unregistered feature points").

[0077] Here, the detection method of the feature point FP can vary according to the form of the autonomous sensor 200 and the content of the detection information. Alternatively, it can also be that the feature point FP is detected in the autonomous sensor 200 and provided as the detection information.

[0078] After step S200, the processing proceeds to step S210.

[0079] In step S210 (second feature point position calculation process), the function evaluation unit 211 calculates information related to the positions of one or more feature points FP detected in step S200 (second feature point position information). Figure 7 It is a conceptual diagram for explaining the second feature point position calculation process. Figure 7 It shows a case where the function evaluation unit 211 detects three feature points, namely FP1, FP2, and FP3, in step S200. Here, the feature point FP1 is the second unregistered feature point, and the feature points FP2 and FP3 are the second registered feature points. In step S210, the function evaluation unit 211 calculates the distances between the feature point FP1, which is the second unregistered feature point, and the feature points FP2 and FP3, which are the second registered feature points, as the second feature point position information.

[0080] More specifically, as follows. The function evaluation unit 211 respectively provides position vectors b1, b2, and b3 based on the detection information of the autonomous sensor 200 with the position of the vehicle 2 as the reference point for the detected feature points FP1, FP2, and FP3. Then, the differences between the position vectors, that is, b2 - b1 and b3 - b1, are used as the second feature point position information.

[0081] After step S210, the process proceeds to step S220.

[0082] In step S220 (difference calculation process), the function evaluation unit 211 calculates the difference between the first feature point position information and the second feature point position information related to the same one or more feature points FP.

[0083] More specifically, as follows. Assume that the function evaluation unit 211 obtains the first feature point position information described in Figure 5 and calculates the second feature point position information described in Figure 7 . Here, assume that Figure 5 and Figure 7 show that the feature points FP1, FP2, and FP3 are the same. Therefore, the function evaluation unit 211 calculates the difference between the first feature point position information and the second feature point position information related to the feature points FP1, FP2, and FP3. That is, it calculates the difference Δ2 between a2 - a1 and b2 - b1 and the difference Δ3 between a3 - a1 and b3 - b1. This is given by the following equations (1) and (2). In the following equations, ||v|| represents the Euclidean norm of the vector v.

[0084] [Equation 1]

[0085] Δ2 = ||(a2 - a1) - (b2 - b1)|| …(1)

[0086] [Formula 2]

[0087] Δ3 = ||(a3 - a1) - (b3 - b1)|| …(2)

[0088] It should be noted that since the feature points FP detected by the function evaluation unit 211 do not include the feature points FP that are the same as the feature point FP4 detected by the infrastructure information processing device 110, in the differential calculation process, a4 - a1 of the first feature point position information is not considered.

[0089] Refer again to Figure 6 . After step S220, the process proceeds to step S230.

[0090] In step S230 (function evaluation process), the function evaluation unit 211 evaluates the function of the autonomous sensor 200 based on the magnitude of the difference calculated in step S220. When the difference calculated by the function evaluation unit 211 in step S220 becomes equal to or greater than a specified threshold k1, the function of the autonomous sensor 200 is evaluated as abnormal. The threshold k1 is a value provided to the program in advance and is determined through compliance experiments of the sensor evaluation system 10, etc.

[0091] More specifically, as follows. Assume that in step S220, the function evaluation unit 211 calculates the differences Δ2 and Δ3 represented by formulas (1) and (2). The function evaluation unit 211 compares the differences Δ2 and Δ3 with the threshold k1 respectively. Then, when either of Δ2 and Δ3 becomes equal to or greater than the threshold k1, the function of the autonomous sensor 200 is evaluated as abnormal. Alternatively, it is also possible to calculate the sum of the squares of the differences Δ2 and Δ3, and when the sum of the squares becomes equal to or greater than the threshold k1, the function of the autonomous sensor 200 is evaluated as abnormal.

[0092] Through the processing described above, the function evaluation of the autonomous sensor 200 can be continuously performed without observing the behavior of the vehicle 2. In addition, the control unit 212 can judge whether the control function based on the detection information of the autonomous sensor 200 can work and perform performance evaluation by obtaining the evaluation result of the function of the autonomous sensor 200.

