CN-121984969-A - Vehicle zero offset processing method and cloud server

CN121984969ACN 121984969 ACN121984969 ACN 121984969ACN-121984969-A

Abstract

The application provides a vehicle zero offset processing method and a cloud server, and relates to the technical field of data processing, wherein the processing method comprises the steps of receiving track data of each of a plurality of first running tracks sent by a target vehicle in a first processing period; the method comprises the steps of determining a plurality of second running tracks from a plurality of first running tracks based on track data of the first running tracks, wherein the second running tracks are straight running tracks, performing zero offset estimation processing based on track data of each second running track to obtain estimation processing results corresponding to each second running track, wherein the estimation processing results corresponding to each second running track comprise a first zero offset value and confidence coefficient of the first zero offset value, and performing risk analysis on the estimation processing results corresponding to the second running tracks through an agent to obtain risk analysis results of a target vehicle. The processing method can realize low cost, high efficiency and reliability of vehicle zero offset processing.

Inventors

LING JIAJIA
GUO XIAOLIN
LI SHIYIN

Assignees

驭势（上海）汽车科技有限公司

Dates

Publication Date: 20260505
Application Date: 20251231

Claims (10)

1. The vehicle zero offset processing method is characterized by being applied to a cloud server, wherein the cloud server comprises an agent, and the method comprises the following steps: receiving track data of each of a plurality of first running tracks sent by a target vehicle in a first processing period, wherein the track data comprises a front wheel deflection angle sequence; determining a plurality of second travel tracks from the plurality of first travel tracks based on track data of each of the plurality of first travel tracks, wherein the second travel tracks are straight travel tracks; performing zero offset estimation processing based on track data of each second running track to obtain an estimation processing result corresponding to each second running track, wherein the estimation processing result corresponding to each second running track comprises a first zero offset value and a confidence coefficient of the first zero offset value; And carrying out risk analysis on the estimation processing results corresponding to the second running tracks by the intelligent agent to obtain a risk analysis result of the target vehicle.
2. The method of claim 1, wherein the trajectory data further comprises a sequence of travel positions and a sequence of heading angles; the determining a plurality of second travel tracks from the plurality of first travel tracks based on track data of each of the plurality of first travel tracks includes: determining a plurality of to-be-selected travel tracks belonging to a target road network area from the plurality of first travel tracks based on the position information of the target road network area and the respective travel position sequences of the plurality of first travel tracks; Determining statistical data corresponding to each to-be-selected running track based on the running position sequence and the course angle sequence of each to-be-selected running track, wherein the statistical data corresponding to each to-be-selected running track comprises track mileage, course angle amplitude and transverse banners; and determining a to-be-selected running track meeting a linear track screening condition in the plurality of to-be-selected running tracks as the second running track, wherein the linear track screening condition comprises at least one of the following steps that the track mileage of the to-be-selected running track is larger than a preset mileage threshold, the course angle amplitude of the to-be-selected running track is smaller than a preset course angle threshold, and the transverse amplitude of the to-be-selected running track is smaller than a preset transverse threshold.
3. The method of claim 1, wherein the performing zero offset estimation processing based on the track data of each second running track to obtain an estimation processing result corresponding to each second running track includes: for each second travel track, performing the following steps: Determining an average value of front wheel deflection angles in a front wheel deflection angle sequence of a current second running track, and obtaining a first zero offset value corresponding to the current second running track; and determining the standard deviation of the front wheel deflection angle in the front wheel deflection angle sequence of the current second running track to obtain the confidence coefficient of the first zero offset value corresponding to the current second running track.
4. The method according to any one of claim 1, wherein the risk analysis is performed on the estimation processing results corresponding to each of the plurality of second driving trajectories by an agent to obtain a risk analysis result of the target vehicle, including: determining an average value of the first zero offset values corresponding to the second running tracks to obtain a second zero offset value of the target vehicle in the first processing period; determining the average value of the confidence coefficient of the first zero offset values corresponding to the second running tracks to obtain a variation coefficient of the target vehicle in the first processing period; determining a second running track with a first zero offset value larger than a first offset threshold value in the plurality of second running tracks as an abnormal track, and dividing the number of the abnormal tracks by the total number of the plurality of second running tracks to obtain a track abnormality rate of the target vehicle in the first processing period; Comparing and analyzing the first zero offset values corresponding to the second running tracks to obtain the change trend of the first zero offset values of the target vehicle in the first processing period; And performing risk analysis on the second zero offset value, the variation coefficient, the track abnormality rate and the variation trend of the first zero offset value of the target vehicle in the first processing period by using an intelligent agent to obtain a risk analysis result of the target vehicle.
5. The method according to claim 4, wherein the method further comprises: Acquiring a second zero offset value, a variation coefficient, a track abnormality rate and a variation trend of a first zero offset value of the target vehicle in at least one second processing period through the intelligent agent, wherein the second processing period is earlier in time than the first processing period; The risk analysis is carried out on the second zero offset value, the variation coefficient, the track abnormality rate and the variation trend of the first zero offset value of the target vehicle in the first processing period by the intelligent agent to obtain a risk analysis result of the target vehicle, and the risk analysis result comprises the following steps: And carrying out risk analysis on the second zero offset value, the variation coefficient, the track abnormality rate and the variation trend of the first zero offset value of the target vehicle in the at least one second processing period and the first processing period by the intelligent agent to obtain a risk analysis result of the target vehicle.
6. The method of claim 5, wherein the risk analysis results of the target vehicle include a risk level of the target vehicle, a target anomaly cause for a steering wheel null bias of the target vehicle, and a target treatment of the steering wheel null bias of the target vehicle; The risk analysis is performed on the second zero offset value, the variation coefficient, the track anomaly rate and the variation trend of the first zero offset value of the target vehicle in the at least one second processing period and the first processing period by the intelligent agent, so as to obtain a risk analysis result of the target vehicle, wherein the risk analysis result comprises: Determining, by the agent, a risk level of the target vehicle from a plurality of preset risk levels based on at least one of a trajectory anomaly rate of the target vehicle during the at least one second processing period and the first processing period, an average of second zero offset values of the target vehicle during the at least one second processing period and the first processing period, and a maximum of second zero offset values of the target vehicle during the at least one second processing period and the first processing period; determining, by the agent, a target processing manner corresponding to a risk level of the target vehicle from a correspondence between a plurality of preset processing manners and a plurality of preset risk levels; And determining, by the agent, the target abnormality cause from a plurality of preset abnormality causes based on a plurality of second zero offset values and a plurality of coefficient of variation of the target vehicle in the at least one second processing period and the first processing period.
7. The method of claim 6, wherein the method further comprises: The method comprises the steps of obtaining target prompt words, wherein the target prompt words comprise a judging rule of a plurality of preset abnormal reasons, a judging standard of a plurality of preset risk levels and a judging standard of a plurality of preset processing modes; And inputting the target prompt word into an initial large language model to obtain the intelligent agent.
8. The method according to claim 1, wherein before the risk analysis is performed on the estimation processing results corresponding to each of the plurality of second travel tracks by the agent, the method further comprises: Deleting the second running tracks with the first zero offset value larger than a second offset threshold value from the plurality of second running tracks to obtain filtered second running tracks; the risk analysis is performed on the estimation processing results corresponding to the second running tracks by the intelligent agent to obtain a risk analysis result of the target vehicle, including: And processing and analyzing the estimation processing results corresponding to the filtered second running tracks by the intelligent agent to obtain risk analysis results of the target vehicle.
9. A cloud server, comprising one or more processors, storage means for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the method for vehicle zero offset processing according to any one of claims 1-8.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program, wherein the computer program, when executed by a processor, implements the method of handling zero offset of a vehicle according to any one of claims 1-8.

