Sensor Evaluation Device

Abstract

To provide a sensor evaluation device usable in evaluation of a sensor.SOLUTION: A sensor evaluation device 1 evaluates a first sensor 5 mounted in a sensor mounting object. The sensor evaluation device includes a specific event detection unit and a recording unit. The specific event detection unit executes (a) unrecognized event detection processing of detecting an unrecognized event where second sensors 7, 9, and 11 mounted on the sensor mounting object recognize a first target but the first sensor does not recognize the first target, and (b) erroneously recognized event detection processing of detecting an erroneously recognized event where the first sensor recognizes a second target but the second sensors do not recognize the second target. When a specific event that is the unrecognized event or the erroneously recognized event is detected, the recording unit stores information concerning the specific event.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a sensor evaluation device. [Background technology]

[0002] A boundary line recognition device is described in Patent Document 1. The boundary line recognition device recognizes boundary lines by using a plurality of sensors mounted on a vehicle. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-219569 Summary of the Invention [Problem to be solved by the invention]

[0004] The performance of a sensor is affected by the mounting position of the sensor in a vehicle, weather, etc. Furthermore, the suitable mounting position may differ depending on the type of sensor. In order to determine the type of sensor, the mounting position of the sensor, etc., it is necessary to evaluate the sensor in the environment in which the sensor will be used. In one aspect of the present disclosure, it is preferable to provide a sensor evaluation device that can be used to evaluate sensors. [Means for solving the problem]

[0005] One aspect of the present disclosure is a sensor evaluation device (1) for evaluating a first sensor (5) mounted on a sensor mounted object (3). The sensor evaluation device (1) evaluates a first sensor (5) mounted on the sensor mounted object (3). The sensor evaluation device (1) evaluates a first target object by: (a) detecting a second sensor (7, 9, 11) mounted on the sensor mounted object; the distance from the sensor mounted object to the first target is less than a threshold, and the second sensor includes a camera Unrecognized Event Then, the unrecognized event detection process is performed to detect b) the first sensor recognizes a second target, and the second sensor recognizes the second target; the distance from the sensor-equipped object to the second target is equal to or greater than a threshold, and the second sensor includes a LIDAR Misrecognition Event Detect false recognition events by performing the process a specific event detection unit (17) configured to It is an unrecognized event or the above-mentioned misrecognized event.and a recording unit (19) for recording information relating to a specific event when the specific event is detected.

[0006] A sensor evaluation device according to one aspect of the present disclosure can detect a specific event when a first sensor is mounted on a sensor-equipped object. When the sensor evaluation device according to one aspect of the present disclosure detects a specific event, the sensor evaluation device can record information about the specific event. For example, a user can evaluate the first sensor by analyzing the information about the specific event. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram illustrating a configuration of a sensor evaluation device. [Figure 2] FIG. 2 is a block diagram showing the functional configuration of the sensor evaluation device. [Figure 3] 4 is a flowchart showing a process executed by the sensor evaluation device. [Figure 4] 10 is a flowchart showing an unrecognized event detection process executed by the sensor evaluation device. [Figure 5] 10 is a flowchart showing an evaluation target lane calculation process executed by the sensor evaluation device. [Figure 6] 10 is a flowchart showing an erroneous recognition event detection process executed by the sensor evaluation device. DETAILED DESCRIPTION OF THE INVENTION

[0008] Exemplary embodiments of the present disclosure will now be described with reference to the drawings. First Embodiment 1. Configuration of sensor evaluation device 1 The configuration of the sensor evaluation device 1 will be described with reference to Fig. 1 and Fig. 2. As shown in Fig. 1, the sensor evaluation device 1 is mounted on a vehicle 3. The vehicle 3 corresponds to a sensor-mounted object. The vehicle 3 is equipped with a sensor 5 to be evaluated, a millimeter-wave radar 7, a lidar 9, and a camera 11.

[0009] The evaluation target sensor 5 is, for example, a camera, a LIDAR, a millimeter-wave radar, or a sonar. The evaluation target sensor 5 corresponds to the first sensor. The millimeter-wave radar 7, the LIDAR 9, and the camera 11 correspond to the second sensor.

[0010] The sensor 5 to be evaluated, the millimeter-wave radar 7, the LIDAR 9, and the camera 11 are each capable of recognizing targets present in the vicinity of the vehicle 3. The sensor 5 to be evaluated, the millimeter-wave radar 7, the LIDAR 9, and the camera 11 have a common recognition range. Therefore, if the environment when recognizing the target is favorable, the sensor 5 to be evaluated, the millimeter-wave radar 7, the LIDAR 9, and the camera 11 can each recognize one target present in the common recognition range.

