Data processing method and system for L4-level autonomous vehicle, and medium

Abstract

The invention relates to a data processing method and system for an L4-level automatic driving vehicle and a medium, and the method comprises the steps: U1, collecting the motion state data information of the vehicle in the driving process of the vehicle, and carrying out the deserialization and cleaning processing, and obtaining the processed motion state data information of the vehicle; and U2, inputting the processed motion state data information of the vehicle into a trained multivariate nonlinear regression model based on contribution degree verification, and representing the correlation between the motion state and the operation quality of the vehicle to obtain correlation data information between the motion state and the operation quality of the vehicle. According to the invention, the data of the vehicle can be accurately processed, the communication between the automatic driving vehicle and the platform and the cloud uploading of the data packet do not need to depend on a private network, the function can be realized through a public network, and the project construction cost can be reduced.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a data processing method, system and medium for an L4 level autonomous driving vehicle. Background Art

[0002] With the continuous development of science and technology, autonomous driving technology is also constantly developing and becoming a topic of concern and research direction.

[0003] In the prior art, a Chinese patent (application number: 202210862664.3, publication number: CN115098554A) discloses a vehicle autonomous driving data processing method, device, electronic device and storage medium, receiving at least one task to be processed sent from the cloud; collecting the specified data of the vehicle according to the task to be processed; filtering the specified data according to the preset filtering rules corresponding to the task to be processed to obtain the filtered data; uploading the filtered data to the cloud, so that the user can download the data corresponding to the task to be processed from the cloud based on the terminal device and process it. In this solution, the connection between the vehicle and the platform is based on the local area network in the test road. For areas without a local area network, it is impossible to upload data to the cloud in real time, and the construction cost of the local area network is relatively high; at the same time, this technology collects L3 level autonomous driving data, and does not involve higher level L4 level autonomous driving data processing and simulation.

[0004] In the prior art, a Chinese patent (application number: 202311802565.7, publication number: CN117950334A) discloses a simulation data processing method, device, storage medium and vehicle for an autonomous driving vehicle, the method comprising: in response to receiving a simulation operation on an autonomous driving vehicle, obtaining simulation data of the autonomous driving vehicle, wherein the simulation data is used to represent data generated by accident simulation of the autonomous driving vehicle; determining alarm data and a vehicle speed set of the autonomous driving vehicle based on the simulation data, wherein the alarm data is used to represent the alarm state of the autonomous driving vehicle during the accident simulation process, and the vehicle speed set includes the vehicle speed of the autonomous driving vehicle at different driving time points; determining the start interception time point and the end interception time point according to the alarm data and the vehicle speed set; intercepting the simulation data based on the start interception time point and the end interception time point to obtain the target simulation data. In this scheme, the simulation data is not fully obtained, but partially intercepted, and manual acquisition of simulation data is not supported, which limits the tracking of problems. At the same time, the analysis and processing of simulation data does not form a batch processing of data sets, and there are still limitations in efficiency. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a data processing method, system and medium for an L4 level autonomous driving vehicle, which can not only accurately process the vehicle data, but also the communication between the autonomous driving vehicle and the platform and the cloud uploading of data packets do not need to rely on a dedicated network. The functions can be realized through the public network, which can reduce the project construction cost.

[0006] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows: a data processing method for an L4 level autonomous driving vehicle, the method comprising:

[0007] U1. During the driving process of the vehicle, the vehicle's motion state data information is collected, and deserialized and cleaned to obtain the processed vehicle's motion state data information;

[0008] U2. Input the processed vehicle motion state data information into the trained multivariate nonlinear regression model based on contribution verification, characterize the correlation between the vehicle motion state and the running quality, and obtain the correlation data information between the vehicle motion state and the running quality;

[0009] U3. Based on the correlation data information between the motion state and the running quality of the vehicle, a neural network algorithm based on an adaptive feedback adjustment factor is used to characterize the running quality of the vehicle to obtain the running quality data information of the vehicle;

[0010] U4. Based on the running quality data information of the vehicle, construct a running quality evaluation function F of the vehicle, evaluate the running quality of the vehicle, and obtain the running quality evaluation data information of the vehicle.

