Vehicle data processing method, device, storage medium and vehicle

Abstract

The present invention discloses a vehicle data processing method, device, storage medium and vehicle. The method comprises: obtaining first driving behavior data of the vehicle in an active driving mode; obtaining second driving behavior data of the vehicle in an automatic driving mode, wherein the second driving behavior data comprises first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the vehicle's environmental perception system, and the second expected behavior data is obtained by the vehicle's planning control system; at least the first driving behavior data and the second driving behavior data are recorded. The present invention solves the technical problem of low accuracy in collecting abnormal data of an automatic driving system.

Description

Technical Field

[0001] The present invention relates to the field of vehicles, and in particular to a method, device, storage medium and vehicle for processing vehicle data. Background Art

[0002] At present, in the process of developing vehicle autonomous driving systems, using driving behavior big data as a driver has become a mainstream research trend. However, the control output of the autonomous driving system is greatly affected by the recognition of the perception system. If the perception system misidentifies, it will also cause abnormal control of the autonomous driving, resulting in low accuracy in the collection of abnormal data of the autonomous driving system.

[0003] Currently, no effective solution has been proposed to the above-mentioned problem of low accuracy in collecting abnormal data of the autonomous driving system. Summary of the invention

[0004] Embodiments of the present invention provide a method, device, storage medium and vehicle for processing vehicle data, so as to at least solve the technical problem of low accuracy in collecting abnormal data of an autonomous driving system.

[0005] According to one aspect of an embodiment of the present invention, a method for processing vehicle data is provided, comprising: obtaining first driving behavior data of a vehicle in an active driving mode; obtaining second driving behavior data of the vehicle in an automatic driving mode, wherein the second driving behavior data comprises first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the vehicle's environmental perception system, and the second expected behavior data is obtained by the vehicle's planning control system; at least the first driving behavior data and the second driving behavior data are recorded.

[0006] Optionally, the output information of the environmental perception system is analyzed based on a neural network model to obtain first expected behavior data, wherein the output information is information perceived by the environmental perception system on the current road scene, and the neural network is a model generated based on analysis of operation instructions.

[0007] Optionally, based on the output information, second expected behavior data is determined.

[0008] Optionally, recording at least the first driving behavior data and the second driving behavior data includes: determining target driving behavior data based on the second driving behavior data; and in response to the target driving behavior data not including the first driving behavior data, recording at least the first driving behavior data and the second driving behavior data.

[0009] Optionally, determining the target driving behavior data based on the second driving behavior data includes: determining data in the second driving behavior data that is greater than a first threshold and less than a second threshold as the target driving behavior data, wherein the second threshold is greater than the first threshold.

[0010] Optionally, obtaining second driving behavior data of the vehicle in the automatic driving mode includes: based on the first expected behavior data, determining that the working state of the environment perception system is a normal working state, and then determining the second expected behavior data as the second driving behavior data.

[0011] Optionally, data in the first expected behavior data that is greater than a third threshold and less than a fourth threshold is determined as first target behavior data, wherein the fourth threshold is greater than the third threshold; in response to the first target behavior data including first driving behavior data, the state of the environmental perception system is determined to be a normal state.

[0012] According to another aspect of an embodiment of the present invention, a vehicle data processing device is also provided, including: a first acquisition unit, used to acquire first driving behavior data of the vehicle in an active driving mode; a second acquisition unit, used to acquire second driving behavior data of the vehicle in an automatic driving mode, wherein the second driving behavior data includes first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the vehicle's environmental perception system, and the second expected behavior data is obtained by the vehicle's planning control system; a recording unit, used to record at least the first driving behavior data and the second driving behavior data.

[0013] According to another aspect of an embodiment of the present invention, a vehicle is provided, which is used to execute the vehicle data processing method of the embodiment of the present invention.