[0011] The sensor evaluation device 1 includes a microcomputer having a CPU 13 and a semiconductor memory such as a RAM or a ROM (hereinafter referred to as a memory 15). Each function of the sensor evaluation device 1 is realized by the CPU 13 executing a program stored in a non-transient physical recording medium. In this example, the memory 15 corresponds to the non-transient physical recording medium storing the program. Furthermore, the execution of this program results in the execution of a method corresponding to the program. The sensor evaluation device 1 may include one microcomputer or multiple microcomputers.

[0012] As shown in FIG. 2, the sensor evaluation device 1 includes a specific event detection unit 17, a recording unit 19, a notification unit 21, and an input unit 23. The sensor evaluation device 1 can acquire the recognition results of the target object from the evaluation target sensor 5, the millimeter wave radar 7, the lidar 9, and the camera 11.

[0013] The sensor evaluation device 1 is connected to a monitor 25, an input device 27, and a speaker 28. The monitor 25, the input device 27, and the speaker 28 are mounted on the vehicle 3. The monitor 25 can display an image based on a signal sent from the sensor evaluation device 1. An operator on board the vehicle 3 can see the image displayed on the monitor 25.

[0014] The input device 27 accepts input from an operator. The input device 27 sends a signal corresponding to the input to the sensor evaluation device 1. The input device 27 is, for example, a touch panel superimposed on the screen of the monitor 25. The input device 27 may also be various types of switches, a keyboard, etc.

[0015] The speaker 28 generates a sound based on a signal sent from the sensor evaluation device 1. The sound includes a warning sound, which will be described later. 2. Processing executed by the sensor evaluation device 1 The processing executed by the sensor evaluation device 1 will be described with reference to Fig. 3 to Fig. 6. In step 1 of Fig. 3, the specific event detection unit 17 executes unrecognized event detection processing. The unrecognized event detection processing will be described with reference to Fig. 4.

[0016] In step 11 of Fig. 4, the specific event detection unit 17 acquires target recognition results from the evaluation target sensor 5, the millimeter-wave radar 7, the LIDAR 9, and the camera 11. Based on the acquired target recognition results, the specific event detection unit 17 determines whether or not there are any unrecognized targets. An unrecognized target is a target that is recognized by any of the millimeter-wave radar 7, the LIDAR 9, and the camera 11, but not by the evaluation target sensor 5.

[0017] That is, it is determined that there is an unrecognized target when the evaluation target sensor 5 does not recognize a target at a position where an unrecognized target has been recognized by any of the millimeter-wave radar 7, the lidar 9, and the camera 11. If there is an unrecognized target, the process proceeds to step 12. If there is no unrecognized target, the unrecognized event detection process ends.

[0018] In step 12, the specific event detection unit 17 performs a process of calculating the lane to be evaluated. The process of calculating the lane to be evaluated will be described with reference to FIG. 5. In step 21 of FIG. 5, the specific event detection unit 17 determines whether or not the camera 11 has recognized a white line. If the camera 11 has recognized a white line, the process proceeds to step 22. If the camera 11 has not recognized a white line, the process proceeds to step 26.

[0019] In step 22, the specific event detection unit 17 calculates the lane to be evaluated based on the recognized white lines. The lane to be evaluated is a lane divided by white lines and is the lane on which the vehicle 3 is traveling.

[0020] In step 23, the specific event detection unit 17 determines whether at least one of the recognized white lines is a broken line. If at least one of the recognized white lines is a broken line, the process proceeds to step 24. If none of the recognized white lines is a broken line, the process proceeds to step 25.

[0021] In step 24, the specific event detection unit 17 expands the lane under evaluation to include the adjacent lanes on either side of the dashed line. In step 25, the specific event detection unit 17 determines the lane to be evaluated.

[0022] In step 26, the specific event detection unit 17 predicts the future driving trajectory of the vehicle 3 based on the yaw rate of the vehicle 3. The specific event detection unit 17 sets an area having a certain width centered on the future driving trajectory as the lane to be evaluated.

[0023] In step 27, the specific event detection unit 17 determines whether the vehicle 3 is changing lanes. The specific event detection unit 17 can determine whether the vehicle 3 is changing lanes based on, for example, the position information of the vehicle 3, map information, steering amount information, etc. If the vehicle 3 is changing lanes, the process proceeds to step 28. If the vehicle 3 is not changing lanes, the process proceeds to step 25.

