Abnormality determination system for a vehicle's peripheral monitoring sensor

Abstract

The present application aims to unify the abnormality determination of the surrounding monitoring sensor of a vehicle. The abnormality determination system includes: a position information acquisition unit that acquires first position information indicating the position of an object when the surrounding monitoring sensor detects the object; a received power acquisition unit that acquires first received power received when the object is detected; a coincidence degree determination unit that determines the coincidence degree of the first position information and second position information indicating the position of each of a plurality of arbitrary objects, with reference to a database in which the second position information is associated with second received power received when the arbitrary object is detected; an axis deviation determination unit that determines axis deviation based on the deviation of the first position information and the second position information within a permissible range of the coincidence degree; and a dirt determination unit that determines dirt based on the deviation of the first received power and the second received power associated with the second position information within the permissible range of the coincidence degree.

Description

Technical Field

[0001] This invention relates to an anomaly detection system for a vehicle's peripheral monitoring sensor. Background Technology

[0002] Previously, there was a known radar device that compared the calculated distance to a building, obtained from the transmitted radio waves and the received reflected waves, with a reference distance calculated by a navigation device based on current location data and map data, to determine whether the distance detection performance was within the normal range (for example, see Patent Document 1).

[0003] Patent Document 1: Japanese Patent Application Publication No. 2005-134231 Summary of the Invention

[0004] However, in the technology described in the aforementioned patent documents, the abnormality of the distance detection performance of the radar device is determined, but it is necessary to separately implement the abnormality detection based on the dirt of the radar device, which has the problem of not being able to uniformly determine the abnormality of the radar device.

[0005] Therefore, the purpose of this invention is to provide an anomaly detection system for vehicle perimeter monitoring sensors that can uniformly perform anomaly detection.

[0006] The purpose of this invention is as follows:

[0007] (1) An anomaly detection system for a vehicle's surrounding monitoring sensor, comprising:

[0008] The system includes: a location information acquisition unit that acquires first location information indicating the location of an object detected by a perimeter monitoring sensor mounted on a vehicle; a power acquisition unit that acquires first received power received by the perimeter monitoring sensor when it detects the object; a consistency determination unit that determines the consistency between the first location information and the second location information by referring to a database that stores a corresponding association between second location information indicating the locations of multiple arbitrary objects and the second received power received by the perimeter monitoring sensor of any vehicle when the arbitrary object is detected; an axle deviation determination unit that determines the axle deviation of the perimeter monitoring sensor based on the deviation between the first location information and the second location information within an acceptable range of consistency with the first location information; and a dirt determination unit that determines the dirt on the perimeter monitoring sensor based on the deviation between the first received power and the second received power, wherein the second received power is associated with the second location information within an acceptable range of consistency with the first location information.

[0009] (2) The anomaly detection system for the vehicle's surrounding monitoring sensors as described in (1) above further comprises:

[0010] The axis deviation correction unit corrects the axis deviation of the peripheral monitoring sensor in such a way that the position of the object detected by the peripheral monitoring sensor becomes the position represented by the second position information.

[0011] (3) An anomaly determination system for the vehicle's surrounding monitoring sensors as described in (1) or (2) above, wherein,

[0012] The database stores the second location information and the second received power in association with vehicle model and year. The consistency determination unit establishes the corresponding association of the second location information with the same vehicle model and year as the vehicle equipped with the surrounding monitoring sensor as the object of anomaly determination, and determines the consistency between the first location information and the second location information.

[0013] (4) The anomaly detection system for the vehicle's surrounding monitoring sensors as described in (1) or (2) above further comprises:

[0014] The comparison unit compares the first location information obtained from the arbitrary vehicle with the location information of an object registered in a high-precision 3D map, and extracts the object whose location information corresponds to the first location information and is registered in the high-precision 3D map; and the storage processing unit treats the extracted object as the arbitrary object, stores the second location information and the second received power in the database, stores the location information of the object registered in the high-precision 3D map as the second location information, and stores the distribution of the first received power compared during extraction as the second received power.