Description

Vehicle zero offset processing method and cloud server Technical Field The application relates to the technical field of data processing, in particular to a vehicle zero offset processing method and a cloud server. Background In the prior art, the determination of the zero offset of the vehicle and the analysis of related data have obvious limitations. In the zero offset calibration link, the existing zero calibration mode generally depends on a special calibration flow, wherein a wider straight running calibration method is applied, a special calibration site is required to be planned and built in advance, a vehicle is controlled to keep a uniform straight running state in the site, relevant data such as a front wheel deflection angle and the like are recorded, and finally a zero offset value is determined by means of mean value calculation and the like. The calibration mode is extremely strongly constrained by site conditions, has the problems of high cost, complex calibration flow and long time consumption, and is difficult to adapt to the dynamic zero offset updating requirement in the daily running process of the vehicle. In the zero offset analysis link, the existing method mainly uses engineers to screen effective information from mass data generated by vehicle running by virtue of self-accumulated technical experience, judges the rationality of zero offset related parameters and identifies potential zero offset abnormal risks. The analysis mode relying on manual work is low in efficiency, and deviation of analysis results is easy to occur due to the difference of manual subjective judgment, so that the automatic and accurate requirements of vehicles on zero offset analysis cannot be met. In summary, the existing vehicle zero offset processing method has the problems of high cost, low efficiency and incapability of ensuring accuracy. Disclosure of Invention The application provides a vehicle zero offset processing method and a cloud server, which are used for at least solving the problems of high cost, low efficiency and incapability of ensuring accuracy in vehicle zero offset processing in the related technology. The application provides a vehicle zero offset processing method which is applied to a cloud server, wherein the cloud server comprises an agent, the processing method comprises the steps of receiving track data of each of a plurality of first running tracks sent by a target vehicle in a first processing period, wherein the track data comprise a front wheel deflection angle sequence, determining a plurality of second running tracks from the plurality of first running tracks based on the track data of each of the plurality of first running tracks, wherein the second running tracks are straight running tracks, performing zero offset estimation processing based on track data of each of the second running tracks to obtain estimation processing results corresponding to each of the second running tracks, wherein the estimation processing results corresponding to each of the second running tracks comprise a first zero offset value and a confidence coefficient of the first zero offset value, and performing risk analysis on the estimation processing results corresponding to each of the plurality of second running tracks through the agent to obtain risk analysis results of the target vehicle. Optionally, the track data further comprises a sequence of driving positions and a sequence of heading angles. The method comprises the steps of determining a plurality of second running tracks from a plurality of first running tracks based on track data of the first running tracks, determining a plurality of to-be-selected running tracks belonging to a target road network area from the first running tracks based on position information of the target road network area and running position sequences of the first running tracks, determining statistical data corresponding to each to-be-selected running track based on running position sequences and course angle sequences of each to-be-selected running track, determining the statistical data corresponding to each to-be-selected running track to be a second running track, wherein the to-be-selected running track meeting a linear track screening condition in the to-be-selected running tracks comprises at least one of a track mileage of the to-be-selected running track being greater than a preset mileage threshold, a course angle amplitude of the to-be-selected running track being less than a preset course angle threshold and a transverse amplitude of the to-be-selected running track being less than a preset transverse threshold. Optionally, the estimating processing of zero offset is performed based on the track data of each second running track to obtain an estimating processing result corresponding to each second running track, and the method comprises the following steps of determining an average value of front wheel deflection angles in a front wheel deflection ang