[0024] In step 28, the specific event detection unit 17 expands the lane to be evaluated to include the adjacent lane into which the lane change is to occur. Returning to Fig. 4, in step 13, the specific event detection unit 17 determines whether or not the unrecognized object is in the lane to be evaluated. If the unrecognized object is in the lane to be evaluated, the process proceeds to step 14. If the unrecognized object is not in the lane to be evaluated, the unrecognized event detection process ends.

[0025] In step 14, the specific event detection unit 17 determines whether the unrecognized target is a moving object. A moving object is a target that is moving. If the unrecognized target is a moving object, the process proceeds to step 18. If the unrecognized target is not a moving object, the process proceeds to step 15. Note that if the process proceeds to step 15, the unrecognized target is a stationary object. A stationary object is a target that is stationary.

[0026] In step 15, the specific event detection unit 17 determines whether the distance from the vehicle 3 to the unrecognized target is equal to or greater than a threshold. If the distance from the vehicle 3 to the unrecognized target is less than the threshold, the process proceeds to step 16. If the distance from the vehicle 3 to the unrecognized target is equal to or greater than the threshold, the process proceeds to step 17.

[0027] In step 16, the specific event detection unit 17 determines whether the camera 11 recognizes the unrecognized object as a vehicle. If the camera 11 recognizes the unrecognized object as a vehicle, the process proceeds to step 18. If the camera 11 does not recognize the unrecognized object as a vehicle, the unrecognized event detection process ends.

[0028] Note that a positive determination in step 16 corresponds to the occurrence of an unrecognized event. The camera 11 corresponds to the second sensor. The unrecognized target corresponds to the first target. If a positive determination is made in step 16, the camera 11 recognizes the unrecognized target, and the evaluation target sensor 5 does not recognize the unrecognized target.

[0029] In step 17, the specific event detection unit 17 determines whether the millimeter-wave radar 7 and the LIDAR 9 recognize an unrecognized target. If the millimeter-wave radar 7 and the LIDAR 9 recognize an unrecognized target, the process proceeds to step 18. If the millimeter-wave radar 7 or the LIDAR 9 does not recognize an unrecognized target, the unrecognized event detection process ends.

[0030] Note that a positive determination in step 17 corresponds to the occurrence of an unrecognized event. The millimeter-wave radar 7 and LIDAR 9 correspond to the second sensor. The unrecognized target corresponds to the first target. If a positive determination is made in step 17, the millimeter-wave radar 7 and LIDAR 9 recognize the unrecognized target, and the evaluation target sensor 5 does not recognize the unrecognized target.

[0031] In step 18, the specific event detection unit 17 detects unrecognized events. 3, the specific event detection unit 17 executes the erroneous recognition event detection process in step 2. The erroneous recognition event detection process will be described with reference to FIG.

[0032] In step 31 of Figure 6, the specific event detection unit 17 acquires target recognition results from the sensor 5 to be evaluated, the millimeter-wave radar 7, the LIDAR 9, and the camera 11. The specific event detection unit 17 determines whether the sensor 5 to be evaluated has recognized a vehicle. If the sensor 5 to be evaluated has recognized a vehicle, the process proceeds to step 32. If the sensor 5 to be evaluated has not recognized a vehicle, the erroneous recognition event detection process ends. Note that the vehicle recognized by the sensor 5 to be evaluated will be referred to as the recognized vehicle hereinafter.

[0033] In step 32, the specific event detection unit 17 performs the process of calculating the lane to be evaluated. The process of calculating the lane to be evaluated is the same as that in step 12. In step 33, the specific event detection unit 17 determines whether the recognized vehicle is in the lane to be evaluated. If the recognized vehicle is in the lane to be evaluated, the process proceeds to step 34. If the recognized vehicle is not in the lane to be evaluated, the false recognition event detection process ends.

[0034] In step 34, the specific event detection unit 17 determines whether the recognized vehicle is a moving object. If the recognized vehicle is a moving object, the process proceeds to step 38. If the recognized vehicle is not a moving object, the process proceeds to step 35. Note that if the process proceeds to step 35, the recognized vehicle is a stationary object.