[0093] 2 - 3. Variation

[0094] The sensor evaluation system 10 of the first embodiment can also adopt a variation as follows.

[0095] 2 - 3 - 1. Variation 1

[0096] The first feature point position information may also include information on the time when the infrastructure sensor 100 detects the position of the feature point related to the first feature point position information (hereinafter also referred to as "the first sensing time"). In addition, the second feature point position information may also include information on the time when the autonomous sensor 200 detects the position of the feature point related to the second feature point position information (hereinafter also referred to as "the second sensing time"). Moreover, the function evaluation unit 211 may also calculate the difference between the first feature point position information and the second feature point position information where the first sensing time is equal to the second sensing time in the differential calculation process.

[0097] Thus, the feature point FP whose position changes over time (for example, the feature point FP related to a pedestrian) can be used as the detection target. Moreover, the function evaluation unit 211 can evaluate the function of the autonomous sensor 200 based on the differential data calculated at consecutive times in the function evaluation process. For example, the difference can be calculated for the data of 10 seconds of the first feature point position information and the second feature point position information related to a specific feature point FP, and the function related to the tracking of the autonomous sensor 200 can be evaluated based on the magnitude of the difference.

[0098] 2-3-2. Modification Example 2

[0099] It may also be assumed that the vehicle 2 is a vehicle that can receive remote support through the operation of a remote support device located at a separated distance. In the case where the function evaluation unit 211 evaluates the function of the autonomous sensor 200 as abnormal in the function evaluation process, the ECU 210 requests remote support from the remote support device.

[0100] Figure 8 It is a block diagram showing a configuration example of the sensor evaluation system 10 of Modification Example 2. In Figure 8 the shown configuration, compared with the configuration shown in Figure 3 a remote support device 3 is provided. The remote support device 3 is configured to be able to communicate with the vehicle 2 to transmit and receive various information (communication information). The remote support device 3 performs remote support for the vehicle 2 through the transmission and reception of communication information. The remote support performed by the remote support device 3 is, for example, remote driving of the vehicle 2. In this case, the communication information sent by the remote support device 3 to the vehicle 2 includes driving operation information for performing remote driving. The communication information received by the remote support device 3 from the vehicle 2 includes the driving image of the vehicle 2.

[0101] When the function evaluation unit 211 evaluates the function of the autonomous sensor 200 as abnormal during the function evaluation process, the ECU 210 outputs a signal indicating a request for remote support to the communication device 230. Then, the remote support device 3 receives the request for remote support from the vehicle 2 via the communication device 230 and performs remote support for the vehicle 2. For example, the remote driving of the vehicle 2 by the remote support device 3 is started.

[0102] Thereby, it is possible to perform remote support for the vehicle 2 at an appropriate timing without always monitoring the state of the vehicle 2. For example, during the autonomous driving control of the vehicle 2, when the performance of the autonomous driving control cannot be sufficiently obtained due to the deterioration of the function of the autonomous sensor 200, the deterioration of the function of the autonomous sensor 200 can be sensed and a request for remote support can be made.

[0103] 3. Second Embodiment

[0104] Hereinafter, the second embodiment will be described. However, matters that have already been described in the foregoing will be appropriately omitted.

[0105] 3-1. Configuration

[0106] Figure 9 It is a block diagram showing a configuration example of the sensor evaluation system 10 of the second embodiment. In the Figure 9 shown configuration, compared with the Figure 3 shown configuration, a storage device 4 is provided. The storage device 4 stores a map database MDB as a database related to map information. In addition, the map database MDB includes information on the positions of several feature points FP on the map.

[0107] The storage device 4 is configured to be able to communicate with the vehicle 2 and the infrastructure device 1. The storage device 4 is typically the memory of a server configured on a network. The vehicle 2 and the infrastructure device 1 acquire map information from the storage device 4 via the network. The function evaluation unit 211 and the infrastructure information processing device 110 perform processing based on the map information acquired from the storage device 4. Therefore, in the sensor evaluation system 10 of the second embodiment, the memories of the ECU 210 and the infrastructure information processing device 110 may not store map information independently.