[0015] (5) The anomaly detection system for the vehicle's surrounding monitoring sensors as described in (4) above, wherein,

[0016] The storage processing unit stores the second location information and the second received power in association with the vehicle model and year obtained from the arbitrary vehicle.

[0017] Invention Effects

[0018] According to the present invention, an anomaly detection system for vehicle perimeter monitoring sensors that can uniformly perform anomaly detection is provided. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the anomaly detection system of the vehicle's surrounding monitoring sensors.

[0020] Figure 2 This is a schematic diagram showing the structure of a vehicle.

[0021] Figure 3 This is a schematic diagram illustrating the structure of a server.

[0022] Figure 4 This is a schematic diagram showing the state of objects detected by surrounding monitoring sensors.

[0023] Figure 5 It is a diagram showing the structure of radar target clouds.

[0024] Figure 6 This is a schematic diagram representing the function block of the ECU processor in a vehicle.

[0025] Figure 7 This is a schematic diagram illustrating the method for determining shaft deviation based on the shaft deviation determination unit.

[0026] Figure 8 This is a schematic diagram illustrating a method for determining dirt based on a dirt determination unit.

[0027] Figure 9 This is a schematic diagram representing the functional blocks of a server's processor.

[0028] Figure 10 This is a flowchart illustrating the processing of the anomaly detection system.

[0029] Figure 11 This is a flowchart illustrating the processing of the anomaly detection system.

[0030] Figure 12 This is a flowchart illustrating the processing of the anomaly detection system. Detailed Implementation

[0031] Figure 1 This is a schematic diagram illustrating the structure of an anomaly detection system 1000 based on a vehicle's surrounding monitoring sensors according to one embodiment. The anomaly detection system 1000 includes multiple vehicles 100 and a server 200. Each vehicle 100 has the functionality of an Advanced Driver-Assistance System (ADAS). Each vehicle 100 and the server 200 can communicate with each other via a communication network 300 composed of optical communication lines, etc., and a wireless base station 400 connected via a gateway (not shown) connected to the communication network 300. That is, the communication network 300 and the wireless base station 400 relay the communication between each vehicle 100 and the server 200.

[0032] Figure 2This is a schematic diagram showing the structure of vehicle 100. Vehicle 100 includes a positioning information receiver 110, a surrounding monitoring sensor 120, a wireless terminal 130, a display device 140, a speaker 150, an electronic control unit (ECU) 160, and an onboard camera 170. These components are communicatively connected via an in-vehicle network. Furthermore, in this embodiment, each vehicle 100 has the same structure, and the server 200 applies the same processing to each vehicle 100; therefore, unless otherwise specified, only one vehicle 100 will be described below.

[0033] The positioning information receiver 110 acquires positioning information indicating the current position and orientation of the vehicle 100. For example, the positioning information receiver 110 can be configured as a GPS (Global Positioning System) receiver. Each time positioning information is received, the positioning information receiver 110 outputs the acquired positioning information to the ECU 160 via the in-vehicle network.

[0034] The surrounding monitoring sensor 120 is a sensor used to monitor the area around the vehicle, including the front, rear, and sides, detecting objects around the vehicle (traffic lights, road signs, white lines on the road, buildings, etc., hereinafter referred to as objects). Specifically, the surrounding monitoring sensor 120 is a radar sensor. Electron waves (millimeter waves) emitted from the radar sensor towards the outside of the vehicle 100 are reflected by objects outside the vehicle 100 and received by the radar sensor. Then, by analyzing the received reflected waves, the position of the object relative to the vehicle 100 is obtained, i.e., the orientation and distance of the object relative to the vehicle 100. Alternatively, the surrounding monitoring sensor 120 can be a lidar (light detection and ranking) sensor.

