Intelligent driving evaluation method, device, terminal and system

Abstract

The invention discloses an intelligent driving evaluation method, device, terminal and system, and the method comprises the steps: obtaining the first data of each vehicle sensor, the second data of each standard sensor, and the legal value range of each sensor data in real time through an intelligent driving evaluation terminal in a testing vehicle driving process, and obtaining the driving environment information from a cloud control platform; if the two pieces of data are within the legal value range, the difference between the two pieces of data is within a preset difference range, or if only the first data is within the legal value range, marking the first data as normal data; if the first data and the second data are within the legal value range, the difference between the first data and the second data is not within the preset difference range, or if only the second data is within the legal value range, marking the first data as abnormal data; and if the two pieces of data are not in the legal value range, marking the first data as normal data when judging that the first data is driving data generated by a driving behavior adopted for avoiding the collision risk in combination with the driving environment information, and otherwise, marking the first data as abnormal data.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and more specifically, to an intelligent driving evaluation method, device, terminal and system. Background Art

[0002] Intelligent connected vehicles are gradually evolving from L1 / L2 assisted driving to L3 / L4 high-level intelligent driving, but there is a large gap between the current level of intelligent driving and human driving experts in terms of situational understanding, edge case processing, driving safety, and traffic efficiency. In order to conduct real-time evaluation of the driving behavior of intelligent driving vehicles, the current main method is: an expert driver drives the test vehicle, observes in real time and manually records whether the driving behavior of the intelligent driving vehicle is legal and reasonable, and finally summarizes the evaluation results of the expert driver and outputs an evaluation report. However, this evaluation method is labor-intensive and not suitable for summary, resulting in low efficiency of the entire evaluation. Summary of the invention

[0003] The present application provides an intelligent driving evaluation method, device, terminal and system, which can realize automatic evaluation of the real-time driving behavior of intelligent driving vehicles, thereby greatly improving the efficiency of the entire evaluation.

[0004] The specific technical solutions are as follows:

[0005] In a first aspect, an embodiment of the present application provides an intelligent driving evaluation method, which is applied to an intelligent driving evaluation terminal, wherein the intelligent driving evaluation terminal includes a plurality of standard sensors and a mobile communication module, and the intelligent driving evaluation terminal is installed on a test vehicle, and the method includes:

[0006] During the driving process of the test vehicle, first state data generated by each vehicle sensor in the test vehicle is obtained in real time through the controller area network CAN protocol, second state data generated by each standard sensor is obtained in real time, a legal value range of each sensor data on the driving road is obtained, and driving environment information around the current driving road is obtained from the cloud control platform in real time through the mobile communication network of the mobile communication module;

[0007] If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is within a preset difference range, marking the first state data as normal data;

[0008] If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is not within the preset difference range, marking the first state data as abnormal data;

[0009] If the first state data is not within the legal value range, and the second state data is within the legal value range, marking the first state data as abnormal data;

[0010] If the first state data is within the legal value range and the second state data is not within the legal value range, marking the first state data as normal data and marking the second state data as abnormal data;

[0011] If both the first state data and the second state data are not within the legal value range, combined with the driving environment information, determine whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, mark the first state data as normal data; when it is determined that the first state data is not the target type of data, mark the first state data as abnormal data, wherein the target type of data includes driving data generated by driving behaviors taken to avoid collision risks.

[0012] In a possible implementation, the method further includes:

[0013] The marked first state data and the marked second state data are sent to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormality detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform.

[0014] In a possible implementation, when there is at least one of a traffic light, a pedestrian, a motor vehicle, and a non-motor vehicle around the test vehicle, the driving environment information includes at least one of the status information of the traffic light, the status information of the pedestrian, the status information of the motor vehicle, and the status information of the non-motor vehicle.

[0015] In a second aspect, an embodiment of the present application provides an intelligent driving evaluation device, which is applied to an intelligent driving evaluation terminal, the intelligent driving evaluation terminal includes a plurality of standard sensors and a mobile communication module, and the intelligent driving evaluation terminal is installed on a test vehicle, and the device includes:

[0016] an acquisition unit, used to acquire, in real time during the driving process of the test vehicle, first state data generated by each vehicle sensor in the test vehicle through a controller area network (CAN) protocol, acquire second state data generated by each standard sensor in real time, acquire a legal value range of each sensor data on the driving road, and acquire driving environment information around the current driving road from a cloud control platform in real time through the mobile communication network of the mobile communication module;

[0017] a judging and marking unit, configured to mark the first state data as normal data when both the first state data and the second state data are within the corresponding legal value range and the difference between the first state data and the second state data is within a preset difference range; mark the first state data as abnormal data when both the first state data and the second state data are within the corresponding legal value range and the difference between the first state data and the second state data is not within the preset difference range; mark the first state data as abnormal data when the first state data is not within the legal value range and the second state data is within the legal value range; mark the first state data as abnormal data when the first state data is within the legal value range When the first state data is within the legal value range and the second state data is not within the legal value range, the first state data is marked as normal data and the second state data is marked as abnormal data; when both the first state data and the second state data are not within the legal value range, combined with the driving environment information, it is determined whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, the first state data is marked as normal data, and when it is determined that the first state data is not the target type of data, the first state data is marked as abnormal data, wherein the target type of data includes driving data generated by driving behavior taken to avoid collision risks.

[0018] In a possible implementation, the device further includes:

[0019] A sending unit is used to send the marked first state data and the marked second state data to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormality detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform.

[0020] In a possible implementation, when there is at least one of a traffic light, a pedestrian, a motor vehicle, and a non-motor vehicle around the test vehicle, the driving environment information includes at least one of the status information of the traffic light, the status information of the pedestrian, the status information of the motor vehicle, and the status information of the non-motor vehicle.

[0021] In a third aspect, an embodiment of the present application provides an intelligent driving evaluation terminal, which includes a plurality of standard sensors, a mobile communication module, a processor, and a memory, and is installed on a test vehicle;

[0022] The standard sensor is used to measure the state of the test vehicle in real time and generate second state data;

[0023] The mobile communication module is used for wireless communication with the cloud control platform;

[0024] The memory is used to store one or more programs and a legal value range of each sensor data on the traveled road;

[0025] The processor is used to execute one or more programs to implement the method described in any possible implementation manner of the first aspect.