[0035] In step 35, the specific event detection unit 17 determines whether the distance from the vehicle 3 to the recognized vehicle is equal to or greater than a threshold. If the distance from the vehicle 3 to the recognized vehicle is less than the threshold, the process proceeds to step 36. If the distance from the vehicle 3 to the recognized vehicle is equal to or greater than the threshold, the process proceeds to step 37.

[0036] In step 36, the specific event detection unit 17 determines whether the millimeter-wave radar 7 and the camera 11 recognize the recognized vehicle. If the millimeter-wave radar 7 and the camera 11 do not recognize the recognized vehicle, the process proceeds to step 38. If the millimeter-wave radar 7 or the camera 11 recognizes the recognized vehicle, the false recognition event detection process ends. The millimeter-wave radar 7 and the camera 11 correspond to sensors other than the LIDAR 9.

[0037] Note that a positive determination in step 36 corresponds to the occurrence of an erroneous recognition event. The millimeter-wave radar 7 and the camera 11 correspond to the second sensor. The recognized vehicle corresponds to the second target. When a positive determination is made in step 36, the evaluation target sensor 5 recognizes the recognized vehicle, but the millimeter-wave radar 7 and the camera 11 do not recognize the recognized vehicle.

[0038] In step 37, the specific event detection unit 17 determines whether the millimeter-wave radar 7, LIDAR 9, and camera 11 recognize the recognized vehicle. If the millimeter-wave radar 7, LIDAR 9, and camera 11 do not recognize the recognized vehicle, the process proceeds to step 38. If the millimeter-wave radar 7, LIDAR 9, or camera 11 recognizes the recognized vehicle, the erroneous recognition event detection process ends.

[0039] Note that a positive determination in step 37 corresponds to the occurrence of an erroneous recognition event. The millimeter-wave radar 7, LIDAR 9, and camera 11 correspond to the second sensor. The recognized vehicle corresponds to the second target. When a positive determination is made in step 37, the evaluation target sensor 5 recognizes the recognized vehicle, but the millimeter-wave radar 7, LIDAR 9, and camera 11 do not recognize the recognized vehicle.

[0040] In step 38, the specific event detection unit 17 detects the false recognition event. Returning to FIG. 3, in step 3, the specific event detection unit 17 determines whether a specific event has been detected. The specific events include an unrecognized event and an erroneous recognition event. If either an unrecognized event or an erroneous recognition event is detected, a specific event is detected. If a specific event is detected, the process proceeds to step 4. If a specific event is not detected, the process ends.

[0041] In step 4, the notification unit 21 issues a notification using the monitor 25. The content of the notification may indicate, for example, that a specific event has occurred. The content of the notification may also indicate, for example, whether the specific event is an unrecognized event or an erroneous recognition event. If an unrecognized event has occurred, the content of the notification may also indicate, for example, the position of the unrecognized target. One way of indicating the position of the unrecognized target is to display the position of the unrecognized target superimposed on an image showing the scenery around the vehicle 3.

[0042] Furthermore, when an erroneous recognition event occurs, the content of the notification may, for example, indicate the location of the recognized vehicle. One way of indicating the location of the recognized vehicle is to display the location of the recognized vehicle superimposed on an image showing the scenery around the vehicle 3.

[0043] Furthermore, the notification unit 21 displays a first option and a second option on the monitor 25. The first option indicates an input (hereinafter referred to as a first input) to be made to the input device 27 when the operator determines that a specific event has actually occurred. The second option indicates an input (hereinafter referred to as a second input) to be made to the input device 27 when the operator determines that a specific event has not actually occurred.

[0044] The first input is, for example, the operator pressing the "YES" area displayed on the monitor 25. The second input is, for example, the operator pressing the "NO" area displayed on the monitor 25. In this case, the input device 27 is a touch panel superimposed on the screen of the monitor 25.

[0045] In step 5, the input unit 23 determines whether or not any input made by the operator to the input device 27 has been received. If any input has been received, the process proceeds to step 6. If no input has been received, the process proceeds to step 8.

[0046] In step 6, the input unit 23 determines whether the received input is the first input. If the received input is the first input, the process proceeds to step 7. If the received input is not the first input, the process ends.

[0047] In step 7, the recording unit 19 records the log in the memory 15. The contents of the log include, for example, information indicating the occurrence of a specific event, information indicating whether an unrecognized event has occurred, information indicating whether an erroneous recognition event has occurred, information indicating the time when the specific event occurred, information indicating the location where the specific event occurred, information indicating which sensor among the millimeter-wave radar 7, the LIDAR 9, and the camera 11 recognized the unrecognized target when an unrecognized event occurred, information indicating which sensor among the millimeter-wave radar 7, the LIDAR 9, and the camera 11 recognized the recognized vehicle when an erroneous recognition event occurred, information indicating the weather, ambient light, etc. when the specific event occurred, information indicating the state of the vehicle 3 when the specific event occurred, etc. The log corresponds to information related to the specific event.