[0014] In an embodiment of the present invention, first driving behavior data of the vehicle in the active driving mode is obtained; second driving behavior data of the vehicle in the automatic driving mode is obtained, wherein the second driving behavior data includes first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the vehicle's environmental perception system, and the second expected behavior data is obtained by the vehicle's planning control system; at least the first driving behavior data and the second driving behavior data are recorded. That is to say, the embodiment of the present invention determines the state of the environmental perception system by comparing the first driving behavior data with the first expected behavior data. If the environmental perception system is normal, the first driving behavior data is further compared with the second expected behavior data to determine the state of the planning system control. When an abnormality occurs in the environmental perception system or the planning control system, data recording is triggered to achieve the purpose of collecting abnormal data of the automatic driving system, thereby solving the technical problem of low accuracy in collecting abnormal data of the automatic driving system and achieving the technical effect of improving the accuracy of collecting abnormal data of the automatic driving system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0016] Figure 1 is a flow chart of a method for processing vehicle data according to an embodiment of the present invention;

[0017] Figure 2 is a schematic diagram of a vehicle data processing system according to an embodiment of the present invention;

[0018] Figure 3 is a schematic diagram of a vehicle data processing device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] Example 1

[0022] According to an embodiment of the present invention, an embodiment of a method for processing vehicle data is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0023] Figure 1 is a flow chart of a method for processing vehicle data according to an embodiment of the present invention. Figure 1 As shown, the vehicle data processing method may include the following steps:

[0024] Step S102, obtaining first driving behavior data of the vehicle in the active driving mode.

[0025] In the technical solution provided in the above step S102 of the present invention, when the vehicle is in an active driving mode, first driving behavior data of the vehicle is acquired, wherein the active driving mode may be a driving mode in which the driver manually controls the vehicle, and the first driving behavior data may be the current actual driving instruction data of the vehicle.

[0026] Optionally, the first driving behavior data can be obtained through the actual motion state of the vehicle, or based on the operation instructions actually issued by the driver. The first driving behavior data can be the vehicle's current actual driving instruction data, which can include the current actual steering wheel angle data, speed data and acceleration data. This is only an example and is not specifically limited.

[0027] Step S104, obtaining second driving behavior data of the vehicle in the automatic driving mode.

[0028] In the technical solution provided in the above step S104 of the present invention, second driving behavior data of the vehicle in the automatic driving mode is obtained, wherein the second driving behavior data may include first expected behavior data obtained by the vehicle's environmental perception system or second expected behavior data obtained by the vehicle's planning control system, the first expected behavior data may be expected instruction data output by the vehicle's environmental perception system, the second expected behavior data may be expected instruction data output by the vehicle's planning control system, the expected instruction data may include expected steering wheel angle data, expected speed data and expected acceleration data, which are merely examples here and are not specifically limited.

[0029] Optionally, the autonomous driving mode may acquire the second driving behavior data by virtually running the vehicle when the vehicle is in an active driving mode, wherein the autonomous driving mode may be a driving mode in which the vehicle's autonomous driving system controls the vehicle.

[0030] Optionally, the first driving data is compared with the first expected data to determine the state of the environment perception system, and if the working state of the environment perception system is normal, the second expected behavior data is determined as the second driving behavior data.

[0031] Optionally, data within a threshold range in the first expected data is extracted to obtain a set of data, and it is determined whether the first driving behavior data is within the set of data. If so, it indicates that the vehicle's environmental perception system is in a normal state.

[0032] Step S106: recording at least the first driving behavior data and the second driving behavior data.

[0033] In the technical solution provided in the above step S106 of the present invention, the first driving behavior data is compared with the second driving behavior data to determine whether there is an abnormality in the vehicle's automatic driving system. If an abnormality occurs, the first driving behavior data, the second driving behavior data and the vehicle's motion state data are recorded.

[0034] Optionally, data within a threshold range is extracted from the second driving behavior data to obtain a set of data, and it is determined whether the first driving behavior data is within the set of data. If not, it indicates that an abnormality occurs in the vehicle's automatic driving system.

[0035] For example, when the second driving behavior data is the first expected behavior data obtained based on the environmental perception system, a certain threshold range is set respectively, and the data within the threshold range is extracted to obtain set data, such as the steering wheel angle reference instruction set and the acceleration reference instruction set output by the environmental perception system. The first driving behavior data is obtained according to the vehicle motion state, such as the driver's operating instructions, that is, the current actual steering wheel angle and acceleration. It is determined whether the actual steering wheel angle and acceleration are in the reference instruction set. If not, it is determined that the perception system is abnormal, and the first driving behavior data, the second driving behavior data and the vehicle's motion state data are recorded. If so, it is determined that the perception system is normal, and the difference between the planning control system and the driver's behavior is further analyzed.