[0026] In a fourth aspect, an embodiment of the present application provides an intelligent driving evaluation system, the system comprising a test vehicle, an intelligent driving evaluation terminal and a cloud control platform, the intelligent driving evaluation terminal comprising a plurality of standard sensors and a mobile communication module, and the intelligent driving evaluation terminal being installed on the test vehicle;

[0027] The test vehicle is used for each vehicle sensor to generate first state data during driving based on the automatic driving system;

[0028] The cloud control platform is used to send the driving environment information around the current driving road of the test vehicle to the intelligent driving evaluation terminal;

[0029] The intelligent driving evaluation terminal is used to:

[0030] During the driving process of the test vehicle, the first state data is obtained in real time through the controller area network CAN protocol, the second state data generated by each standard sensor is obtained in real time, the legal value range of each sensor data on the driving road is obtained, and the driving environment information is obtained from the cloud control platform in real time through the mobile communication network of the mobile communication module;

[0031] If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is within a preset difference range, marking the first state data as normal data;

[0032] If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is not within the preset difference range, marking the first state data as abnormal data;

[0033] If the first state data is not within the legal value range, and the second state data is within the legal value range, marking the first state data as abnormal data;

[0034] If the first state data is within the legal value range and the second state data is not within the legal value range, marking the first state data as normal data and marking the second state data as abnormal data;

[0035] If both the first state data and the second state data are not within the legal value range, combined with the driving environment information, determine whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, mark the first state data as normal data; when it is determined that the first state data is not the target type of data, mark the first state data as abnormal data, wherein the target type of data includes driving data generated by driving behaviors taken to avoid collision risks.

[0036] In a possible implementation, the system further includes an intelligent driving evaluation platform;

[0037] The intelligent driving evaluation platform is used to obtain the marked first state data and the marked second state data sent by the intelligent driving evaluation terminal through the mobile communication module, so as to replay the marked first state data and the marked second state data to verify the abnormality detection result of the first state data.

[0038] In a possible implementation, the intelligent driving evaluation platform is further used to obtain the operational design domain (ODD) data of the test vehicle, and decompose the ODD data into a plurality of ODD elements;

[0039] Matching the multiple ODD elements with the test scenario library to obtain at least one successfully matched target test scenario, wherein the evaluation dimension corresponding to the at least one target test scenario is a safety and comfort dimension;

[0040] Generating a test route corresponding to each of the target test scenarios so that the test vehicle travels along each of the test routes;

[0041] Acquiring driving data generated by the test vehicle during driving along each of the test routes;

[0042] For each evaluation indicator in each target test scenario in the safety and comfort dimension, determine the indicator score of the evaluation indicator according to the driving data and the scoring rule of the evaluation indicator;

[0043] For each experience indicator in each preset experience scenario, determining an indicator score of the experience indicator according to the driving data and a scoring rule of the experience indicator;

[0044] For the communication efficiency index in the traffic efficiency dimension, determining an index score of the communication efficiency index according to the driving data and a scoring rule of the communication efficiency index;

[0045] The total score of the test vehicle is determined according to the indicator score of each of the evaluation indicators, the indicator weight of each of the evaluation indicators, the scenario weight of each of the target test scenarios, the number of scenarios of the target test scenarios, the dimension weight of the safety and comfort dimension, the indicator score of each of the experience indicators, the indicator weight of each of the experience indicators, the number of scenarios of the preset experience scenarios, the dimension weight of the experience dimension, the indicator score of the communication efficiency indicator and the dimension weight of the traffic efficiency dimension.

[0046] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed on a computer or a processor, the computer or the processor executes the method described in any possible implementation of the first aspect.

[0047] The intelligent driving evaluation method, device, terminal and system provided in the embodiments of the present application can pre-install an intelligent driving evaluation terminal including a plurality of standard sensors and a mobile communication module on the test vehicle. In this way, during the driving of the test vehicle, the vehicle sensors on the test vehicle and the standard sensors in the intelligent driving evaluation terminal will measure the real-time status of the test vehicle, and can also obtain the legal value range of each sensor data on the driving road, and obtain the driving environment information around the current driving road from the cloud control platform in real time. Therefore, for the same type of sensor, when at least one of the first state data generated by the vehicle sensor and the second state data generated by the standard sensor is within the corresponding legal value range, , the first state data can be marked as normal or abnormal based on the difference between the two state data and which state data is within the legal value range. When both state data are not within the legal value range, the first state data can be further judged based on the driving environment information to determine whether the first state data is driving data generated by the driving behavior taken to avoid collision risk. When the first state data is driving data generated by the driving behavior taken to avoid collision risk, it indicates that the first state data is a reasonable behavior made in a special scenario, and the first state data can be marked as normal data. When the first state data is not driving data generated by the driving behavior taken to avoid collision risk, the first state data can be marked as abnormal data. It can be seen from this that the intelligent driving evaluation terminal provided in the embodiment of the present application can not only automatically evaluate the driving behavior of the test vehicle in real time, but also reasonably evaluate the special driving behavior in special scenarios, thereby ensuring the fairness and rationality of the evaluation on the basis of improving the evaluation efficiency.

[0048] In addition, the intelligent driving evaluation terminal can also send the marked first state data and the marked second state data to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormal detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform. When the abnormal detection result of a certain first state data is incorrect, the tester can also correct the abnormal detection result. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art description are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work.

[0050] Figure 1 A flowchart of an intelligent driving evaluation method provided in an embodiment of the present application;

[0051] Figure 2 A schematic diagram of an intelligent driving evaluation system architecture provided in an embodiment of the present application;

[0052] Figure 3 A schematic diagram of another intelligent driving evaluation system architecture provided in an embodiment of the present application;

[0053] Figure 4 A block diagram of the composition of an intelligent driving evaluation device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0055] It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The terms "including" and "having" in the embodiments of the present application and the accompanying drawings and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0056] Figure 1 The present invention is a flow chart of an intelligent driving evaluation method, which can be applied to an intelligent driving evaluation terminal, which includes a plurality of standard sensors and a mobile communication module, and is installed on a test vehicle, for example, on the top of the outer side of the vehicle, and communicates with the vehicle system through a CAN (Controller Area Network) bus. The method includes:

[0057] S110: During the driving process of the test vehicle, the first state data generated by each vehicle sensor in the test vehicle is obtained in real time through the controller area network CAN protocol, the second state data generated by each standard sensor is obtained in real time, the legal value range of each sensor data on the driving road is obtained, and the driving environment information around the current driving road is obtained from the cloud control platform in real time through the mobile communication network of the mobile communication module.