[0048] In step 8, the input unit 23 determines whether a certain time has passed since the notification was started in step 4. If the certain time has passed, the process proceeds to step 9. If the certain time has not passed, the process proceeds to step 5.

[0049] In step 9, the notification unit 21 generates a warning sound using the speaker 28. 3. Effects of the sensor evaluation device 1 (1A) The sensor evaluation device 1 can detect a specific event. When the sensor evaluation device 1 detects a specific event, it can record a log. For example, a user can evaluate the sensor 5 to be evaluated by analyzing the recorded log.

[0050] (1B) When the sensor evaluation device 1 detects a specific event, it notifies the operator. The operator determines whether the specific event has actually occurred, and performs a first input when it determines that the specific event has actually occurred. The sensor evaluation device 1 records a log on the condition that the first input is received. Therefore, the sensor evaluation device 1 can prevent the log from being recorded when a specific event has not actually occurred.

[0051] (1C) The sensor 5 to be evaluated, the millimeter-wave radar 7, the lidar 9, and the camera 11 are mounted on the vehicle 3. Therefore, the sensor evaluation device 1 can evaluate the sensor 5 to be evaluated that is mounted on the vehicle 3.

[0052] (1D) The camera 11 can accurately recognize targets at close range. When the distance from the vehicle 3 to the unrecognized target is less than a threshold, the sensor evaluation device 1 detects an unrecognized event on the condition that the camera 11 recognizes the unrecognized target. Therefore, the sensor evaluation device 1 can detect an unrecognized event more accurately.

[0053] Furthermore, the millimeter-wave radar 7 and LIDAR 9 can accurately recognize targets at long distances. When the distance from the vehicle 3 to the unrecognized target is equal to or greater than a threshold, the sensor evaluation device 1 detects an unrecognized event on the condition that the millimeter-wave radar 7 and LIDAR 9 recognize the unrecognized target. Therefore, the sensor evaluation device 1 can detect an unrecognized event more accurately.

[0054] (1E) When the recognized vehicle is close to vehicle 3, LIDAR 9 may mistakenly recognize the exhaust gas of the recognized vehicle as a target. When the distance from vehicle 3 to the recognized vehicle is less than a threshold, the sensor evaluation device 1 detects an erroneous recognition event on the condition that no sensors other than LIDAR 9 have recognized the recognized vehicle. Therefore, the sensor evaluation device 1 can more accurately detect an erroneous recognition event. <Other embodiments> Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments and can be implemented in various modified forms.

[0055] (1) The evaluation target sensor 5, millimeter wave radar 7, LIDAR 9, and camera 11 may be mounted on a sensor-mounted object other than the vehicle 3. Examples of the sensor-mounted object other than the vehicle 3 include a fixed object installed in a position facing the road.

[0056] (2) The vehicle 3 may be equipped with a sonar in addition to or in place of some of the millimeter-wave radar 7, the lidar 9, and the camera 11. The vehicle 3 may also be equipped with a sensor unit that combines GPS and map information.

[0057] (3) If the sensor evaluation device 1 makes a positive determination in step 3, the sensor evaluation device 1 may always record a log. (4) The sensor evaluation device 1 may detect only one of an unrecognized event and an erroneous recognition event.

[0058] (5) The first sensor and the second sensor can be appropriately selected from known sensors. (6) The first target and the second target may be targets other than a vehicle. The first target and the second target may be located outside the lane being evaluated.

[0059] (7) The sensor evaluation device 1 and the method described herein may be implemented by a special-purpose computer configured by configuring a processor and memory programmed to execute one or more functions embodied in a computer program. Alternatively, the sensor evaluation device 1 and the method described herein may be implemented by a special-purpose computer configured by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the sensor evaluation device 1 and the method described herein may be implemented by one or more special-purpose computers configured by combining a processor and memory programmed to execute one or more functions with a processor configured with one or more hardware logic circuits. Furthermore, the computer program may be stored in a computer-readable non-transitory tangible recording medium as instructions to be executed by a computer. The method for implementing the functions of each unit included in the sensor evaluation device 1 does not necessarily need to include software; all of the functions may be implemented using one or more hardware devices.