[0036] To take another example, when the second driving behavior data is the second expected behavior data obtained based on the planning control system, a certain threshold range is set respectively, and the data within the threshold range is extracted to obtain set data, such as the steering wheel angle reference instruction set and the acceleration reference instruction set output by the planning control system. The first driving behavior data, that is, the current actual steering wheel angle and acceleration, are obtained according to the vehicle motion state, and it is determined whether the actual steering wheel angle and acceleration are in the reference instruction set. If not, it is determined that the planning control system is abnormal, and the first driving behavior data, the second driving behavior data and the vehicle's motion state data are recorded.

[0037] Optionally, the vehicle's autonomous driving system may include an environmental perception system and a planning control system. If any abnormality occurs in either system, data recording will be triggered.

[0038] In the above steps S102 to S106 of this embodiment, the first driving behavior data of the vehicle in the active driving mode is obtained; the second driving behavior data of the vehicle in the automatic driving mode is obtained; and at least the first driving behavior data and the second driving behavior data are recorded. That is to say, the embodiment of the present invention determines the state of the environment perception system by comparing the first driving behavior data with the first expected behavior data. If the environment perception system is normal, the first driving behavior data is further compared with the second expected behavior data to determine the state of the planning system control. When an abnormality occurs in the environment perception system or the planning control system, data recording is triggered to achieve the purpose of collecting abnormal data of the automatic driving system, thereby solving the technical problem of low accuracy in collecting abnormal data of the automatic driving system and achieving the technical effect of improving the accuracy of collecting abnormal data of the automatic driving system.

[0039] The above method of this embodiment is further introduced below.

[0040] As an optional implementation method, step S104 analyzes the output information of the environmental perception system based on a neural network model to obtain first expected behavior data, wherein the output information is information perceived by the environmental perception system on the current road scene, and the neural network is a model generated based on analysis of operation instructions.

[0041] In this embodiment, the output information of the environmental perception system is input into a neural network model generated based on the analysis of the operation instructions for analysis to generate first expected behavior data, wherein the output information may include vehicle information, obstacle information around the vehicle, and road information on which the vehicle is located.

[0042] Optionally, the vehicle information may include the vehicle's position, speed, acceleration and other information; the obstacle information around the vehicle may include the type, position, speed and other information of the obstacles around the vehicle; the road information where the vehicle is located may include the number of lanes and intersection information of the road where the vehicle is located. These are merely examples and are not specifically limited.

[0043] Optionally, the first expected behavior data may include lateral operation instruction data and longitudinal operation instruction data. The lateral operation instruction data may be steering wheel angle data, and the longitudinal operation instruction data may be acceleration data. This is only used as an example and is not specifically limited.

[0044] Optionally, the neural network model can be based on a convolutional neural network. The neural network model is first established based on the image information of the environmental perception system, the lidar point cloud information and the motion information of the vehicle collected during the driving process, and then a training set of the operation instruction analysis model is constructed. The training of the neural network model is completed through active labeling.

[0045] As an optional implementation manner, the second expected behavior data is determined based on the output information.

[0046] In this embodiment, based on the output information of the environmental perception system, the output information of the environmental perception system is planned and controlled by the planning control system of the vehicle to obtain second expected behavior data, wherein the second expected behavior data may be expected driving instruction data output by the planning control system of the vehicle, and the expected driving instruction data may include expected steering wheel angle data and expected acceleration data.

[0047] As an optional implementation method, obtaining second driving behavior data of the vehicle in the automatic driving mode includes: based on the first expected behavior data, determining that the working state of the environmental perception system is a normal working state, and then determining the second expected behavior data as the second driving behavior data.

[0048] In this embodiment, data within a threshold range in the first expected data is extracted to obtain a set of data, and the first driving behavior data is compared with the set of data to determine the working state of the environmental perception system. When the working state of the environmental perception system is judged to be a normal working state, the second expected behavior data can be determined as the second driving behavior data, wherein the set data may include a steering wheel angle reference instruction set and an acceleration reference instruction set, and the working state of the environmental perception system may include a normal working state and an abnormal working state.