[0058] Among them, the standard sensor is a reliable sensor that meets the accuracy requirements. The sensors on the test vehicle include image sensors, laser radars, speed sensors, angular velocity sensors, acceleration sensors, etc. Similarly, the standard sensors can also include these sensors, and the types of standard sensors and vehicle sensors on the test vehicle correspond one to one. The first state data are the data generated by these sensors, including image data, laser point cloud data, speed, acceleration, angular velocity, heading angle, etc.

[0059] The legal value range of each sensor data on the road on which the vehicle is traveling refers to the state range of the vehicle on the road in compliance with traffic laws. For example, if there is a speed limit on each road, the speed measured by the speed sensor has a corresponding legal speed range. For another example, when the vehicle is driving straight, it is not allowed to cross the lane line. At this time, the positional relationship between the vehicle and the lane line collected by the image sensor, or the distance between the center point of the vehicle's rear axle and the lane line measured by the distance sensor, can have a corresponding legal value range.

[0060] The cloud control platform can combine roadside equipment to obtain the motion status of each element on each road in real time, so the intelligent driving evaluation terminal can obtain the driving environment information around the current driving road from the cloud control platform in real time through the mobile communication network of the mobile communication module. When there is at least one of traffic lights, pedestrians, motor vehicles and non-motor vehicles around the test vehicle, the driving environment information includes at least one of the status information of traffic lights, pedestrians, motor vehicles and non-motor vehicles. In addition, there may be static obstacles in the test vehicle, so the driving environment information can also include the location information of static obstacles.

[0061] It should be noted that, in the abstract of the specification, the first state data and the second state data are referred to as first data and second data, respectively.

[0062] S120: If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is within a preset difference range, the first state data is marked as normal data.

[0063] When the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is within the preset difference range, it means that the first state data and the second state data are both legal, and because the difference between the two is small, it can be considered that the first state data meets the accuracy requirements and is marked as normal data, or no marking is performed and it is normal data by default.

[0064] S130: If the first state data and the second state data are both within the corresponding legal value ranges, and the difference between the first state data and the second state data is not within a preset difference range, the first state data is marked as abnormal data.

[0065] When the first state data and the second state data are both within the corresponding legal value range, but the difference between the first state data and the second state data is not within the preset difference range, it means that although the first state data is legal, it is unreasonable and does not meet the accuracy requirements, and it can be marked as abnormal data.

[0066] S140: If the first state data is not within the legal value range, and the second state data is within the legal value range, the first state data is marked as abnormal data.

[0067] When the first state data is not within the legal value range, but the second state data is within the legal value range, it means that the first state data is illegal and unreasonable, and the decision made by the test vehicle based on the first state data will also be unreasonable, so the first state data can be marked as abnormal data.

[0068] S150: If the first state data is within the legal value range and the second state data is not within the legal value range, the first state data is marked as normal data and the second state data is marked as abnormal data.

[0069] When the first state data is within the legal value range, but the second state data is not within the legal value range, it means that there may be certain abnormalities in the second state data, and it is necessary to feed it back to the intelligent driving evaluation platform or the cloud control platform to perform accuracy detection and / or repair and improvement work on the standard sensor corresponding to the second state data.

[0070] S160: If both the first state data and the second state data are outside the legal value range, determine whether the first state data belongs to the target type of data in combination with the driving environment information, and when it is determined that the first state data is the target type of data, mark the first state data as normal data; when it is determined that the first state data is not the target type of data, mark the first state data as abnormal data, wherein the target type of data includes driving data generated by driving behaviors taken to avoid collision risks.

[0071] When both the first state data and the second state data are outside the legal value range, if the first state data is directly marked as abnormal data, misjudgment may occur. For example, in actual applications, whether it is a manually driven vehicle or an intelligently driven vehicle, emergency avoidance operations may occur due to unexpected driving environments, but the vehicle state caused by the emergency avoidance operation may not meet the legal value range. This ensures the safety of both or even multiple parties, so it is reasonable.

[0072] Therefore, when both the first state data and the second state data are not within the legal value range, the driving environment information can also be combined to determine whether the first state data belongs to the target type of data; when it is determined that the first state data is driving data generated by the driving behavior taken to avoid the risk of collision, the first state data is marked as normal data; and when it is determined that the first state data is not driving data generated by the driving behavior taken to avoid the risk of collision, the first state data is marked as abnormal data. In other words, in special scenarios, parameters such as speed, acceleration, deceleration, line-crossing distance, line-crossing time, body offset, braking distance, and sidewalk parking distance may not meet their corresponding legal value ranges.

[0073] For example, if the test vehicle should have gone straight, but could not avoid the obstacle by braking due to an emergency, but could solve the problem by changing lanes, the abnormalities such as acceleration, deceleration, and crossing the line will be treated as special processing, that is, as the correct data generated under normal behavior. For another example, if there is a foreign object on the road, the abnormalities such as changing lanes to avoid obstacles, acceleration, deceleration, and body deviation will be treated as special processing.

[0074] In order to further determine the accuracy of the first state data, when it is determined that the first state data and the second state data are both driving data generated by driving behaviors taken to avoid collision risks, it is necessary to judge again whether the difference between the two is within the preset difference range. If it is within the preset difference range, the first state data is marked as normal data. If it is not within the preset difference range, the first state data is still marked as abnormal data.