[0011] Furthermore, the method further comprises:

[0012] U5. Based on the operation quality assessment data information of the vehicle, a preset threshold is set. If the operation quality assessment data of the vehicle is less than the preset threshold, the operation quality of the vehicle does not meet the requirements and the operation status of the vehicle needs to be adjusted. If the operation quality assessment data of the vehicle is greater than the preset threshold, the operation quality of the vehicle meets the requirements and drives normally.

[0013] Furthermore, in step U2, the trained multivariate nonlinear regression model based on contribution verification includes:

[0014] U21. Based on the processed vehicle motion state data information, construct a vehicle motion state data set, and establish a contribution function Q of the vehicle motion state data set i ,

[0015]

[0016] Among them, xi is the motion state data information of the vehicle at the i-th moment after processing, x j is the motion state data information of the vehicle at the jth moment after processing, m is the sample size, and the contribution of the vehicle's motion state data set is calculated to obtain the contribution data information of the vehicle's motion state data set;

[0017] U22. Based on the contribution data information of the vehicle's motion state data set, the vehicle's motion state data set is arranged in ascending order of contribution, and the vehicle's motion state data set is divided into a training data set Train and a verification data set Val,

[0018]

[0019] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, x k is the motion state data information of the vehicle at the kth moment after processing, m is the sample size, k is the cutoff point, n is a constant parameter, and the training data set of the motion state of the vehicle and the verification data set of the motion state of the vehicle are obtained;

[0020] U23. Based on the training data set of the vehicle's motion state, establish a first characterization function W1 of the vehicle's running quality,

[0021]

[0022] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, n is a constant parameter, r is the weight vector of the vehicle's motion state, α is the first characterization parameter of the vehicle's running quality, g is the high-dimensional feature space function, b is the bias constant parameter of the model, the first characterization parameter of the vehicle's running quality is optimized to obtain the optimized first characterization parameter data information of the vehicle's running quality, based on the verification data set of the vehicle's motion state, a second characterization function W2 of the vehicle's running quality is established,

[0023]

[0024] Among them, x k is the vehicle motion state data information at the kth moment after processing, k is the demarcation point, n is the constant parameter, m is the sample size, δ is the second characterization parameter of the vehicle running quality, b is the bias constant parameter of the model, β k To optimize the factor, the second characterization parameter of the vehicle running quality is optimized to obtain optimized data information of the second characterization parameter of the vehicle running quality;

[0025] U24. Based on the second parameter data information representing the optimized vehicle running quality and the first parameter data information representing the optimized vehicle running quality, a multivariate nonlinear regression R for contribution verification is established.

[0026]

[0027] Where m is the sample size, x i is the motion state data information of the vehicle at the i-th moment after processing, γ i is the relationship determining factor of the vehicle running quality, α0 is the first characterization parameter data information of the optimized vehicle running quality, δ0 is the second characterization parameter data information of the optimized vehicle running quality, b is the bias constant parameter of the model, f is the multivariate relationship vector regression function, which characterizes the correlation between the vehicle's motion state and running quality, and obtains the correlation data information between the vehicle's motion state and running quality.

[0028] Furthermore, the high-dimensional feature space function g is,

[0029]

[0030] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, η i is the weight vector of the high-dimensional feature space.

[0031] Furthermore, the multivariate relationship vector regression function f is,

[0032]

[0033] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, α0 is the first characterization parameter data information of the optimized vehicle running quality, and δ0 is the second characterization parameter data information of the optimized vehicle running quality.

[0034] Furthermore, the optimization factor β k The constraints are:

[0035]

[0036] Furthermore, in step U3, the characterization of the vehicle running quality by using a neural network algorithm based on an adaptive feedback adjustment factor includes:

[0037] U31. Based on the correlation data information of the motion state and running quality of the vehicle, construct a data set of correlation between the motion state and running quality of the vehicle;

[0038] U32. Input the data set of correlation between the vehicle's motion state and running quality into a neural network model based on an adaptive feedback adjustment factor for training and learning, and determine the neuron function S,

[0039]

[0040] Among them, y is the correlation data set between the vehicle's motion state and running quality, λ1, λ2 and λ3 are the adaptive feedback adjustment factors of the neural network model, and the trained neural network model based on the adaptive feedback adjustment factor is obtained;

[0041] U33. Based on the trained neural network model based on the adaptive feedback adjustment factor, the correlation data information between the vehicle's motion state and operation quality is input to characterize the vehicle's operation quality and obtain the vehicle's operation quality data information.