[0049] Optionally, if the first driving behavior data is within the above-mentioned set data, it indicates that the vehicle's environmental perception system is in a normal state; if not, it indicates that the vehicle's environmental perception system is in an abnormal state.

[0050] As an optional implementation method, data in the first expected behavior data that is greater than a third threshold and less than a fourth threshold is determined as first target behavior data, wherein the fourth threshold is greater than the third threshold; in response to the first target behavior data including the first driving behavior data, it is determined that the state of the environmental perception system is a normal state.

[0051] In this embodiment, data greater than the third threshold and less than the fourth threshold in the first expected behavior data is extracted, and the extracted data set is determined as the first target driving behavior data. When the first driving behavior data is judged to be the data in the first target behavior data, it is determined that the state of the environmental perception system is normal, wherein the third threshold and the fourth threshold can be used to characterize the preset threshold range.

[0052] For example, when the threshold range of the acceleration reference instruction set in the first target driving behavior data is 5 to 10, and the threshold range of the steering wheel angle reference instruction set in the first target driving behavior data is 2 to 5, the acceleration in the first driving behavior data can be any value in the range of 5 to 10, and the steering wheel angle in the first driving behavior data can be any value in the range of 2 to 5. If the acceleration in the first driving behavior data is 7 and the steering wheel angle is 3, it is determined that the state of the environmental perception system is normal.

[0053] As an optional implementation method, step S106, recording at least the first driving behavior data and the second driving behavior data, includes: determining target driving behavior data based on the second driving behavior data; in response to the target driving behavior data not including the first driving behavior data, recording at least the first driving behavior data and the second driving behavior data.

[0054] In this embodiment, data within a threshold range is extracted from the second driving behavior data to obtain a set of data, which is determined as the target driving behavior data. When it is determined that the target driving behavior data does not include the first driving behavior data, at least the first driving behavior data and the second driving behavior data are recorded.

[0055] Optionally, a comparative analysis of the target driving behavior data and the first driving behavior data can be used to determine the working state of the autonomous driving system. For example, when it is determined that the first driving behavior data does not fall within the range of the target driving behavior data, it can be determined that the working state of the autonomous driving system is an abnormal state.

[0056] For example, when the target driving behavior data ranges from 5 to 10 and the first driving behavior data is 3, it is determined that the working state of the automatic driving system is abnormal, and the recording of the first driving behavior data and the second driving behavior data is triggered.

[0057] As an optional implementation method, determining target driving behavior data based on the second driving behavior data includes: determining data in the second driving behavior data that is greater than a first threshold and less than a second threshold as target driving behavior data, wherein the second threshold is greater than the first threshold.

[0058] In this embodiment, data greater than the first threshold and less than the second threshold in the second driving behavior data is extracted, and the extracted data set is determined as the target driving behavior data, wherein the first threshold and the second threshold can be used to characterize a preset threshold range, and the target driving behavior data can be used to respond to trigger data recording.

[0059] For example, when the threshold range of the acceleration reference instruction set in the target driving behavior data is 5 to 10, and the threshold range of the steering wheel angle reference instruction set in the target driving behavior data is 2 to 5, the acceleration in the second driving behavior data can be any value in the range of 5 to 10, and the steering wheel angle in the second driving behavior data can be any value in the range of 2 to 5. If the acceleration in the second driving behavior data is 7 and the steering wheel angle is 3, it is determined that the state of the planned control system is normal; if the acceleration of the second driving behavior data is 12 and the steering wheel angle is 9, it is determined that the state of the planned control system is abnormal.

[0060] This embodiment determines the state of the environment perception system by comparing the first driving behavior data with the first expected behavior data. If the environment perception system is normal, the first driving behavior data is further compared with the second expected behavior data to determine the state of the planning system control. When an abnormality occurs in the environment perception system or the planning control system, data recording is triggered to achieve the purpose of collecting abnormal data of the automatic driving system, thereby solving the technical problem of low accuracy in collecting abnormal data of the automatic driving system and achieving the technical effect of improving the accuracy of collecting abnormal data of the automatic driving system.

[0061] Example 2

[0062] The technical solution of the embodiment of the present invention is illustrated below in conjunction with preferred implementation modes.