[0108] 3-2. Processing

[0109] In the sensor evaluation system 10 of the second embodiment, the calculated first feature point position information and the second feature point position information are different from those of the sensor evaluation system 10 of the first embodiment. Hereinafter, regarding the processing performed by the infrastructure information processing device 110 and the function evaluation unit 211 of the second embodiment, the differences from the sensor evaluation system 10 of the first embodiment will be focused on and described.

[0110] 3-2-1. Infrastructure Information Processing Device

[0111] The flowchart of the processing executed by the infrastructure information processing device 110 of the second embodiment is the same as Figure 4 . However, the content of the processing executed in step S110 (first feature point position calculation processing) is different from that described in the first embodiment. In addition, in step S100 (first feature point detection processing), the feature points FP detected by the infrastructure information processing device 110 include the feature points FP (hereinafter, also referred to as "registered feature points") whose positions on the map are stored in the map database MDB, and the feature points FP (hereinafter, also referred to as "unregistered feature points") whose positions on the map are not stored in the map database MDB.

[0112] Figure 10 is a conceptual diagram for explaining the first feature point position calculation processing executed by the infrastructure information processing device 110 of the second embodiment. In Figure 10 , a case is shown where the infrastructure information processing device 110 detects four feature points, namely FP1, FP2, FP3, and FP4, in the first feature point detection processing. Here, the feature points FP1 and FP4 are unregistered feature points, and the feature points FP2 and FP3 are registered feature points. In the first feature point position calculation processing, the infrastructure information processing device 110 calculates the positions of the feature points FP1 and FP4, which are unregistered feature points, on the map as the first feature point position information.

[0113] More specifically, as follows. The infrastructure information processing device 110 provides the position vectors p2 and p3 of the feature points FP2 and FP3, which are registered feature points, based on the map information obtained from the storage device 4. Then, based on the position vectors p2 and p3 and the detection information of the infrastructure sensor 100 related to the feature points FP2 and FP3, the position of the infrastructure device 1 on the map is estimated. Alternatively, the position of the infrastructure device 1 on the map may be provided to the program in advance. Then, based on the position of the infrastructure device 1 on the map and the detection information of the infrastructure sensor 100 related to the feature points FP1 and FP4, the position vectors s1 and s4 of the feature points FP1 and FP4, which are unregistered feature points, are calculated as the first feature point position information.

[0114] 3-2-2. Function Evaluation Unit

[0115] The flowchart of the processing executed by the function evaluation unit 211 of the second embodiment is the same as Figure 6The same. However, the content of the processing performed in step S210 (second feature point position calculation process) and step S220 (differential calculation process) is different from the content described in the first embodiment. In addition, in step S200 (second feature point detection process), the feature points FP detected by the function evaluation unit 211 include registered feature points and unregistered feature points.

[0116] Figure 11 FIG. is a conceptual diagram for explaining the second feature point position calculation process performed by the function evaluation unit 211 in the second embodiment. In Figure 11 shows a case where the function evaluation unit 211 detects three feature points, namely FP1, FP2, and FP3, in the second feature point detection process. Here, the feature point FP1 is an unregistered feature point, and the feature points FP2 and FP3 are registered feature points. In the second feature point position calculation process, the function evaluation unit 211 calculates the position of the feature point FP1, which is an unregistered feature point, on the map as the second feature point position information.

[0117] More specifically, as follows. The function evaluation unit 211 provides the position vectors p2 and p3 of the feature points FP2 and FP3, which are registered feature points, based on the map information obtained from the storage device 4. Then, based on the position vectors p2 and p3 and the detection information of the autonomous sensor 200 related to the feature points FP2 and FP3, the position of the vehicle 2 on the map is estimated. Then, based on the position of the vehicle 2 on the map and the detection information of the autonomous sensor 200 related to the feature point FP1, the position vector t1 of the feature point FP1, which is an unregistered feature point, is calculated as the second feature point position information.

[0118] In the differential calculation process, the function evaluation unit 211 calculates the difference between the first feature point position information and the second feature point position information related to the same one or more feature points FP.