[0042] Furthermore, the vehicle operation quality evaluation function F is:

[0043]

[0044] Among them, z is the running quality data information of the vehicle, and μ1, μ2 and μ3 are the evaluation factors of the running quality of the vehicle.

[0045] In order to achieve the above-mentioned objectives and other related objectives, the present invention also provides a data processing system for an L4 level autonomous driving vehicle, including a computer device, which is programmed or configured to execute the steps of any one of the data processing methods for an L4 level autonomous driving vehicle.

[0046] In order to achieve the above-mentioned objectives and other related objectives, the present invention also provides a computer-readable storage medium, which stores a computer program programmed or configured to execute any one of the data processing methods for L4 level autonomous driving vehicles described.

[0047] The present invention has the following positive effects:

[0048] 1. The present invention characterizes the correlation between the vehicle's motion state and operation quality through a trained multivariate nonlinear regression model based on contribution verification, and characterizes the vehicle's operation quality in combination with a neural network algorithm based on an adaptive feedback adjustment factor. It can not only accurately process the vehicle data, but also the communication between the autonomous driving vehicle and the platform and the cloud upload of data packets do not need to rely on a dedicated network. The functions can be realized through the public network, which can reduce the project construction cost.

[0049] 2. The present invention evaluates the vehicle's operating quality by constructing a vehicle operating quality evaluation function F, which can not only further process the L4 level autonomous driving vehicle, but also complete driving tasks and monitor the driving environment. At this stage, all driving-related tasks have nothing to do with the driver and passengers. L4 level autonomous driving technology is the future development direction of the industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0051] Figure 2 It is a flow chart of the multivariate nonlinear regression model trained based on contribution verification of the present invention;

[0052] Figure 3 It is a schematic diagram of the flow of the neural network algorithm based on the adaptive feedback adjustment factor of the present invention;

[0053] Figure 4 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0054] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0055] Example 1: Figure 1 As shown, a data processing method for an L4 level autonomous driving vehicle, the method comprising:

[0056] U1. During the driving process of the vehicle, the vehicle's motion state data information is collected, and deserialized and cleaned to obtain the processed vehicle's motion state data information;

[0057] U2. Input the processed vehicle motion state data information into the trained multivariate nonlinear regression model based on contribution verification, characterize the correlation between the vehicle motion state and the running quality, and obtain the correlation data information between the vehicle motion state and the running quality;

[0058] U3. Based on the correlation data information between the motion state and the running quality of the vehicle, a neural network algorithm based on an adaptive feedback adjustment factor is used to characterize the running quality of the vehicle to obtain the running quality data information of the vehicle;

[0059] U4. Based on the running quality data information of the vehicle, construct a running quality evaluation function F of the vehicle, evaluate the running quality of the vehicle, and obtain the running quality evaluation data information of the vehicle.

[0060] In this embodiment, the method further includes:

[0061] U5. Based on the operation quality assessment data information of the vehicle, a preset threshold is set. If the operation quality assessment data of the vehicle is less than the preset threshold, the operation quality of the vehicle does not meet the requirements and the operation status of the vehicle needs to be adjusted. If the operation quality assessment data of the vehicle is greater than the preset threshold, the operation quality of the vehicle meets the requirements and drives normally.