[0035] The wireless terminal 130 includes, for example, an antenna and signal processing circuitry that performs various processes related to wireless communication, such as modulation and demodulation of wireless signals. The wireless terminal 130 then receives downlink wireless signals from the wireless base station 400 and transmits uplink wireless signals to the wireless base station 400. Specifically, the wireless terminal 130 extracts the signals transmitted from the server 200 to the vehicle 100 from the downlink wireless signals received from the wireless base station 400 and transmits them to the ECU 160. Furthermore, the wireless terminal 130 generates an uplink wireless signal that includes the signals received from the ECU 160 and transmitted to the server 200, and transmits this wireless signal.

[0036] The display device 140, for example, is a liquid crystal display (LCD) that displays a warning when the peripheral monitoring sensor 120 malfunctions. The speaker 150 issues an audible warning when the peripheral monitoring sensor 120 malfunctions.

[0037] ECU 160 includes a processor 162, a memory 164, and a communication interface 166. The processor 162 has one or more central processing units (CPUs) and their peripheral circuitry. The processor 162 may also include other arithmetic circuits such as logic units, numerical processing units, or graphics processing units. The memory 164 includes, for example, volatile and non-volatile semiconductor memories, storing data related to the processing involved in this embodiment. The communication interface 166 has interface circuitry for connecting the ECU 160 to an in-vehicle network.

[0038] The vehicle-mounted camera 170 has a two-dimensional detector composed of an array of photoelectric conversion elements such as CCD or C-MOS that are sensitive to visible light, and an imaging optical system that images the area to be photographed onto the two-dimensional detector. The vehicle-mounted camera 170 is positioned inside the vehicle, near the dashboard or windshield, etc., and captures images of the surroundings of the vehicle (e.g., in front of the vehicle), generating an image representing the environment around the vehicle. Each time an image is generated, the vehicle-mounted camera 170 outputs the generated image to the ECU 160 via an in-vehicle network.

[0039] Figure 3 This is a schematic diagram showing the structure of server 200. Server 200 has a control device 210 and a storage device 220.

[0040] The control device 210 includes a processor 212, a memory 214, and a communication interface 216. The processor 212 has one or more central processing units (CPUs) and their peripheral circuitry. The processor 212 may also include other arithmetic circuits such as logic units, numerical processing units, or graphics processing units. The memory 214 includes, for example, volatile semiconductor memory and non-volatile semiconductor memory. The communication interface 216 has interface circuitry for connecting the control device 210 to a network or communication network 300 within the server 200. The communication interface 216 is configured to communicate with the vehicle 100 via the communication network 300 and the wireless base station 400. That is, the communication interface 216 transmits information received from the vehicle 100 via the wireless base station 400 and the communication network 300 to the processor 212. Furthermore, the communication interface 216 transmits information received from the processor 212 to the vehicle 100 via the communication network 300 and the wireless base station 400.

[0041] Storage device 220 may include, for example, a hard disk drive or an optical recording medium and access means thereof. A high-resolution three-dimensional map (HD map) is stored in storage device 220. Furthermore, information about radar target clouds, described later, is stored in storage device 220. Additionally, storage device 220 may store a computer program for executing processes that run on processor 212.

[0042] If the surrounding monitoring sensor 120 of vehicle 100 detects an object, then as follows Figure 4 As shown, the vehicle ID (xxx), the location information (xR, yR, zR) indicating the location of object 500, and the received power (dBsm) of the surrounding monitoring sensor 120 when it detects object 500 are associated. Additionally, in Figure 4 In the example, a traffic signal is shown as object 500. The vehicle ID includes identification information for each vehicle 100 and information indicating the vehicle model and year. The vehicle ID, location information, and received power are sent from vehicle 100 to server 200 as object information.

[0043] Server 200 receives object information sent from vehicle 100. Server 200 compares the location information representing the location of the received object 500 with the locations of objects registered in a high-precision 3D map stored in storage device 220, and extracts objects matching the location information from the high-precision 3D map. Then, server 200 associates the object location information (X, Y, Z) obtained from the high-precision 3D map based on the vehicle model and year of vehicle 100 with the received power from vehicle 100 and stores it in storage device 220. In this embodiment, the information thus stored is also referred to as a radar target cloud.