[0119] More specifically, as follows. Assume that the function evaluation unit 211 obtains the first feature point position information described in Figure 10 from the infrastructure device 1 and calculates the second feature point position information described in Figure 11 . Here, assume that Figure 10 and Figure 11 the feature points FP1, FP2, and FP3 shown in are the same. Therefore, the function evaluation unit 211 calculates the difference between the first feature point position information and the second feature point position information related to the feature points FP1, FP2, and FP3. That is, it calculates the difference Δ between s1 and t1. This is given by the following equation (3).

[0120] [Equation 3]

[0121] Δ = ||s1 - t1|| …(3)

[0122] Note that, since the feature points FP detected by the function evaluation unit 211 do not include the feature points FP that are the same as the feature point FP4 detected by the infrastructure information processing device 110, in the differential calculation process, s4 of the first feature point position information is not considered.

[0123] Thus, in the sensor evaluation system 10 of the second embodiment, the calculated first feature point position information and second feature point position information are different from those of the sensor evaluation system 10 of the first embodiment. However, the same effect as that of the sensor evaluation system 10 of the first embodiment can be provided.

[0124] 3-3. Variation

[0125] The sensor evaluation system 10 of the second embodiment may also adopt a modified scheme as follows in the same manner as the first embodiment.

[0126] 3-3-1. Variation 1

[0127] The first feature point position information may also include the time (hereinafter, also referred to as "first sensing time") when the infrastructure sensor 100 detects the position of the feature point related to the first feature point position information. In addition, the second feature point position information may also include the time (hereinafter, also referred to as "second sensing time") when the autonomous sensor 200 detects the position of the feature point related to the second feature point position information. Further, the function evaluation unit 211 may calculate the difference between the first feature point position information and the second feature point position information with the first sensing time being equal to the second sensing time in the differential calculation process.

[0128] 3-3-2. Variation 2

[0129] It may also be assumed that the vehicle 2 is a vehicle that can receive remote support through the operation of a remote support device located at a separated distance. When the function evaluation unit 211 evaluates the function of the autonomous sensor 200 as abnormal in the function evaluation process, the ECU 210 requests remote support from the remote support device.

[0130] 4. Third Embodiment

[0131] Hereinafter, the third embodiment will be described. However, matters that have been described in the foregoing will be appropriately omitted.

[0132] 4-1. Configuration

[0133] The third embodiment shows a case where the autonomous sensor 200 provided in the vehicle 2 is evaluated by a sensor evaluation device. Figure 12It is a block diagram showing a configuration example of the sensor evaluation system 10 according to the third embodiment. The sensor evaluation system 10 includes a sensor evaluation device 5. The configuration of the infrastructure device 1 is the same as the configuration described in the first embodiment. Except for the ECU 210, the configuration of the vehicle 2 is the same as the configuration described in the first embodiment. The ECU 210 of the third embodiment does not include the function evaluation unit 211 described in the first embodiment.

[0134] The sensor evaluation device 5 evaluates the autonomous sensor 200 equipped on the vehicle 2. The sensor evaluation device 5 is configured to be able to communicate with the vehicle 2 and the infrastructure device 1 to send various information (communication information). The sensor evaluation device 5 includes a memory and a processor that executes a program. The sensor evaluation device 5 is typically a server configured on the network to which the vehicle 2 and the infrastructure device 1 are connected.

[0135] The sensor evaluation device 5 obtains information on the execution result of the infrastructure information processing device 110 from the infrastructure device 1 (including the first feature point position information). In addition, it obtains the detection information of the autonomous sensor 200 from the vehicle 2. Then, the sensor evaluation device 5 outputs the evaluation result of the autonomous sensor 200 by the processor executing the same processing as the processing described in Figure 6 The evaluation result output is transmitted to the ECU 210 via communication, and the control unit 212 executes the processing corresponding to the evaluation result.

[0136] In this way, in the third embodiment, the evaluation of the autonomous sensor 200 equipped on the vehicle 2 is performed by the sensor evaluation device 5. In the third embodiment, the same effect as the first embodiment can also be provided.