[0075] The intelligent driving evaluation method provided in the embodiment of the present application can pre-install an intelligent driving evaluation terminal including a plurality of standard sensors and a mobile communication module on the test vehicle. In this way, during the driving of the test vehicle, the vehicle sensors on the test vehicle and the standard sensors in the intelligent driving evaluation terminal can measure the real-time status of the test vehicle, and can also obtain the legal value range of each sensor data on the driving road, and obtain the driving environment information around the current driving road from the cloud control platform in real time. Therefore, for the same type of sensor, when at least one of the first state data generated by the vehicle sensor and the second state data generated by the standard sensor is within the corresponding legal value range, it can be combined The difference between the two state data, and which state data is within the legal value range, is used to mark the first state data as normal or abnormal. When both state data are not within the legal value range, the driving environment information can be combined to further determine whether the first state data is driving data generated by the driving behavior taken to avoid collision risks. When the first state data is driving data generated by the driving behavior taken to avoid collision risks, it means that the first state data is a reasonable behavior made in a special scenario, and the first state data can be marked as normal data. When the first state data is not driving data generated by the driving behavior taken to avoid collision risks, the first state data can be marked as abnormal data. It can be seen from this that the intelligent driving evaluation terminal provided in the embodiment of the present application can not only automatically evaluate the driving behavior of the test vehicle in real time, but also reasonably evaluate the special driving behavior in special scenarios. On the basis of improving the evaluation efficiency, the fairness and rationality of the evaluation are guaranteed.

[0076] In a possible implementation, in order to further verify whether the evaluation of the first state data by the intelligent driving evaluation terminal is accurate, the intelligent driving evaluation terminal can also send the marked first state data and the marked second state data to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormal detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform. When verifying that the abnormal detection result of a certain first state data is incorrect, the tester can also correct the abnormal detection result. The abnormal detection result includes the marking of normal data or the marking of abnormal data.

[0077] In addition, the intelligent driving evaluation platform can also output and display the statistical results of the abnormal detection results of the first state data. For example, it can count the number of abnormalities and the duration of abnormalities for different driving scenarios.

[0078] Based on the above method embodiment, another embodiment of the present application provides an intelligent driving evaluation system, such as Figure 2As shown, the system includes a test vehicle 210, an intelligent driving evaluation terminal 220 and a cloud control platform 230, wherein the intelligent driving evaluation terminal 220 includes a plurality of standard sensors 221 and a mobile communication module 222, and the intelligent driving evaluation terminal 220 is installed on the test vehicle 210;

[0079] The test vehicle 210 is used for each vehicle sensor 211 to generate first state data during driving based on the automatic driving system;

[0080] The cloud control platform 230 is used to send the driving environment information around the current driving road of the test vehicle 210 to the intelligent driving evaluation terminal 220;

[0081] The intelligent driving evaluation terminal 220 is used to:

[0082] During the driving process of the test vehicle 210, the first state data is obtained in real time through the controller area network CAN protocol, the second state data generated by each standard sensor 221 is obtained in real time, the legal value range of each sensor data on the driving road is obtained, and the driving environment information is obtained from the cloud control platform 230 in real time through the mobile communication network of the mobile communication module 222;

[0083] If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is within a preset difference range, marking the first state data as normal data;

[0084] If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is not within the preset difference range, marking the first state data as abnormal data;

[0085] If the first state data is not within the legal value range, and the second state data is within the legal value range, marking the first state data as abnormal data;

[0086] If the first state data is within the legal value range and the second state data is not within the legal value range, marking the first state data as normal data and marking the second state data as abnormal data;

[0087] If both the first state data and the second state data are not within the legal value range, combined with the driving environment information, determine whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, mark the first state data as normal data; when it is determined that the first state data is not the target type of data, mark the first state data as abnormal data, wherein the target type of data includes driving data generated by driving behaviors taken to avoid collision risks.

[0088] The intelligent driving evaluation system provided in the embodiment of the present application can pre-install an intelligent driving evaluation terminal including a plurality of standard sensors and a mobile communication module on the test vehicle. In this way, during the driving of the test vehicle, the vehicle sensors on the test vehicle and the standard sensors in the intelligent driving evaluation terminal can measure the real-time status of the test vehicle, and can also obtain the legal value range of each sensor data on the driving road, and obtain the driving environment information around the current driving road from the cloud control platform in real time. Therefore, for the same type of sensor, when at least one of the first state data generated by the vehicle sensor and the second state data generated by the standard sensor is within the corresponding legal value range, it can be combined The difference between the two state data, and which state data is within the legal value range, is used to mark the first state data as normal or abnormal. When both state data are not within the legal value range, the driving environment information can be combined to further determine whether the first state data is driving data generated by the driving behavior taken to avoid collision risks. When the first state data is driving data generated by the driving behavior taken to avoid collision risks, it means that the first state data is a reasonable behavior made in a special scenario, and the first state data can be marked as normal data. When the first state data is not driving data generated by the driving behavior taken to avoid collision risks, the first state data can be marked as abnormal data. It can be seen from this that the intelligent driving evaluation terminal provided in the embodiment of the present application can not only automatically evaluate the driving behavior of the test vehicle in real time, but also reasonably evaluate the special driving behavior in special scenarios. On the basis of improving the evaluation efficiency, the fairness and rationality of the evaluation are guaranteed.

[0089] In a possible implementation, Figure 3 As shown, the system further includes an intelligent driving evaluation platform 240;

[0090] The intelligent driving evaluation platform 240 is used to obtain the marked first state data and the marked second state data sent by the intelligent driving evaluation terminal 220 through the mobile communication module 222, so as to replay the marked first state data and the marked second state data to verify the abnormality detection result of the first state data.

[0091] In a possible implementation, the intelligent driving evaluation platform 240 is further used to obtain the operational design domain (ODD) data of the test vehicle and decompose the ODD data into a plurality of ODD elements;

[0092] Matching the multiple ODD elements with the test scenario library to obtain at least one successfully matched target test scenario, wherein the evaluation dimension corresponding to the at least one target test scenario is a safety and comfort dimension;

[0093] Generating a test route corresponding to each of the target test scenarios so that the test vehicle travels along each of the test routes;

[0094] Acquiring driving data generated by the test vehicle during driving along each of the test routes;

[0095] For each evaluation indicator in each target test scenario in the safety and comfort dimension, determine the indicator score of the evaluation indicator according to the driving data and the scoring rule of the evaluation indicator;

[0096] For each experience indicator in each preset experience scenario, determining an indicator score of the experience indicator according to the driving data and a scoring rule of the experience indicator;

[0097] For the communication efficiency index in the traffic efficiency dimension, determining an index score of the communication efficiency index according to the driving data and a scoring rule of the communication efficiency index;

[0098] The total score of the test vehicle is determined according to the indicator score of each of the evaluation indicators, the indicator weight of each of the evaluation indicators, the scenario weight of each of the target test scenarios, the number of scenarios of the target test scenarios, the dimension weight of the safety and comfort dimension, the indicator score of each of the experience indicators, the indicator weight of each of the experience indicators, the number of scenarios of the preset experience scenarios, the dimension weight of the experience dimension, the indicator score of the communication efficiency indicator and the dimension weight of the traffic efficiency dimension.