[0062] In this embodiment, if Figure 2 As shown, in step U2, the trained multivariate nonlinear regression model based on contribution verification includes:

[0063] U21. Based on the processed vehicle motion state data information, construct a vehicle motion state data set, and establish a contribution function Q of the vehicle motion state data set i ,

[0064]

[0065] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, x j is the motion state data information of the vehicle at the jth moment after processing, m is the sample size, and the contribution of the vehicle's motion state data set is calculated to obtain the contribution data information of the vehicle's motion state data set;

[0066] U22. Based on the contribution data information of the vehicle's motion state data set, the vehicle's motion state data set is arranged in ascending order of contribution, and the vehicle's motion state data set is divided into a training data set Train and a verification data set Val,

[0067]

[0068] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, x k is the motion state data information of the vehicle at the kth moment after processing, m is the sample size, k is the cutoff point, n is a constant parameter, and the training data set of the motion state of the vehicle and the verification data set of the motion state of the vehicle are obtained;

[0069] U23. Based on the training data set of the vehicle's motion state, establish a first characterization function W1 of the vehicle's running quality,

[0070]

[0071] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, n is a constant parameter, r is the weight vector of the vehicle's motion state, α is the first characterization parameter of the vehicle's running quality, g is the high-dimensional feature space function, b is the bias constant parameter of the model, the first characterization parameter of the vehicle's running quality is optimized to obtain the optimized first characterization parameter data information of the vehicle's running quality, based on the verification data set of the vehicle's motion state, a second characterization function W2 of the vehicle's running quality is established,

[0072]

[0073] Among them, x k is the vehicle motion state data information at the kth moment after processing, k is the demarcation point, n is the constant parameter, m is the sample size, δ is the second characterization parameter of the vehicle running quality, b is the bias constant parameter of the model, β k To optimize the factor, the second characterization parameter of the vehicle running quality is optimized to obtain optimized data information of the second characterization parameter of the vehicle running quality;

[0074] U24. Based on the second parameter data information representing the optimized vehicle running quality and the first parameter data information representing the optimized vehicle running quality, a multivariate nonlinear regression R for contribution verification is established.

[0075]

[0076] Where m is the sample size, x i is the motion state data information of the vehicle at the i-th moment after processing, γ i is the relationship determining factor of the vehicle running quality, α0 is the first characterization parameter data information of the optimized vehicle running quality, δ0 is the second characterization parameter data information of the optimized vehicle running quality, b is the bias constant parameter of the model, f is the multivariate relationship vector regression function, which characterizes the correlation between the vehicle's motion state and running quality, and obtains the correlation data information between the vehicle's motion state and running quality.

[0077] In this embodiment, the high-dimensional feature space function g is,

[0078]

[0079] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, η i is the weight vector of the high-dimensional feature space.

[0080] In this embodiment, the multivariate relationship vector regression function f is,

[0081]

[0082] Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, α0 is the first characterization parameter data information of the optimized vehicle running quality, and δ0 is the second characterization parameter data information of the optimized vehicle running quality.

[0083] In this embodiment, the optimization factor β k The constraints are:

[0084]

[0085] Example 2: Based on the data processing method for an L4 level autonomous driving vehicle in Example 1, the present invention is further illustrated and described below.

[0086] like Figure 1 As shown, a data processing method for an L4 level autonomous driving vehicle, the method comprising:

[0087] U1. During the driving process of the vehicle, the vehicle's motion state data information is collected, and deserialized and cleaned to obtain the processed vehicle's motion state data information;

[0088] U2. Input the processed vehicle motion state data information into the trained multivariate nonlinear regression model based on contribution verification, characterize the correlation between the vehicle motion state and the running quality, and obtain the correlation data information between the vehicle motion state and the running quality;

[0089] U3. Based on the correlation data information between the motion state and the running quality of the vehicle, a neural network algorithm based on an adaptive feedback adjustment factor is used to characterize the running quality of the vehicle to obtain the running quality data information of the vehicle;

[0090] U4. Based on the running quality data information of the vehicle, construct a running quality evaluation function F of the vehicle, evaluate the running quality of the vehicle, and obtain the running quality evaluation data information of the vehicle.

[0091] In this embodiment, if Figure 3 As shown, in step U3, the use of a neural network algorithm based on an adaptive feedback adjustment factor to characterize the vehicle running quality includes:

[0092] U31. Based on the correlation data information of the motion state and running quality of the vehicle, construct a data set of correlation between the motion state and running quality of the vehicle;

[0093] U32. Input the data set of correlation between the vehicle's motion state and running quality into a neural network model based on an adaptive feedback adjustment factor for training and learning, and determine the neuron function S,

[0094]

[0095]

[0096] Among them, y is the correlation data set between the vehicle's motion state and running quality, λ1, λ2 and λ3 are the adaptive feedback adjustment factors of the neural network model, and the trained neural network model based on the adaptive feedback adjustment factor is obtained;

[0097] U33. Based on the trained neural network model based on the adaptive feedback adjustment factor, the correlation data information between the vehicle's motion state and operation quality is input to characterize the vehicle's operation quality and obtain the vehicle's operation quality data information.