[0044] For the advanced driver assistance system to function properly, the proper functioning of the surrounding monitoring sensor 120 is a prerequisite. In this embodiment, a radar target cloud is created that associates objects registered in a high-precision 3D map with the received power obtained by the surrounding monitoring sensor 120 of each vehicle 100. By matching these targets with objects detected by the vehicle 100 during driving, the received power corresponding to each target is sent to the vehicle 100, and any abnormalities in the surrounding monitoring sensor 120 and its surrounding components (e.g., components that transmit radar waves) are determined.

[0045] At this time, the types of peripheral monitoring sensors 120 and their mounting locations in vehicle 100 vary depending on the vehicle model and year. Therefore, if the received power is sent to vehicle 100 without considering the vehicle model and year, it is impossible to accurately determine the anomaly of the peripheral monitoring sensors 120. Therefore, the radar target cloud stores the location information and received power in a corresponding association with the vehicle model and year. Thus, by sending the received power corresponding to the vehicle model and year to vehicle 100, deviations in received power corresponding to the vehicle model and year can be suppressed to determine anomalies.

[0046] Figure 5 This is a diagram showing the structure of a radar target cloud. For example... Figure 5 As shown, the radar target cloud stores the names of objects (traffic lights, road signs, white lines on the road, buildings, etc.), the locations of objects, and the received power of surrounding monitoring sensors 120 when detecting objects, as well as the vehicle model (A, B, C, D, E, ...) and year of the vehicle 100. The names and locations of objects stored in the radar target cloud are obtained from a high-precision 3D map based on a comparison of the location information received from the vehicle 100 with the high-precision 3D map. Furthermore, the received power stored in the radar target cloud follows a power distribution obtained by analyzing the received power from the vehicle 100, based on the vehicle model and year.

[0047] Figure 6 This is a schematic diagram of the functional block diagram of the processor 162 of the ECU 160 of the vehicle 100. The processor 162 of the ECU 160 includes a position information acquisition unit 162a, a power receiving acquisition unit 162b, an object determination unit 162c, an object information transmission unit 162d, a reference information acquisition unit 162e, an axle deviation determination unit 162f, an axle deviation correction unit 162g, a dirt determination unit 162h, and a notification unit 162i. These units of the processor 162 are, for example, functional modules implemented by a computer program running on the processor 162. That is, the functional block diagram of the processor 162 consists of the processor 162 and the program (software) used to enable its functions. Furthermore, this program can be recorded in the memory 164 of the ECU 160 or in a recording medium connected externally. Alternatively, these units of the processor 162 can be dedicated arithmetic circuits provided in the processor 162.

[0048] When the surrounding monitoring sensor 120 mounted on the vehicle 100 detects an object, the location information acquisition unit 162a acquires first location information indicating the position of the object detected by the surrounding monitoring sensor 120. Specifically, the location information acquisition unit 162a acquires the first location information indicating the position of the object based on the position information of the vehicle 100 detected by the positioning information receiver 110 and the position information of the object relative to the vehicle 100 detected by the surrounding monitoring sensor 120.

[0049] The power acquisition unit 162b acquires the first received power received by the surrounding monitoring sensor 120 when it detects an object.

[0050] The object determination unit 162c determines the attributes of objects based on information received by the surrounding monitoring sensor 120 or images from the vehicle-mounted camera 170. Specifically, the object determination unit 162c determines the attributes (names) of objects such as traffic signals, road signs, white lines on the road, and buildings.

[0051] When determining the attributes of an object based on an image from the vehicle-mounted camera 170, the object determination unit 162c determines the attributes of the object, for example, by matching a template image with a template of the image generated by the vehicle-mounted camera 170, or by inputting the image generated by the vehicle-mounted camera 170 into a machine learning recognizer for object detection.