[0099] In the embodiment of the present application, the vehicle can be comprehensively evaluated and tested in three dimensions: safety and comfort, experience, and communication efficiency. The following introduces the relevant information of these three dimensions respectively:

[0100] 1. Safety and comfort dimensions

[0101] ODD (Operational Design Domain) refers to the specific conditions under which the intelligent driving system is set to function. In simple terms, it is the "safety boundary" of the intelligent driving vehicle, defining the conditions under which the intelligent driving system can work normally. Outside of these conditions, intelligent driving cannot guarantee safety.

[0102] After obtaining the ODD data of the test vehicle, the ODD data can be decomposed into multiple ODD elements according to certain rules. The element types of ODD elements include at least one of roads, traffic facilities, traffic rules and control, weather environment, traffic participants and road intelligence. Each type of ODD element can be shown in Table 1.

[0103] Table 1 List of ODD elements

[0104]

[0105]

[0106] Any ODD element, or any combination of multiple ODD elements may correspond to different test scenarios, and the mapping relationship between the ODD element or the combination of ODD elements and the test scenario can be pre-set, and the mapping relationship can be stored in the test scenario library, so that the electronic device can determine which test scenarios the multiple ODD elements obtained can match by looking up the mapping relationship, and determine the successfully matched test scenario as the target test scenario. In actual applications, when a vehicle with at least one ODD element can run normally in one or some test scenarios, the at least one ODD element can be set to match the test scenario, and when a vehicle with at least one ODD element cannot run normally in one or some test scenarios, the at least one ODD element can be set to not match the test scenario. Among them, the evaluation dimension corresponding to at least one target test scenario is the safety and comfort dimension.

[0107] The test scene library includes static scenes and / or dynamic scenes.

[0108] When the static scene includes driving on a road section and specifically driving on a straight road, the evaluation index includes at least one of lane line recognition and response, traffic sign recognition and response, and longitudinal acceleration; and / or,

[0109] When the static scene includes driving on a road section and specifically driving on a curve, the evaluation indicators include at least one of lane line recognition and response, traffic sign recognition and response, longitudinal acceleration, and lateral acceleration; and / or,

[0110] When the static scene includes traffic at an intersection, and specifically traffic at an intersection with traffic lights, the evaluation indicators include at least one of traffic light recognition and response, intersection guide line recognition and response, and longitudinal acceleration; and / or,

[0111] When the static scene includes crossing an intersection and specifically crossing an intersection without traffic lights, the evaluation indicators include at least one of intersection guide line recognition and response, intersection conflict crossing and response, and longitudinal acceleration; and / or,

[0112] When the static scene includes tunnel driving, the evaluation indicators include at least one of driving safety and response, driving safety and response, lane recognition and response, following distance, and longitudinal acceleration; and / or,

[0113] When the static scene includes a roundabout, and specifically a complex roundabout with traffic lights, the evaluation indicators include at least one of traffic light recognition and response, intersection guide line recognition and response, safe following distance, longitudinal acceleration, and lateral acceleration; and / or,

[0114] When the static scene includes a roundabout and specifically a complex roundabout without traffic lights, the evaluation indicators include at least one of intersection guide line recognition and response, safe following distance, longitudinal acceleration, and lateral acceleration; and / or,

[0115] When the static scene includes the passage of the auxiliary road ramp entrance and exit, the evaluation indicators include at least one of the intersection guide line recognition and response, following distance, longitudinal acceleration, and lateral acceleration; and / or,

[0116] When the static scene includes bridge traffic, the evaluation indicators include at least one of lane line recognition and response, traffic sign and marking recognition and response, and longitudinal acceleration; and / or,

[0117] When the static scene includes mountain road driving, the evaluation indicators include at least one of curve passing, lane line recognition and response ramp passing, longitudinal acceleration, and lateral acceleration; and / or,

[0118] When the static scene includes driving on a narrow road, the evaluation indicators include at least one of lane line recognition and response, traffic sign and marking recognition and response, travel speed, and longitudinal acceleration; and / or,

[0119] When the static scene includes obstacle recognition and response, the evaluation index includes at least one of obstacle recognition distance, obstacle avoidance distance, and longitudinal acceleration; and / or,

[0120] When the static scenario includes a U-turn, the evaluation indicators include U-turn execution efficiency and / or lateral acceleration; and / or,

[0121] When the dynamic scene includes stable following driving and specifically following a vehicle on a straight road, the evaluation indicators include at least one of lane line recognition and response, target vehicle recognition, following distance, following speed, and longitudinal acceleration; and / or,

[0122] When the dynamic scene includes stable following driving and specifically following a car on a curve, the evaluation indicators include at least one of lane line recognition and response, lane line recognition and response, following distance, following speed, longitudinal acceleration, and lateral acceleration; and / or,

[0123] When the dynamic scene includes pedestrian and non-motor vehicle avoidance, the evaluation index includes at least one of the shortest distance to the target, starting efficiency, and longitudinal acceleration; and / or,

[0124] When the dynamic scene includes merging, the evaluation index includes at least one of safe headway, turn signal control, and longitudinal acceleration; and / or,

[0125] When the dynamic scene includes overtaking, the evaluation index includes at least one of the following: lane change time, safe distance to the preceding vehicle, safe distance to the adjacent vehicle, overtaking completion time, and longitudinal acceleration; and / or,

[0126] When the dynamic scene includes oncoming vehicles, the evaluation indicators include vehicle body offset and / or lateral distance; and / or,

[0127] When the dynamic scenario includes borrowing from the oncoming lane, the evaluation indicators include lateral distance and / or longitudinal distance; and / or,

[0128] When the dynamic scenario includes congested traffic, the evaluation indicators include at least one of the following distance, starting time, and longitudinal acceleration.