[0063] The autonomous driving of vehicles must cope with complex real-world traffic scenarios, especially high-level autonomous driving. The most important feature of high-level autonomous driving is that it has no scene restrictions. That is, the development of the autonomous driving system should be based on unrestricted scenes as verification assumptions. The ability to cover complex environments and unfamiliar and sudden scenes is the core issue in the development of the autonomous driving system.

[0064] Since vehicles can collect a large amount of real scene data while driving, in order to efficiently filter out the scenes required for autonomous driving from a large amount of scene data, the autonomous driving system can be compared with the driver's behavior in the active driving state. When the two are inconsistent, the scene is determined to be a valid scene. By using driving behavior big data as a driver, the development of vehicle autonomous driving systems has become an international mainstream research trend.

[0065] In a related technology, autonomous vehicles can obtain road traffic environment data, vehicle driving status data, driver control behavior data and other data through the environmental perception system. When actively driving, the autonomous driving system can be operated virtually, and the autonomous driving control instructions can be compared with the driver's control instructions to analyze the differences, and the scene data with differences can be screened out to support the optimization development and testing of the autonomous driving system.

[0066] In another related technology, a control method for triggering recording of human-vehicle differential data based on a dynamic time warping algorithm is proposed. According to the driver's control input and the information input of the environmental perception system, the perception results of the environmental perception system are transmitted to the automatic driving algorithm, and the expected acceleration / deceleration and the expected steering wheel angle are calculated and output. The throttle / brake pedal opening and the steering wheel angle are used as the driver's control input. Through a calibrated model, the driver's pedal opening is converted into acceleration / deceleration. Through the differential real-time triggering control recording method based on the time warping algorithm, the automatic triggering recording of differential data is realized, replacing the "active calibration, manual triggering" method that requires the intervention of assistants in the traditional solution, which can greatly reduce the amount of redundant data. However, there is still a problem that the control output of the automatic driving system is greatly affected by the recognition of the perception system. If the perception is misidentified, it will also cause abnormal control of the automatic driving, and the difference cannot be well compared.

[0067] In another related technology, an autonomous driving data recorder is also proposed, including a camera, an image sensor, a data recording processor, an image and machine vision processor, a memory, and a high-speed solid-state storage medium, wherein the image sensor is connected to the data recording processor; the camera is connected to the image sensor for real-time shooting, and the captured original image is sent to the data recording processor through the image sensor; the data recording processor is connected to the image and machine vision processor, the memory, and the high-speed solid-state storage medium. By adding a recording processor between the traditional machine vision processor and the image sensor, the consistency of the recorded image and the image used for the reasoning of the machine vision algorithm is guaranteed, but no content is mentioned about effectively supporting the development and testing of automatic control systems.

[0068] In another related technology, a storage method, device and storage medium for autonomous driving data are also proposed, the method comprising: during the autonomous driving process, obtaining the autonomous driving data of the vehicle, the autonomous driving data including at least the driving data of the vehicle, the environmental visual data of the current environment and the traffic data; storing the autonomous driving data through a recording terminal installed in the vehicle; when it is detected that the storage period has been reached, transferring the autonomous driving data stored in the storage period to an enterprise data recording platform, the enterprise data recording platform is used to store the autonomous driving data of the vehicle, and sending the data in the autonomous driving data that meets the upload requirements to the regulatory data platform, thereby improving the comprehensiveness of the autonomous driving data storage and the utilization rate of the autonomous driving data, but there is still a problem of low accuracy in collecting abnormal data of the autonomous driving system.

[0069] In another related technology, a management system and method for autonomous driving data, and a data processing system are proposed. A data storage system and a data processing system are provided in the management system for autonomous driving data. The data storage system stores the autonomous driving data actually generated by the autonomous driving control system, and classifies and stores the output data of each module of the autonomous driving control system. The data processing system provides a user interface, receives the time selected by the user for viewing, and operates the data of the modules in the autonomous driving control system. According to the data operation method, the input data or output data of the module selected by the user is read from the data storage system, and the module is operated using the data operation method selected by the user and the read data, thereby improving the utilization rate and interactivity of the autonomous driving data. However, no content related to effectively supporting the development and testing of automatic control systems is mentioned.