[0052] For example, the object determination unit 162c can use such a recognizer to output the probability of an object represented by that pixel in the input image, based on the possible types of objects represented by each pixel, and pre-learn a segmentation recognizer that represents the object with the highest probability. As such a recognizer, the object determination unit 162c can, for example, use a deep neural network (DNN) with a segmentation convolutional neural network type (CNN) architecture, such as a fully convolutional neural network (FCN).

[0053] The object information sending unit 162d sends object information to the server 200. Specifically, the object information sending unit 162d sends the first location information acquired by the location information acquisition unit 162a and the second received power information acquired by the received power acquisition unit 162b, along with the vehicle ID, to the server 200. Furthermore, the object information sending unit 162d may send the name of the object determined by the object determination unit 162c to the server 200.

[0054] When the server 200 determines that reference information corresponding to the object information exists in the radar target cloud and sends reference information from the server 200, the reference information acquisition unit 162e acquires the reference information sent from the server 200.

[0055] The axis deviation determination unit 162f determines the axis deviation of the peripheral monitoring sensor 120 based on the deviation of the second position information from the first position information, where the consistency between the first position information and the second position information is within an acceptable range. Specifically, the axis deviation determination unit 162f determines the axis deviation based on the deviation between the first position information and the second position information acquired by the position information acquisition unit 162a, based on the second position information determined by the server 200 to have a consistency with the first position information within an acceptable range. Here, the second position information is the position information stored in the radar target cloud on the server 200 side, and it is the position information whose consistency with the first position information is determined to be within an acceptable range. If the deviation of the first position information relative to the second position information is above a predetermined threshold, the axis deviation determination unit 162f determines that an axis deviation has occurred in the peripheral monitoring sensor 120.

[0056] Figure 7 This is a schematic diagram illustrating the method for determining axis deviation based on the axis deviation determination unit 162f. The axis deviation determination unit 162f compares the first position information (xR, yR, zR) acquired by the position information acquisition unit 162a with the second position information (X, Y, Z) included in the reference information acquired from the server 200. Then, if the first position information (xR, yR, zR) deviates from the second position information (X, Y, Z) acquired from the server 200 by a threshold or more, it is determined that an axis deviation has occurred in the peripheral monitoring sensor 120.

[0057] The axle deviation determination unit 162f determines that an axle deviation has occurred in the peripheral monitoring sensor 120 if the position of the peripheral monitoring sensor 120 relative to the vehicle 100 is set as the origin, and the deviation of the angle between the origin and the two straight lines connecting the positions (X, Y, Z) and (xR, yR, zR) is greater than or equal to a threshold. For example, the axle deviation determination unit 162f can determine whether the deviation of the angle between the two straight lines is greater than or equal to a threshold based on the relative angle of the two straight lines projected onto the XY plane (horizontal axle deviation θ) and the relative angle of the two straight lines projected onto the YZ plane (vertical axle deviation φ).

[0058] The shaft deviation correction unit 162g corrects the shaft deviation of the peripheral monitoring sensor 120 by making the position of the object detected by the peripheral monitoring sensor 120 the position represented by the second position information.

[0059] The dirt determination unit 162h determines the dirt on the surrounding monitoring sensor 120 based on the deviation of a second received power that is associated with the second location information, where the consistency between the first received power and the first location information is within an acceptable range. Here, the second received power is the received power associated with the second location information within the radar target cloud. The dirt determination unit 162h determines dirt based on the deviation between the second received power associated with the second location information within the radar target cloud and the first received power acquired by the received power acquisition unit 162b, based on the second location information determined by the server 200 to have a consistency with the first location information within an acceptable range, and the second location information. Furthermore, in radar sensors, the distance between the vehicle 100 and the target object has virtually no effect on the received power. Therefore, dirt can be determined based on the received power.