[0098] In this embodiment, the vehicle running quality evaluation function F is:

[0099]

[0100] Among them, z is the running quality data information of the vehicle, and μ1, μ2 and μ3 are the evaluation factors of the running quality of the vehicle.

[0101] In this embodiment, the present invention also provides a data processing system for an L4 level autonomous driving vehicle, comprising a computer device, which is programmed or configured to execute the steps of any one of the methods for processing data for an L4 level autonomous driving vehicle.

[0102] In this embodiment, if Figure 4 As shown, pts is the vehicle dispatching platform, vms is the vehicle fault diagnosis platform, and dms is the autonomous driving data processing platform. For rosbag package slicing, the collected problem rosbag packages can be played through webviz to view their vehicle running trajectories and related parameters. During playback, you can click on the time node to mark and generate relevant label information. At the same time, according to the selected time node, the rosbag package fragment is intercepted, and the relevant script is used to generate a scene set to reproduce the problem. Scene set problem reproduction and repair, for the generated scene set, you can use perception, positioning, high-precision map, planning, control and other scripts to repair it. Each type of script will have multiple versions, and these versions will be continuously updated. You can also upload the local script version through the platform for temporary simulation testing. According to the previous label information, select different types of scripts to simulate the scene set collection of the same vehicle type, and efficiently batch process the scene set collection to improve the efficiency of simulation testing.

[0103] In this embodiment, the present invention provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the data processing methods for an L4 level autonomous driving vehicle.

[0104] Any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM (RDRAM).

[0105] In summary, the present invention can not only process vehicle data accurately, but also the communication between the autonomous driving vehicle and the platform and the cloud uploading of data packets do not need to rely on a dedicated network. The functions can be realized through the public network, which can reduce project construction costs.

[0106] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A data processing method for an L4 level autonomous driving vehicle, characterized in that: The method comprises: U1. During the driving process of the vehicle, the vehicle's motion state data information is collected, and deserialized and cleaned to obtain the processed vehicle's motion state data information; U2. Input the processed vehicle motion state data information into the trained multivariate nonlinear regression model based on contribution verification, characterize the correlation between the vehicle motion state and the running quality, and obtain the correlation data information between the vehicle motion state and the running quality; U3. Based on the correlation data information between the motion state and the running quality of the vehicle, a neural network algorithm based on an adaptive feedback adjustment factor is used to characterize the running quality of the vehicle to obtain the running quality data information of the vehicle; U4. Based on the running quality data information of the vehicle, construct a running quality evaluation function F of the vehicle, evaluate the running quality of the vehicle, and obtain the running quality evaluation data information of the vehicle.

2. The data processing method for an L4 level autonomous driving vehicle according to claim 1, characterized in that: The method further comprises: U5. Based on the operation quality assessment data information of the vehicle, a preset threshold is set. If the operation quality assessment data of the vehicle is less than the preset threshold, the operation quality of the vehicle does not meet the requirements and the operation status of the vehicle needs to be adjusted. If the operation quality assessment data of the vehicle is greater than the preset threshold, the operation quality of the vehicle meets the requirements and drives normally.