[0070] In order to solve the above problems, the embodiments of the present invention are based on the differences between the information of the autonomous driving system and the information of the active driving system. By establishing a model to compare and analyze the differences between the environmental perception system and the active driving system, relevant data is collected when an abnormality of the autonomous driving system is triggered. After the perception system fails to compare the differences, the differences between the planning control system and the active driving system are further analyzed to obtain scene data that is inconsistent between the autonomous driving system and the active driving system, so as to identify abnormal scenes using different modules, and apply the corresponding data to the development and testing of different modules, so as to provide data support for the development and testing of the autonomous driving system.

[0071] The following is an example of a vehicle data processing system.

[0072] Figure 2 is a schematic diagram of a vehicle data processing system according to an embodiment of the present invention. Figure 2 As shown, the vehicle data processing system 20 may include: an automatic driving system 202 , an active driving system 204 , and an automatic driving abnormal data triggering collection system 206 .

[0073] The autonomous driving system 202 is used to obtain autonomous driving information.

[0074] The autonomous driving system may include an environment perception system and a planning control system, and the autonomous driving information may include environment perception information obtained through the environment perception system and planning control information obtained through the planning control system.

[0075] Optionally, environmental perception information may include vehicle information, information about obstacles around the vehicle, and information about the road the vehicle is on. The vehicle information may include information such as the vehicle's position, speed, acceleration, etc. The obstacle information around the vehicle may include information such as the type, position, speed, etc. of the obstacles around the vehicle. The road information where the vehicle is located may include the number of lanes and intersection information of the road where the vehicle is located. This is only an example and is not specifically limited. The road information where the vehicle is located can be obtained through a high-precision map and the vehicle's position.

[0076] Optionally, the planning and control information may be based on an autonomous driving system that is virtually running in an active driving state, which plans and controls the environmental perception system information to obtain the desired steering wheel angle and acceleration output, and may include angle requests and acceleration requests, which are only used as examples here without specific limitations.

[0077] The active driving system 204 is used to obtain the driver's operating instructions.

[0078] The active driving system can obtain the driver's operating instructions based on the vehicle's motion state, wherein the driver's operating instructions can be operating instructions obtained based on the vehicle's current actual driving information, and can include the current actual steering wheel angle and acceleration.

[0079] The autonomous driving abnormal data triggering collection system 206 is used to analyze the difference between the autonomous driving system and the active driving system, and trigger data recording based on the analysis results, wherein the analysis of the difference between the autonomous driving system and the active driving system may be the analysis of the difference between the environmental perception system and the active driving system, or the analysis of the difference between the planning control system and the active driving system.

[0080] The following is a further introduction to the difference analysis between the environmental perception system and the active driving system.

[0081] An operation instruction analysis model based on environmental perception is established, and the acquired environmental perception information is input into the operation instruction analysis model to output the vehicle operation instructions, wherein the operation instruction analysis model can be a neural network model with a convolutional neural network as the main body. The operation instruction analysis model based on environmental perception can be established based on the environmental perception system image information, lidar point cloud information and vehicle motion information collected during vehicle driving, to construct a training set for the operation instruction analysis model, and the training of the neural network model for perception-based operation instruction analysis is completed through active labeling.

[0082] Inputting the acquired environmental perception information into the operation instruction analysis model can be converting the environmental perception information into a raster image as a model input, and the output vehicle operation instructions can be used to change the driving state of the vehicle, and can include lateral operation instructions and longitudinal operation instructions, wherein the lateral operation instruction can be an expected steering wheel angle instruction, and the longitudinal operation instruction can be an expected acceleration instruction.

[0083] A certain threshold is set to obtain a steering wheel angle reference instruction set and an acceleration reference instruction set. The expected steering wheel angle instruction and vehicle speed instruction obtained are compared with the threshold, and it is determined whether the actual steering wheel angle and acceleration fall within the corresponding reference instruction set. If it is determined that the actual steering wheel angle and acceleration do not fall within the reference instruction set, the environmental perception system is judged to be in an abnormal state, and the environmental perception system data, driver operation data and vehicle motion state data in the current scene are recorded; if it is determined that the actual steering wheel angle and acceleration fall within the reference instruction set, the environmental perception system is judged to be in a normal state, and the difference between the planning control system and the active driving system is further analyzed.