[0060] Figure 8 This is a schematic diagram illustrating the dirt determination method based on the dirt determination unit 162h. The dirt determination unit 162h compares the first received power received by the peripheral monitoring sensor 120 when it detects an object with the second received power obtained from the server 200. Then, if the dirt determination unit 162h determines that the peripheral monitoring sensor 120 is contaminated if the first received power received by the peripheral monitoring sensor 120 when it detects an object is below a predetermined threshold determined by the second received power obtained from the server 200, then the peripheral monitoring sensor 120 is contaminated. Figure 8 In the example shown, if the first received power received by the peripheral monitoring sensor 120 is below the threshold TH set based on the distribution 600 of the second received power obtained from the server 200, it is determined that dirt (NG) has occurred in the peripheral monitoring sensor 120.

[0061] Furthermore, the distribution 600 of the second received power corresponding to the vehicle model and year of vehicle 100 within the radar target cloud can be basically normally distributed. Therefore, Figure 8 The threshold TH can be a value specified by the standard deviation σ (or 2σ, 3σ).

[0062] The notification unit 162i issues a warning if it determines that axial misalignment has occurred in the peripheral monitoring sensor 120. Furthermore, the notification unit 162i issues a warning if it determines that dirt has accumulated in the peripheral monitoring sensor 120. Specifically, the notification unit 162i performs processing to display the warning on the display device 140 or to output a warning audibly from the speaker 150.

[0063] Figure 9This is a schematic diagram showing the functional block diagram of the processor 212 of the control device 210 included in the server 200. The processor 212 of the control device 210 includes an object information acquisition unit 212a, a comparison unit 212b, a storage processing unit 212c, a consistency determination unit 212d, and a reference information transmission unit 212e. These units of the processor 212 are, for example, functional modules implemented by a computer program running on the processor 212. That is, these units of the processor 212 consist of the processor 212 and the program (software) used to enable its function. Furthermore, this program can be recorded in the memory 214 included in the control device 210 or in a recording medium connected externally. Alternatively, these units of the processor 212 can be dedicated arithmetic circuits provided in the processor 212.

[0064] Examples of processing performed on server 200 include the creation of radar target clouds and the processing of axis deviation and dirt for determining the surrounding monitoring sensors 120.

[0065] First, the creation process of the radar target cloud will be explained. The object information acquisition unit 212a acquires object information transmitted from the vehicle 100. The object information includes the vehicle ID, the object's first location information, and the first received power.

[0066] The comparison unit 212b compares the first location information obtained from any vehicle 100 with the location information of objects registered in a high-precision 3D map, and extracts the objects whose location information corresponds to the first location information and are registered in the high-precision 3D map. At this time, the comparison unit 212b can extract objects from the high-precision 3D map whose location information differs from the first location information by a predetermined value or less. If the object information sent from the vehicle 100 includes the name of an object, the comparison unit 212b can further compare the name included in the object information with the name of the object registered in the high-precision 3D map, and extract objects from the object registered in the high-precision 3D map whose names are the same as the names received from the vehicle 100.

[0067] The storage processing unit 212c stores the object extracted by the comparison unit 212b as an arbitrary object, and stores the second location information and the second received power in the database. The storage processing unit 212c stores the location information of the object registered in the high-precision 3D map as the second location information, and stores the distribution of the first received power compared during extraction as the second received power. The storage processing unit 212c establishes a corresponding association between the second location information and the second received power and the vehicle model and year obtained from any vehicle 100. The storage processing unit 212c can also store the name of the object extracted by comparison in the database. Thus, by storing the location information obtained from the high-precision 3D map as the second location information in the radar target cloud, the location of the object within the radar target cloud is uniquely determined. Furthermore, the distribution of the first received power obtained under various driving conditions of each vehicle 100 is stored in the radar target cloud as the second received power. The second location information and second received power, which are stored in the radar target cloud and associated with vehicle model and year, are sent to vehicle 100 as reference information for determining the axis deviation and dirt of the surrounding monitoring sensor 120. The content of the radar target cloud is updated when the object information acquisition unit 212a acquires object information or periodically.