3. The data processing method for an L4 level autonomous driving vehicle according to claim 1, characterized in that: In step U2, the trained multivariate nonlinear regression model based on contribution verification includes: U21. Based on the processed vehicle motion state data information, construct a vehicle motion state data set, and establish a contribution function Q of the vehicle motion state data set i , Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, x j is the motion state data information of the vehicle at the jth moment after processing, m is the sample size, and the contribution of the vehicle's motion state data set is calculated to obtain the contribution data information of the vehicle's motion state data set; U22. Based on the contribution data information of the vehicle's motion state data set, the vehicle's motion state data set is arranged in ascending order of contribution, and the vehicle's motion state data set is divided into a training data set Train and a verification data set Val, Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, x k is the motion state data information of the vehicle at the kth moment after processing, m is the sample size, k is the cutoff point, n is a constant parameter, and the training data set of the motion state of the vehicle and the verification data set of the motion state of the vehicle are obtained; U23. Based on the training data set of the vehicle's motion state, establish a first characterization function W1 of the vehicle's running quality, Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, n is a constant parameter, r is the weight vector of the vehicle's motion state, α is the first characterization parameter of the vehicle's running quality, g is the high-dimensional feature space function, b is the bias constant parameter of the model, the first characterization parameter of the vehicle's running quality is optimized to obtain the optimized first characterization parameter data information of the vehicle's running quality, based on the verification data set of the vehicle's motion state, a second characterization function W2 of the vehicle's running quality is established, Among them, x k is the vehicle motion state data information at the kth moment after processing, k is the demarcation point, n is the constant parameter, m is the sample size, δ is the second characterization parameter of the vehicle running quality, b is the bias constant parameter of the model, β k To optimize the factor, the second characterization parameter of the vehicle running quality is optimized to obtain optimized data information of the second characterization parameter of the vehicle running quality; U24. Based on the second parameter data information representing the optimized vehicle running quality and the first parameter data information representing the optimized vehicle running quality, a multivariate nonlinear regression function R for contribution verification is established. Where m is the sample size, x i is the motion state data information of the vehicle at the i-th moment after processing, γ i is the relationship determining factor of the vehicle running quality, α0 is the first characterization parameter data information of the optimized vehicle running quality, δ0 is the second characterization parameter data information of the optimized vehicle running quality, b is the bias constant parameter of the model, f is the multivariate relationship vector regression function, which characterizes the correlation between the vehicle's motion state and running quality, and obtains the correlation data information between the vehicle's motion state and running quality.

4. The data processing method for an L4 level autonomous driving vehicle according to claim 3, characterized in that: The high-dimensional feature space function g is, Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, η i is the weight vector of the high-dimensional feature space.

5. The data processing method for an L4 level autonomous driving vehicle according to claim 3, characterized in that: The multivariate relationship vector regression function f is, Among them, x i is the motion state data information of the vehicle at the i-th moment after processing, α0 is the first characterization parameter data information of the optimized vehicle running quality, and δ0 is the second characterization parameter data information of the optimized vehicle running quality.

6. The data processing method for an L4 level autonomous driving vehicle according to claim 3, characterized in that: The optimization factor β k The constraints are:

7. The data processing method for an L4 level autonomous driving vehicle according to claim 1, characterized in that: In step U3, the characterization of the vehicle running quality using a neural network algorithm based on an adaptive feedback adjustment factor includes: U31. Based on the correlation data information of the motion state and running quality of the vehicle, construct a data set of correlation between the motion state and running quality of the vehicle; U32. Input the data set of correlation between the vehicle's motion state and running quality into a neural network model based on an adaptive feedback adjustment factor for training and learning, and determine the neuron function S, Among them, y is the correlation data set between the vehicle's motion state and running quality, λ1, λ2 and λ3 are the adaptive feedback adjustment factors of the neural network model, and the trained neural network model based on the adaptive feedback adjustment factor is obtained; U33. Based on the trained neural network model based on the adaptive feedback adjustment factor, the correlation data information between the vehicle's motion state and operation quality is input to characterize the vehicle's operation quality and obtain the vehicle's operation quality data information.

8. The data processing method for an L4 level autonomous driving vehicle according to claim 1, characterized in that: The vehicle's running quality evaluation function F is: Among them, z is the running quality data information of the vehicle, and μ1, μ2 and μ3 are the evaluation factors of the running quality of the vehicle.

9. A data processing system for an L4 level autonomous driving vehicle, comprising a computer device, characterized in that: The computer device is programmed or configured to execute the steps of the data processing method for an L4 level autonomous driving vehicle as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that is programmed or configured to execute the data processing method for an L4 level autonomous driving vehicle as described in any one of claims 1 to 8.

Images