[0137] 4 - 3. Variation

[0138] The sensor evaluation system 10 of the third embodiment can also adopt the following modified schemes in the same way as the first embodiment.

[0139] 4 - 3 - 1. Variation 1

[0140] The first feature point position information may also include the time (hereinafter, also referred to as the "first sensing time") when the infrastructure sensor 100 detects the position of the feature point related to the first feature point position information. In addition, the second feature point position information may also include the time (hereinafter, also referred to as the "second sensing time") when the autonomous sensor 200 detects the position of the feature point related to the second feature point position information. And the processor may also calculate the difference between the first feature point position information and the second feature point position information with the same first sensing time and second sensing time in the differential calculation process.

[0141] 4-3-2. Variant Example 2

[0142] Alternatively, assume that the vehicle 2 is a vehicle that can receive remote support through the operation of a remote support device located at a separated distance. When the processor evaluates the function of the autonomous sensor 200 as abnormal in the function evaluation process, the ECU 210 requests remote support from the remote support device.

[0143] 5. Effects

[0144] As described above, according to the sensor evaluation system 10, the sensor evaluation device 5, and the vehicle 2 of the present embodiment, the function evaluation of the autonomous sensor 200 can be continuously performed without observing the behavior of the vehicle 2. For example, assume that due to the deterioration of the function of the autonomous sensor 200, the position of the vehicle itself relative to the stop line cannot be well estimated during the autonomous driving control of the vehicle 2. At this time, the deterioration of the function of the autonomous sensor 200 can be detected without observing the behavior that the vehicle 2 does not stop well in front of the stop line.

[0145] In addition, the control unit 212 can determine whether the control function based on the detection information of the autonomous sensor 200 can work and evaluate the performance by obtaining the evaluation result of the function of the autonomous sensor 200.

Claims

1. A sensor evaluation system for evaluating a sensor equipped in a vehicle and sensing the environment around the vehicle, the sensor evaluation system being characterized by comprising: An infrastructure device, disposed outside the vehicle; and an information processing device, The infrastructure device includes: an infrastructure sensor for sensing the environment around the infrastructure device; and an infrastructure information processing device for performing information processing, The infrastructure information processing device executes: storing first map information, the first map information including information on the positions of at least one feature point, a first feature point detection process for detecting a plurality of feature points representing characteristic portions of the environment around the infrastructure device based on the detection information of the infrastructure sensor; and a first feature point position calculation process for calculating, as first feature point position information, the distances between registered feature points whose position information is stored in the first map information and unregistered feature points whose position information is not stored in the first map information among the plurality of feature points detected by the first feature point detection process, The information processing device executes: storing second map information, the second map information including information on the positions of at least one feature point, a second feature point detection process for detecting a plurality of the feature points of the environment around the vehicle based on the detection information of the sensor; a second feature point position calculation process for calculating, as second feature point position information, the distances between registered feature points whose position information is stored in the second map information and unregistered feature points whose position information is not stored in the second map information among the plurality of feature points detected by the second feature point detection process; a difference calculation process for calculating the difference between the first feature point position information and the second feature point position information for combinations of the registered feature points and the unregistered feature points that are the same; and a function evaluation process for evaluating the function of the sensor based on the magnitude of the difference.

2. The sensor evaluation system according to claim 1, wherein the first feature point position information includes information on a first sensing time, the information on the first sensing time indicating the time when the infrastructure sensor detected the position information of the feature point related to the first feature point position information, the second feature point position information includes information on a second sensing time, the information on the second sensing time indicating the time when the sensor detected the position information of the feature point related to the second feature point position information, In the difference calculation process, the information processing device calculates the difference between the first feature point position information and the second feature point position information for which the first sensing time and the second sensing time are equal.

3. The sensor evaluation system according to claim 1 or 2, wherein in the function evaluation process, the information processing device evaluates the function of the sensor as abnormal when the difference becomes equal to or greater than a prescribed threshold.

4. The sensor evaluation system according to claim 1 or 2, wherein The vehicle is a vehicle that can receive remote support through the operation of a remote support device located at a separated distance. In the function evaluation process, when the information processing device evaluates the function of the sensor as abnormal, the information processing device requests the remote support device for the remote support.