[0068] As described above, if the radar target cloud is stored in the storage device 220, the server 200 performs processing to determine the axial deviation and dirt of the surrounding monitoring sensor 120. The object information acquisition unit 212a acquires object information sent from the vehicle 100 in order to determine the axial deviation and dirt of the surrounding monitoring sensor 120.

[0069] The consistency determination unit 212d refers to a database that stores second location information representing the positions of multiple arbitrary objects and the second received power received by the surrounding monitoring sensor 120 of any vehicle when the arbitrary object is detected, and determines the consistency between the first location information and the second location information. Alternatively, a radar target cloud can be used as a database. Specifically, the consistency determination unit 212d compares the first location information obtained from the vehicle 100 with the second location information within the radar target cloud to determine whether the consistency between the first location information and the second location information is within an acceptable range.

[0070] As described above, the radar target cloud stores the second location information and the second received power in a corresponding association with the vehicle model and year. Therefore, the consistency determination unit 212d can determine the consistency between the first location information and the second location information based on the second location information, which is associated with the same vehicle model and year as the vehicle 100 equipped with the peripheral monitoring sensor 120, which is the object of anomaly determination.

[0071] When the consistency determination unit 212d determines that there is second location information in the radar target cloud that is within the acceptable range of consistency with the first location information of the object information, the reference information transmitting unit 212e transmits reference information including the second location information to the vehicle 100. Thus, reference information obtained from a vehicle of the same model and year as the vehicle 100 that transmitted object information for anomaly determination is transmitted to the vehicle 100 within the radar target cloud.

[0072] in addition, Figure 6 And then Figure 9 The structures of the processors shown are examples, and the components of one processor can be located in other processors. For example, at least a portion of the functional block of processor 212 of server 200 can be located in processor 162 of vehicle 100. Furthermore, the components of one processor can be repeatedly located in other processors. Additionally, the radar target cloud can be stored in memory 164 of vehicle 100.

[0073] Figures 10-12 This is a flowchart illustrating the process performed by the anomaly detection system 1000 according to a predetermined control cycle regarding the determination of shaft deviation and dirt on the peripheral monitoring sensor 120. First, the peripheral monitoring sensor 120 detects an object (step S10). Next, the location information acquisition unit 162a acquires first location information, and the power acquisition unit 162b acquires first received power (step S12). The first location information and the first received power, along with the vehicle ID, are sent to the server 200 and received by the server 200.

[0074] Next, the consistency determination unit 212d refers to the reference information within the radar target cloud (step S14) and determines whether there is reference information within the radar target cloud that is consistent with the object information (step S16). Specifically, the consistency determination unit 212d compares the first position information and the second position information within the radar target cloud to determine whether there is reference information whose consistency between the first position information and the second position information is within an acceptable range. If there is reference information within the radar target cloud that is consistent with the object information, the reference information transmission unit 212e transmits the reference information, including the second position information whose consistency with the first position information is within an acceptable range, to the vehicle 100 (step S18). After step S18, processing proceeds... Figure 11 Step S20 and Figure 12 Steps S20 and S30 are processed in parallel. On the other hand, if no reference information consistent with the object information exists within the radar target cloud in step S16, the process returns to step S10.

[0075] exist Figure 11In step S20, the axis deviation determination unit 162f determines the axis deviation of the peripheral monitoring sensor 120 based on the deviation between the first position information and the second position information included in the reference information obtained from the server 200 (step S20). Then, if the deviation of the first position information relative to the second position information is greater than or equal to a threshold (if "yes" in step S22), the notification unit 162i issues a warning of axis deviation abnormality (step S24). If the deviation is less than the threshold in step S22 (if "no" in step S22), the axis deviation correction unit 162g corrects the axis deviation of the peripheral monitoring sensor 120 (step S26). After step S26, the process returns to step S10. Alternatively, if the deviation is sufficiently small compared to the threshold, the process can return to step S10 without correcting the axis deviation in step S26.