[0060] (8) Multiple functions of one component in the above embodiments may be realized by multiple components, or one function of one component may be realized by multiple components. Also, multiple functions of multiple components may be realized by one component, or one function realized by multiple components may be realized by one component. Also, part of the configuration of the above embodiments may be omitted. Also, at least part of the configuration of the above embodiments may be added to or substituted for the configuration of another of the above embodiments.

[0061] (9) In addition to the sensor evaluation device 1 described above, the present disclosure can also be realized in various forms, such as a system including the sensor evaluation device 1 as a component, a program for causing a computer to function as the sensor evaluation device 1, a non-transient physical recording medium such as a semiconductor memory on which this program is recorded, a sensor evaluation method, etc.

[0062] [Technical idea disclosed in this specification] [Item 1] A sensor evaluation device (1) for evaluating a first sensor (5) mounted on a sensor-mounted object (3), a specific event detection unit (17) configured to detect a specific event that is at least one of (a) an unrecognized event in which a second sensor (7, 9, 11) mounted on the sensor mount recognizes a first target and the first sensor does not recognize the first target, and (b) an erroneous recognition event in which the first sensor recognizes a second target and the second sensor does not recognize the second target; a recording unit (19) for recording information about the specific event when the specific event detection unit detects the specific event; A sensor evaluation device comprising: [Item 2] Item 1. The sensor evaluation device according to item 1, a notification unit (21) that notifies an operator when the specific event detection unit detects the specific event; an input unit (23) for receiving input from the operator; Furthermore, The recording unit is configured to record information about the specific event on the condition that the input unit accepts the input after the alarm unit has issued the alarm. [Item 3] Item 1 or 2, the sensor evaluation device, The sensor evaluation device is a vehicle (3) equipped with a sensor. [Item 4] The sensor evaluation device according to any one of items 1 to 3, A sensor evaluation device, wherein the first sensor and the second sensor each include one or more selected from the group consisting of a camera, a lidar, a millimeter wave radar, and a sonar. [Item 5] The sensor evaluation device according to any one of items 1 to 4, A sensor evaluation device in which, when the specific event detection unit detects the unrecognized event, if the distance from the sensor-mounted object to the first target is equal to or greater than a threshold, the second sensor includes a millimeter-wave radar or a lidar, and if the distance is less than the threshold, the second sensor includes a camera. [Item 6] The sensor evaluation device according to any one of items 1 to 5, When the specific event detection unit detects the false recognition event, if the distance from the sensor-mounted object to the first target is less than a threshold, the second sensor is a sensor evaluation device including a sensor other than a LIDAR. [Explanation of symbols]

[0063] 1...sensor evaluation device, 3...vehicle, 5...sensor to be evaluated, 7...millimeter wave radar, 9...lidar, 11...camera, 17...specific event detection unit, 19...recording unit, 21...alarm unit, 23...input unit

Claims

1. A sensor evaluation device (1) for evaluating a first sensor (5) mounted on a sensor-mounted object (3), a specific event detection unit (17) configured to: (a) execute an unrecognized event detection process to detect an unrecognized event in which a second sensor (7, 9, 11) mounted on the sensor mounted object recognizes a first target, the first sensor does not recognize the first target, and the distance from the sensor mounted object to the first target is less than a threshold; and (b) execute an erroneous recognition event detection process to detect an erroneous recognition event in which the first sensor recognizes a second target, the second sensor does not recognize the second target, and the distance from the sensor mounted object to the second target is equal to or greater than a threshold; a recording unit (19) for recording information about a specific event when the specific event detection unit detects the specific event, which is the unrecognized event or the erroneous recognition event; Equipped with the second sensor in the unrecognized event includes a camera; A sensor evaluation device in which the second sensor in the false recognition event includes a lidar.

2. The sensor evaluation device according to claim 1 , a notification unit (21) that notifies an operator when the specific event detection unit detects the specific event; an input unit (23) for receiving input from the operator; Furthermore, The recording unit is configured to record information about the specific event on the condition that the input unit accepts the input after the alarm unit has issued the alarm.

3. 3. The sensor evaluation device according to claim 1, The sensor evaluation device is a vehicle (3) equipped with a sensor.

4. The sensor evaluation device according to any one of claims 1 to 3, A sensor evaluation device, wherein the first sensor and the second sensor each include one or more selected from the group consisting of a camera, a lidar, a millimeter wave radar, and a sonar.

Images