5. A sensor evaluation device that evaluates a sensor equipped in a vehicle and senses the environment around the vehicle, the sensor evaluation device being characterized by comprising: A memory that stores a program; and A processor coupled to the memory, The processor executes, when the program is executed: A process of acquiring detection information from the sensor; A process of acquiring first feature point position information from an infrastructure device provided outside the vehicle, the first feature point position information representing the distance between a registered feature point of a plurality of feature points representing characteristic parts of the environment around the infrastructure device and a non-registered feature point whose position information is not stored in first map information possessed by the infrastructure device; A feature point detection process of detecting one or more of the feature points of the environment around the vehicle based on the detection information of the sensor; A second feature point position calculation process of calculating second feature point position information, the second feature point position information representing the distance between a registered feature point of a plurality of feature points detected by the feature point detection process and a non-registered feature point whose position information is not stored in second map information possessed by the sensor evaluation device; A difference calculation process of calculating the difference between the first feature point position information and the second feature point position information for combinations of the registered feature points and the non-registered feature points that are the same; And A function evaluation process of evaluating the function of the sensor based on the magnitude of the difference.

6. The sensor evaluation device according to claim 5, wherein The first feature point position information includes information on a first sensing time, the information on the first sensing time representing the time when the infrastructure device detects the position information of the feature point related to the first feature point position information, The second feature point position information includes information on a second sensing time, the information on the second sensing time representing the time when the sensor detects the position information of the feature point related to the second feature point position information, In the difference calculation process, the processor calculates the difference between the first feature point position information and the second feature point position information for which the first sensing time and the second sensing time are equal.

7. The sensor evaluation device according to claim 5 or 6, wherein In the function evaluation process, when the difference becomes equal to or greater than a prescribed threshold, the processor evaluates the function of the sensor as abnormal.

8. The sensor evaluation device according to claim 5 or 6, wherein The vehicle is a vehicle that can receive remote support through the operation of a remote support device located at a separated distance. In the function evaluation process, when the function of the sensor is evaluated as abnormal, the remote support is requested from the remote support device.

9. A vehicle, characterized in that, It includes: a sensor that senses the environment around the vehicle; and an information processing device, The information processing device performs: a process of obtaining detection information from the sensor; a process of obtaining first feature point position information from an infrastructure device provided outside the vehicle, where the first feature point position information represents the distance between a registered feature point whose position information of a plurality of feature points representing characteristic parts of the environment around the infrastructure device is stored in first map information possessed by the infrastructure device and an unregistered feature point whose position information is not stored in the first map information; a feature point detection process that detects one or more of the feature points of the environment around the vehicle based on the detection information of the sensor; a second feature point position calculation process that calculates second feature point position information, where the second feature point position information represents the distance between a registered feature point whose position information of a plurality of feature points detected by the feature point detection process is stored in second map information possessed by a sensor evaluation device and an unregistered feature point whose position information is not stored in the second map information; a difference calculation process that calculates the difference between the first feature point position information and the second feature point position information where the combinations of the registered feature points and the unregistered feature points are the same; and a function evaluation process that evaluates the function of the sensor based on the magnitude of the difference.

10. The vehicle according to claim 9, wherein the first feature point position information includes information on a first sensing time, and the information on the first sensing time represents the time when the infrastructure device detects the position information of the feature point related to the first feature point position information; the second feature point position information includes information on a second sensing time, and the information on the second sensing time represents the time when the sensor detects the position information of the feature point related to the second feature point position information; in the difference calculation process, the information processing device calculates the difference between the first feature point position information and the second feature point position information where the first sensing time and the second sensing time are equal.

11. The vehicle according to claim 9 or 10, wherein in the function evaluation process, the information processing device evaluates the function of the sensor as abnormal when the difference becomes equal to or greater than a specified threshold.

12. The vehicle according to claim 9 or 10, wherein the vehicle can receive remote support through the operation of a remote support device located at a separated distance, in the function evaluation process, the information processing device requests the remote support from the remote support device when the function of the sensor is evaluated as abnormal.

Images