[0076] Furthermore, in Figure 12 In step S30, the dirt determination unit 162h compares the first received power with the second received power contained in the reference information obtained from the server 200 to determine the dirt on the peripheral monitoring sensor 120. Then, if the first received power is below the threshold determined by the second received power obtained from the server 200 (if "yes" in step S32), the notification unit 162i issues a warning of dirt abnormality (step S34). On the other hand, if the first received power exceeds the threshold (if "no" in step S32), the process returns to step S10.

[0077] In addition, Figure 11 Step S24 or Figure 12 When an abnormal warning is issued in step S34, the functions of the advanced driver assistance system can be temporarily disabled.

[0078] As explained above, according to this embodiment, since the database that establishes an association between the location information and received power of the object is determined and the consistency between the location information and received power of the object detected on the vehicle 100 side, the shaft deviation determination and dirt determination of the surrounding monitoring sensor 120 can be performed uniformly.

[0079] Symbol Explanation

[0080] 160 - Electronic Control Unit (ECU), 162 - Processor, 162a - Position Information Acquisition Unit, 162b - Power Acquisition Unit, 162f - Shaft Deviation Determination Unit, 162g - Shaft Deviation Correction Unit, 162h - Dirt Determination Unit, 212b - Comparison Unit, 212c - Storage Processing Unit, 212d - Consistency Determination Unit, 1000 - Anomaly Determination System.

Claims

1. An anomaly detection system for a vehicle's surrounding monitoring sensors, characterized in that, have: The location information acquisition unit acquires first location information indicating the location of the object detected by the surrounding monitoring sensor when the surrounding monitoring sensor mounted on the vehicle detects an object. The power acquisition unit acquires the first received power when the surrounding monitoring sensor detects the object; The consistency determination unit refers to a database that stores second location information representing the positions of multiple arbitrary objects and second received power received by the surrounding monitoring sensors of any vehicle when the arbitrary object is detected, and determines the consistency between the first location information and the second location information. The axis deviation determination unit determines the axis deviation of the peripheral monitoring sensor based on the deviation between the first position information and the second position information, where the consistency between the first position information and the second position information is within an acceptable range. and The dirt determination unit determines the dirt on the surrounding monitoring sensor based on the deviation between the first received power and the second received power. The second received power is associated with the second location information whose consistency with the first location information is within an acceptable range.

2. The anomaly detection system for vehicle perimeter monitoring sensors according to claim 1, characterized in that, It also has: The axis deviation correction unit corrects the axis deviation of the peripheral monitoring sensor in such a way that the position of the object detected by the peripheral monitoring sensor becomes the position represented by the second position information.

3. The anomaly detection system for vehicle perimeter monitoring sensors according to claim 1 or 2, characterized in that, The database stores the second location information and the second received power in association with vehicle model and year. The consistency determination unit establishes a corresponding association with the second location information based on the same model and year as the vehicle equipped with the surrounding monitoring sensor that is the object of anomaly determination, and determines the consistency between the first location information and the second location information.

4. The anomaly detection system for vehicle perimeter monitoring sensors according to claim 1 or 2, characterized in that, It also has: The comparison unit compares the first location information obtained from the arbitrary vehicle with the location information of the object registered in the high-precision three-dimensional map, and extracts the object whose location information corresponds to the first location information and is registered in the high-precision three-dimensional map. and The storage processing unit treats the extracted object as the arbitrary object, stores the second location information and the second received power in the database, stores the location information of the object registered in the high-precision 3D map as the second location information, and stores the distribution of the first received power compared during extraction as the second received power.

5. The anomaly detection system for vehicle perimeter monitoring sensors according to claim 4, characterized in that, The storage processing unit stores the second location information and the second received power in association with the vehicle model and year obtained from the arbitrary vehicle.

Images