[0129] Different evaluation indicators have different scoring rules, which generally include two categories of rules. When the evaluation indicators include the operating state parameters of the test vehicle itself or the quantitative parameters between the test vehicle and the surrounding environment, such as longitudinal acceleration, lateral acceleration, following distance, following speed, and safe headway, the scoring rules include different scores corresponding to different value ranges. The electronic device can match the indicator value of the evaluation indicator calculated based on the driving data of the test vehicle with the different value ranges in the scoring rules, and determine the indicator score of the evaluation indicator according to the successfully matched value range. The mapping relationship between different value ranges and scores can be determined based on actual experience, and the actual experience can be determined by the historical driving data of manual driving or other intelligent driving vehicles. When the evaluation indicators include lane line recognition and response, traffic sign recognition and response, traffic light recognition and response, intersection guide line recognition and response, turn signal control, etc., which involve perception effects and control effects, different effects (perception effects, control effects, etc.) can be set in advance for different evaluation indicators based on actual experience, and different scores can be set for different effects. The electronic device can match the indicator effect (perception effect, control effect, etc.) of the evaluation indicator calculated based on the driving data of the test vehicle with the different effects in the scoring rules, and determine the score corresponding to the successfully matched effect as the indicator score of the evaluation indicator.

[0130] 2. Experience Dimension

[0131] This test project evaluates the overall satisfaction of the intelligent driving system based on the experience of experienced driving reviewers during the driving and riding process of the entire test route. The evaluation of experience indicators takes into account factors such as the practicality of the function, driving comfort, sense of security, and human-computer interaction experience.

[0132] The preset experience scenario may specifically include at least one of interactive experience, driving burden, and functional control experience.

[0133] When the preset experience scenario includes interactive experience, the experience indicators include functional usability and / or information interaction; and / or,

[0134] When the preset experience scenario includes driving burden, the experience indicator includes driving burden; and / or,

[0135] When the preset experience scenario includes functional control experience, the experience indicators include at least one of vehicle longitudinal control, centering control capability, automatic lane change, large vehicle avoidance, on- and off-ramps, and tunnel passing.

[0136] The scoring rules for experience indicators are similar to those for evaluation indicators and will not be elaborated here.

[0137] (III) Communication efficiency dimension

[0138] The scoring rule of the communication efficiency index may correspond different scores to different communication efficiency ranges, by matching the communication efficiency calculated based on driving data with different communication efficiency ranges, and determining the score corresponding to the successfully matched communication efficiency range as the index score of the communication efficiency index.

[0139] The communication efficiency index is designed to evaluate the overall intelligence level of the tested vehicle when it is driving on actual roads when the intelligent driving function is activated, using the traffic efficiency of experienced human drivers on the test route as a reference. The evaluation method is as follows:

[0140] 1. Record the travel time TSV and TRV of the test vehicle and the reference vehicle from the start point to the end point of the test route respectively, and calculate the time difference ΔT=T after the test is completed. SV -T RV ;

[0141] 2. Count the ΔT of more than 3 tests and record them. Remove the maximum and minimum values ​​of ΔT and take the average of the remaining test scores to 2 decimal places to get .

[0142] (IV) Overall score

[0143] Calculate the total score of the test vehicle according to the preset evaluation formula;

[0144] The preset evaluation formulas include:

[0145] M 总 =W 安全 *∑(W 场 *∑(W 评指 *I 评指 ) / N 评指 )+W 体验 *∑(W 场 *∑(W 体指 *I 体指 ) / N 体指 )+W 效率 *I 效率

[0146] Among them, M 总 Indicates the total score, W 安全 represents the dimension weight of the safety and comfort dimension, W 场 represents the scene weight of each target test scene, W 评指 Indicates the indicator weight of each evaluation indicator, I 评指 Indicates the indicator score of each evaluation indicator, N 评指 represents the number of target test scenarios, W 体验 represents the dimension weight of the experience dimension, W 体指 Indicates the indicator weight of each experience indicator, I体指 Indicates the indicator score of each experience indicator, N 体指 Indicates the number of preset experience scenarios, W 效率 Represents the dimension weight of the traffic efficiency dimension, I 效率 Indicates the indicator score of the communication efficiency indicator.

[0147] The above weights can be determined by practical experience, for example, W 安全 It can be 80%, W 体验 It can be 10%, W 效率 It can be 10%. In addition, the weights of the same evaluation indicators in different test scenarios may be the same or different, and need to be determined according to the specific test scenario. For example, in a straight-line driving scenario, the weight of longitudinal acceleration is 30%, in a curved driving scenario, the weight of longitudinal acceleration is 15%, and in a traffic scene with a traffic light, the weight of longitudinal acceleration is 30%.

[0148] The above preset evaluation formula includes the scores for the safety and comfort dimensions, the experience dimension and the traffic efficiency dimension. Therefore, in addition to displaying the total score of the test vehicle, the evaluation report can also display the specific scores of these three dimensions and even the scores of each evaluation indicator for testers to view.

[0149] The embodiment of the present application first obtains the operational design domain (ODD) data of the test vehicle, and decomposes the ODD data into multiple ODD elements, and then matches the multiple ODD elements with the test scenario library to obtain at least one successfully matched target test scenario, and generates a test route corresponding to each target test scenario. After the test vehicle travels along each of the test routes, driving data can also be obtained. For each evaluation indicator in each target test scenario in the safety and comfort dimension, the indicator score of the evaluation indicator is determined according to the scoring rules of the driving data and the evaluation indicator. For each experience indicator in each preset experience scenario, the indicator score of the evaluation indicator is determined according to the scoring rules of the driving data and the experience indicator. Scoring rules, determine the index score of the experience index, for the communication efficiency index in the traffic efficiency dimension, determine the index score of the communication efficiency index according to the scoring rules of driving data and communication efficiency indicators, and finally determine the total score of the test vehicle according to the index score of each evaluation indicator, the index weight of each evaluation indicator, the scene weight of each target test scene, the number of scenes of the target test scene, the dimension weight of the safety and comfort dimension, the index score of each experience indicator, the index weight of each experience indicator, the number of scenes of the preset experience scene, the dimension weight of the experience dimension, the index score of the communication efficiency index and the dimension weight of the traffic efficiency dimension. It can be seen that the embodiment of the present application can not only comprehensively evaluate the overall behavior performance of intelligent driving from multiple dimensions, but also conduct targeted evaluation according to the test scene matched by the ODD data of the tested vehicle, so that automobile manufacturers can more accurately and intuitively determine whether the vehicles they manufacture are consistent with the ODD data.