[0084] For example, at a traffic light intersection, the current actual scene is a green light, and the driver operates the vehicle through the intersection. If the environmental perception system mistakenly identifies the current actual scene as a red light, the output of the autonomous driving system will be to slow down and stop the vehicle. The actual acceleration of the vehicle will be very different from the expected acceleration. Therefore, it can be determined that the environmental perception system is abnormal, triggering the recording of data.

[0085] The following is a further introduction to the difference analysis between the planning control system and the active driving system.

[0086] When the active driving system is in working state, the autonomous driving system operates virtually, performs planning and control according to the acquired environmental perception information, and outputs planning control information, wherein the planning control information may include the desired steering wheel angle and acceleration.

[0087] Based on the acquired expected steering wheel angle and acceleration, as well as the steering wheel angle reference instruction set and the acceleration reference instruction set, it is determined whether the actual steering wheel angle and acceleration fall within the corresponding reference instruction set. If it is determined that the actual steering wheel angle and acceleration do not fall within the reference instruction set, it is determined that the planning control system is in an abnormal state, and the planning control system data, driver operation data and vehicle motion state data in the current scenario are recorded; if it is determined that the actual steering wheel angle and acceleration fall within the reference instruction set, it is determined that the planning control system is in a normal state and data recording is not triggered.

[0088] An embodiment of the present invention discloses a vehicle data processing system. During active driving, the automatic driving system virtually runs, generates vehicle reference instructions according to the environmental perception system, and compares them with the operating instructions of the active driving system to determine the working state of the environmental perception system. When it is judged that the environmental perception system is abnormal, data collection is triggered. When it is judged that the environmental perception system is normal, the difference between the control instructions of the planning control system and the operating instructions of the active driving system is further analyzed. When it is judged that the planning control system is abnormal, data collection is triggered to achieve the purpose of supporting the development and testing of the planning control system, thereby achieving the technical effect of improving the accuracy of collecting abnormal data of the automatic driving system and solving the technical problem of low accuracy in collecting abnormal data of the automatic driving system.

[0089] Example 3

[0090] According to an embodiment of the present invention, a vehicle data processing device is also provided. It should be noted that the vehicle data processing device of this embodiment can be used to execute the vehicle data processing method in embodiment 1 of the present invention.

[0091] Figure 3 is a schematic diagram of a vehicle data processing device according to an embodiment of the present invention. Figure 3 As shown, the vehicle data processing device 30 may include: a first acquisition unit 302 , a second acquisition unit 304 , and a recording unit 306 .

[0092] A first acquisition unit 302 is used to acquire first driving behavior data of the vehicle in the active driving mode;

[0093] A second acquisition unit 304 is used to acquire second driving behavior data of the vehicle in the automatic driving mode, wherein the second driving behavior data includes first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the environment perception system of the vehicle, and the second expected behavior data is obtained by the planning control system of the vehicle;

[0094] The recording unit 306 is configured to record at least the first driving behavior data and the second driving behavior data.

[0095] Optionally, the second acquisition unit 304 may include: an analysis module, which can be used to analyze the output information of the environmental perception system based on a neural network model to obtain first expected behavior data, wherein the output information is information perceived by the environmental perception system of the current road scene, and the neural network is a model generated based on analysis of operation instructions.

[0096] Optionally, the first analysis module may include: a first determination submodule, which may be used to determine second expected behavior data based on the output information.

[0097] Optionally, the second acquisition unit 304 may include: a first determination module, which may be used to determine that the working state of the environment perception system is a normal working state based on the first expected behavior data, and then determine the second expected behavior data as the second driving behavior data.

[0098] Optionally, the first determination module may include: a second determination submodule, which may be used to determine data in the first expected behavior data that is greater than a third threshold and less than a fourth threshold as first target behavior data, wherein the fourth threshold is greater than the third threshold.

[0099] Optionally, the first determination module may include: a third determination submodule, which may be used to determine that the state of the environment perception system is a normal state in response to the first target behavior data including the first driving behavior data.

[0100] Optionally, the recording unit 306 may include: a second determination module, which may be configured to determine target driving behavior data based on the second driving behavior data.

[0101] Optionally, the recording unit 306 may include: a recording module, which may be configured to record at least the first driving behavior data and the second driving behavior data in response to the target driving behavior data not including the first driving behavior data.

[0102] In an embodiment of the present invention, a first acquisition unit is used to acquire first driving behavior data of a vehicle in an active driving mode; a second acquisition unit is used to acquire second driving behavior data of a vehicle in an automatic driving mode, wherein the second driving behavior data includes first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the vehicle's environmental perception system, and the second expected behavior data is obtained by the vehicle's planning and control system; a recording unit is used to record at least the first driving behavior data and the second driving behavior data, thereby achieving a technical effect of improving the accuracy of collecting abnormal data of an automatic driving system and solving the technical problem of low accuracy in collecting abnormal data of an automatic driving system.

[0103] Example 4

[0104] According to an embodiment of the present invention, a computer-readable storage medium is further provided. The computer-readable storage medium includes a stored program, wherein the program executes the method for processing vehicle data described in Embodiment 1.

[0105] Example 5

[0106] According to an embodiment of the present invention, a processor is further provided. The processor is used to run a program, wherein the method for processing vehicle data described in Embodiment 1 is executed when the program is run.

[0107] Example 6

[0108] According to an embodiment of the present invention, a vehicle is further provided. The vehicle is used to execute the vehicle data processing method according to the embodiment of the present invention.

[0109] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0110] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0112] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0113] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0114] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0115] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for processing vehicle data, characterized in that: include: Acquiring first driving behavior data of the vehicle in an active driving mode; Acquiring second driving behavior data of the vehicle in the automatic driving mode, wherein the second driving behavior data includes first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the environment perception system of the vehicle, and the second expected behavior data is obtained by the planning control system of the vehicle; recording at least the first driving behavior data and the second driving behavior data; Wherein, obtaining the second driving behavior data of the vehicle in the automatic driving mode includes: based on the first expected behavior data, determining that the working state of the environment perception system is a normal working state, and determining the second expected behavior data as the second driving behavior data; At least recording the first driving behavior data and the second driving behavior data includes: determining target driving behavior data based on the second driving behavior data; and in response to the target driving behavior data not including the first driving behavior data, recording at least the first driving behavior data and the second driving behavior data.

2. The method according to claim 1, characterized in that include: The output information of the environmental perception system is analyzed based on a neural network model to obtain the first expected behavior data, wherein the output information is information perceived by the environmental perception system on the current road scene, and the neural network is a model generated based on analysis of operation instructions.

3. The method according to claim 2, characterized in that include: Based on the output information, the second expected behavior data is determined.

4. The method according to claim 1, characterized in that: Determining target driving behavior data based on the second driving behavior data includes: The data in the second driving behavior data that is greater than a first threshold and less than a second threshold is determined as the target driving behavior data, wherein the second threshold is greater than the first threshold.

5. The method according to claim 1, characterized in that include: Determine the data in the first expected behavior data that is greater than a third threshold and less than a fourth threshold as first target behavior data, wherein the fourth threshold is greater than the third threshold; In response to the first target behavior data including the first driving behavior data, it is determined that the working state of the environment perception system is a normal working state.

6. A vehicle data processing device, characterized in that: include: A first acquisition unit, configured to acquire first driving behavior data of the vehicle in an active driving mode; a second acquisition unit, configured to acquire second driving behavior data of the vehicle in the automatic driving mode, wherein the second driving behavior data includes first expected behavior data or second expected behavior data, the first expected behavior data is obtained by the environment perception system of the vehicle, and the second expected behavior data is obtained by the planning control system of the vehicle; a recording unit, configured to record at least the first driving behavior data and the second driving behavior data; The second acquisition unit is specifically configured to determine, based on the first expected behavior data, that the working state of the environment perception system is a normal working state, and then determine the second expected behavior data as the second driving behavior data; The recording unit is specifically configured to determine target driving behavior data based on the second driving behavior data; and in response to the target driving behavior data not including the first driving behavior data, record at least the first driving behavior data and the second driving behavior data.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 5.

8. A vehicle, characterized in that: The vehicle is used to execute the vehicle data processing method described in any one of claims 1 to 5.

Images