[0150] Based on the above method embodiment, another embodiment of the present application provides an intelligent driving evaluation device, such as Figure 4 As shown, the device is applied to an intelligent driving evaluation terminal, the intelligent driving evaluation terminal includes a variety of standard sensors and a mobile communication module, and the intelligent driving evaluation terminal is installed on a test vehicle. The device includes:

[0151] The acquisition unit 310 is used to acquire, in real time during the driving process of the test vehicle, the first state data generated by each vehicle sensor in the test vehicle through the controller area network CAN protocol, acquire the second state data generated by each standard sensor in real time, acquire the legal value range of each sensor data on the driving road, and acquire the driving environment information around the current driving road from the cloud control platform in real time through the mobile communication network of the mobile communication module;

[0152] The judging and marking unit 320 is used to mark the first state data as normal data when both the first state data and the second state data are within the corresponding legal value range and the difference between the first state data and the second state data is within a preset difference range; mark the first state data as abnormal data when both the first state data and the second state data are within the corresponding legal value range and the difference between the first state data and the second state data is not within the preset difference range; mark the first state data as abnormal data when the first state data is not within the legal value range and the second state data is within the legal value range; mark the first state data as abnormal data when the first state data is within When the first state data is within the legal value range and the second state data is not within the legal value range, the first state data is marked as normal data and the second state data is marked as abnormal data; when both the first state data and the second state data are not within the legal value range, combined with the driving environment information, it is determined whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, the first state data is marked as normal data, and when it is determined that the first state data is not the target type of data, the first state data is marked as abnormal data, wherein the target type of data includes driving data generated by driving behaviors taken to avoid collision risks.

[0153] In a possible implementation, the device further includes:

[0154] A sending unit is used to send the marked first state data and the marked second state data to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormality detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform.

[0155] In a possible implementation, when there is at least one of a traffic light, a pedestrian, a motor vehicle, and a non-motor vehicle around the test vehicle, the driving environment information includes at least one of the status information of the traffic light, the status information of the pedestrian, the status information of the motor vehicle, and the status information of the non-motor vehicle.

[0156] The intelligent driving evaluation device provided in the embodiment of the present application can pre-install an intelligent driving evaluation terminal including a plurality of standard sensors and a mobile communication module on the test vehicle. In this way, during the driving of the test vehicle, the vehicle sensors on the test vehicle and the standard sensors in the intelligent driving evaluation terminal will measure the real-time status of the test vehicle, and can also obtain the legal value range of each sensor data on the driving road, and obtain the driving environment information around the current driving road from the cloud control platform in real time. Therefore, for the same type of sensor, when at least one of the first state data generated by the vehicle sensor and the second state data generated by the standard sensor is within the corresponding legal value range, it can be combined The difference between the two state data, and which state data is within the legal value range, is used to mark the first state data as normal or abnormal. When both state data are not within the legal value range, the driving environment information can be combined to further determine whether the first state data is driving data generated by the driving behavior taken to avoid collision risks. When the first state data is driving data generated by the driving behavior taken to avoid collision risks, it means that the first state data is a reasonable behavior made in a special scenario, and the first state data can be marked as normal data. When the first state data is not driving data generated by the driving behavior taken to avoid collision risks, the first state data can be marked as abnormal data. It can be seen from this that the intelligent driving evaluation terminal provided in the embodiment of the present application can not only automatically evaluate the driving behavior of the test vehicle in real time, but also reasonably evaluate the special driving behavior in special scenarios. On the basis of improving the evaluation efficiency, the fairness and rationality of the evaluation are guaranteed.

[0157] In addition, the intelligent driving evaluation terminal can also send the marked first state data and the marked second state data to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormal detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform. When the abnormal detection result of a certain first state data is incorrect, the tester can also correct the abnormal detection result.

[0158] Based on the above embodiments, another embodiment of the present application provides an intelligent driving evaluation terminal, which includes a plurality of standard sensors, a mobile communication module, a processor and a memory, and is installed on a test vehicle;

[0159] The standard sensor is used to measure the state of the test vehicle in real time and generate second state data;

[0160] The mobile communication module is used for wireless communication with the cloud control platform;

[0161] The memory is used to store one or more programs and a legal value range of each sensor data on the traveled road;

[0162] The processor is used to execute one or more programs to implement the intelligent driving evaluation method as described in any of the above embodiments.

[0163] Based on the above embodiments, another embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed on a computer or a processor, the computer or the processor executes the method described in any possible implementation manner above.

[0164] The above electronic device embodiments all correspond to the system embodiments and have the same technical effects. For specific descriptions, please refer to the system embodiments. A person skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present application.

[0165] Those skilled in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment according to the description of the embodiment, or can be changed accordingly and located in one or more devices different from the embodiment. The modules in the above embodiment can be combined into one module, or can be further divided into multiple sub-modules.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent driving evaluation method, characterized in that: The method is applied to an intelligent driving evaluation terminal, the intelligent driving evaluation terminal includes a plurality of standard sensors and a mobile communication module, and the intelligent driving evaluation terminal is installed on a test vehicle, and the method includes: During the driving process of the test vehicle, first state data generated by each vehicle sensor in the test vehicle is obtained in real time through the controller area network CAN protocol, second state data generated by each standard sensor is obtained in real time, a legal value range of each sensor data on the driving road is obtained, and driving environment information around the current driving road is obtained from the cloud control platform in real time through the mobile communication network of the mobile communication module; If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is within a preset difference range, marking the first state data as normal data; If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is not within the preset difference range, marking the first state data as abnormal data; If the first state data is not within the legal value range, and the second state data is within the legal value range, marking the first state data as abnormal data; If the first state data is within the legal value range and the second state data is not within the legal value range, marking the first state data as normal data and marking the second state data as abnormal data; If both the first state data and the second state data are not within the legal value range, combined with the driving environment information, determine whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, mark the first state data as normal data; when it is determined that the first state data is not the target type of data, mark the first state data as abnormal data, wherein the target type of data includes driving data generated by driving behaviors taken to avoid collision risks.

2. The method according to claim 1, characterized in that The method further comprises: The marked first state data and the marked second state data are sent to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormality detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform.

3. The method according to claim 1 or 2, characterized in that: When there is at least one of a traffic light, a pedestrian, a motor vehicle and a non-motor vehicle around the test vehicle, the driving environment information includes at least one of the status information of the traffic light, the status information of the pedestrian, the status information of the motor vehicle and the status information of the non-motor vehicle.

4. An intelligent driving evaluation device, characterized in that: The device is applied to an intelligent driving evaluation terminal, the intelligent driving evaluation terminal includes a plurality of standard sensors and a mobile communication module, and the intelligent driving evaluation terminal is installed on a test vehicle, and the device includes: an acquisition unit, used to acquire, in real time during the driving process of the test vehicle, first state data generated by each vehicle sensor in the test vehicle through a controller area network (CAN) protocol, acquire second state data generated by each standard sensor in real time, acquire a legal value range of each sensor data on the driving road, and acquire driving environment information around the current driving road from a cloud control platform in real time through the mobile communication network of the mobile communication module; a judging and marking unit, configured to mark the first state data as normal data when both the first state data and the second state data are within the corresponding legal value range and the difference between the first state data and the second state data is within a preset difference range; mark the first state data as abnormal data when both the first state data and the second state data are within the corresponding legal value range and the difference between the first state data and the second state data is not within the preset difference range; mark the first state data as abnormal data when the first state data is not within the legal value range and the second state data is within the legal value range; mark the first state data as abnormal data when the first state data is within the legal value range When the first state data is within the legal value range and the second state data is not within the legal value range, the first state data is marked as normal data and the second state data is marked as abnormal data; when both the first state data and the second state data are not within the legal value range, combined with the driving environment information, it is determined whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, the first state data is marked as normal data, and when it is determined that the first state data is not the target type of data, the first state data is marked as abnormal data, wherein the target type of data includes driving data generated by driving behavior taken to avoid collision risks.

5. The device according to claim 4, characterized in that The device also includes: A sending unit is used to send the marked first state data and the marked second state data to the intelligent driving evaluation platform through the mobile communication module, so as to verify the abnormality detection result of the first state data by replaying the marked first state data and the marked second state data on the intelligent driving evaluation platform.

6. The device according to claim 4 or 5, characterized in that When there is at least one of a traffic light, a pedestrian, a motor vehicle and a non-motor vehicle around the test vehicle, the driving environment information includes at least one of the status information of the traffic light, the status information of the pedestrian, the status information of the motor vehicle and the status information of the non-motor vehicle.

7. An intelligent driving evaluation terminal, characterized in that: The intelligent driving evaluation terminal includes a variety of standard sensors, a mobile communication module, a processor and a memory, and the intelligent driving evaluation terminal is installed on a test vehicle; The standard sensor is used to measure the state of the test vehicle in real time and generate second state data; The mobile communication module is used for wireless communication with the cloud control platform; The memory is used to store one or more programs and a legal value range of each sensor data on the traveled road; The processor is used to execute one or more programs to implement the method according to any one of claims 1 to 3.

8. An intelligent driving evaluation system, characterized in that: The system includes a test vehicle, an intelligent driving evaluation terminal and a cloud control platform, wherein the intelligent driving evaluation terminal includes a plurality of standard sensors and a mobile communication module, and the intelligent driving evaluation terminal is installed on the test vehicle; The test vehicle is used for each vehicle sensor to generate first state data during driving based on the automatic driving system; The cloud control platform is used to send the driving environment information around the current driving road of the test vehicle to the intelligent driving evaluation terminal; The intelligent driving evaluation terminal is used to: During the driving process of the test vehicle, the first state data is obtained in real time through the controller area network CAN protocol, the second state data generated by each standard sensor is obtained in real time, the legal value range of each sensor data on the driving road is obtained, and the driving environment information is obtained from the cloud control platform in real time through the mobile communication network of the mobile communication module; If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is within a preset difference range, marking the first state data as normal data; If the first state data and the second state data are both within the corresponding legal value range, and the difference between the first state data and the second state data is not within the preset difference range, marking the first state data as abnormal data; If the first state data is not within the legal value range, and the second state data is within the legal value range, marking the first state data as abnormal data; If the first state data is within the legal value range and the second state data is not within the legal value range, marking the first state data as normal data and marking the second state data as abnormal data; If both the first state data and the second state data are not within the legal value range, combined with the driving environment information, determine whether the first state data belongs to the target type of data, and when it is determined that the first state data is the target type of data, mark the first state data as normal data; when it is determined that the first state data is not the target type of data, mark the first state data as abnormal data, wherein the target type of data includes driving data generated by driving behaviors taken to avoid collision risks.

9. The system according to claim 8, characterized in that The system also includes an intelligent driving evaluation platform; The intelligent driving evaluation platform is used to obtain the marked first state data and the marked second state data sent by the intelligent driving evaluation terminal through the mobile communication module, so as to replay the marked first state data and the marked second state data to verify the abnormality detection result of the first state data.

10. The system according to claim 9, characterized in that The intelligent driving evaluation platform is further used to obtain the operational design domain (ODD) data of the test vehicle and decompose the ODD data into multiple ODD elements; Matching the multiple ODD elements with the test scenario library to obtain at least one successfully matched target test scenario, wherein the evaluation dimension corresponding to the at least one target test scenario is a safety and comfort dimension; Generating a test route corresponding to each of the target test scenarios so that the test vehicle travels along each of the test routes; Acquiring driving data generated by the test vehicle during driving along each of the test routes; For each evaluation indicator in each target test scenario in the safety and comfort dimension, determine the indicator score of the evaluation indicator according to the driving data and the scoring rule of the evaluation indicator; For each experience indicator in each preset experience scenario, determining an indicator score of the experience indicator according to the driving data and a scoring rule of the experience indicator; For the communication efficiency index in the traffic efficiency dimension, determining an index score of the communication efficiency index according to the driving data and a scoring rule of the communication efficiency index; The total score of the test vehicle is determined according to the indicator score of each of the evaluation indicators, the indicator weight of each of the evaluation indicators, the scenario weight of each of the target test scenarios, the number of scenarios of the target test scenarios, the dimension weight of the safety and comfort dimension, the indicator score of each of the experience indicators, the indicator weight of each of the experience indicators, the number of scenarios of the preset experience scenarios, the dimension weight of the experience dimension, the indicator score of the communication efficiency indicator and the dimension weight of the traffic efficiency dimension.

Images