CN-122018930-A - Driving behavior model self-adaptive updating method based on cloud platform

CN122018930ACN 122018930 ACN122018930 ACN 122018930ACN-122018930-A

Abstract

The application relates to a cloud platform-based driving behavior model self-adaptive updating method which is applied to vehicle equipment and comprises the steps of obtaining parameter information of each sensor in at least one sensor, sending the parameter information of each sensor in the at least one sensor to a cloud platform, receiving a download link of a target preset driving model returned from the cloud platform, wherein the target preset driving model is determined by the cloud platform based on the parameter information of each sensor in the at least one sensor, downloading the target preset driving model based on the download link, and updating a local driving model of the vehicle equipment according to the target preset driving model. It can be seen that the driving model on the vehicle device is determined based on the parameter information of the sensors carried by the vehicle device itself. The mechanism can ensure that the data acquisition precision of the driving model and the data acquisition precision of the vehicle equipment sensor are matched, so that occurrence of false event triggering caused by difficulty in adapting the driving model to the precision difference of different sensors is effectively avoided, and further, the driving experience of a user is improved.

Inventors

ZHOU ZHIWEN
LIU LIAO
LI LAN

Assignees

深圳市麦谷科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251205

Claims (10)

1. A method for adaptively updating a driving behavior model based on a cloud platform, wherein the method is applied to a vehicle device, the vehicle device comprising at least one sensor, the method comprising: Acquiring parameter information of each sensor in the at least one sensor; sending parameter information of each sensor in the at least one sensor to a cloud platform; Receiving a download link of a target preset driving model returned from the cloud platform, wherein the target preset driving model is determined by the cloud platform based on parameter information of each sensor in the at least one sensor; Downloading the target preset driving model based on the downloading link; and updating a local driving model of the vehicle equipment according to the target preset driving model.
2. The method according to claim 1, wherein the method further comprises: Reporting historical driving data to the cloud platform, so that the cloud platform determines a correction value of a target parameter in the target preset driving model based on the historical driving data; receiving a corrected value of the target parameter returned from the cloud platform; And adjusting the target parameters of the target preset driving model based on the corrected values of the target parameters.
3. The method of claim 2, wherein the historical driving data includes vehicle state data, driving behavior data, and context awareness data, and the target parameters include a jerk threshold and a collision threshold.
4. A driving behavior model self-adaptive updating method based on a cloud platform, which is characterized in that the method is applied to the cloud platform and comprises the following steps: receiving parameter information from each of at least one sensor of the vehicle device; extracting features from the parameter information of each sensor of the at least one sensor to obtain a plurality of feature information; Screening a target preset driving model from a plurality of preset driving models according to the plurality of characteristic information; And returning a download link of the target preset driving model to the vehicle equipment so that the vehicle equipment downloads the target preset driving model based on the download link.
5. The method of claim 4, wherein each of the plurality of preset driving models corresponds to a plurality of tag information, the plurality of tag information corresponds to the plurality of feature information one-to-one, and the selecting the target preset driving model from the plurality of preset driving models according to the plurality of feature information comprises: Screening a preset driving model meeting a first condition from the plurality of preset driving models, and taking the preset driving model as the target preset driving model, wherein the first condition comprises that each piece of label information in the plurality of pieces of label information corresponding to the preset driving model is matched with the characteristic information corresponding to the label information.
6. The method according to claim 4 or 5, characterized in that the method further comprises: receiving historical driving data reported by the vehicle equipment; Determining a correction value of a target parameter in the target preset driving model based on the historical driving data; And returning the corrected value of the target parameter to the vehicle equipment so that the vehicle equipment adjusts the target parameter of the target preset driving model based on the corrected value of the target parameter.
7. The method of claim 6, wherein the historical driving data includes vehicle state data, driving behavior data, and context awareness data, and wherein determining a correction value for a target parameter in the target preset driving model based on the historical driving data comprises: Extracting acceleration response features, pitch features and roll features from the vehicle state data; Extracting driving behavior characteristics from the driving behavior data; extracting road condition characteristics from the environment perception data; And determining a correction value of the target parameter according to the acceleration response characteristic, the bump characteristic, the roll characteristic, the driving behavior characteristic and the road condition characteristic.
8. The method of claim 7, wherein said determining a correction value of said target parameter based on said acceleration response characteristic, said pitch characteristic, said roll characteristic, said driving behavior characteristic, and said road condition characteristic comprises: determining a model feature of the vehicle device from the acceleration response feature, the pitch feature and the roll feature; And determining a correction value of the target parameter according to the vehicle model characteristics, the driving behavior characteristics and the road condition characteristics.
9. The method of claim 8, wherein the driving behavior feature comprises a rapid deceleration frequency and a rapid deceleration average intensity, the target parameter comprises a rapid deceleration threshold, and the determining the correction value of the target parameter based on the model feature, the driving behavior feature, and the road condition feature comprises: And determining a correction value of the rapid deceleration threshold according to the vehicle model characteristics, the rapid deceleration frequency, the rapid deceleration average intensity and the road condition characteristics.
10. The method of claim 9, wherein the driving behavior feature further comprises a collision frequency and a collision average intensity, the target parameter further comprises a collision threshold, and the determining the correction value of the target parameter based on the model feature, the driving behavior feature, and the road condition feature comprises: and determining a correction value of the collision threshold according to the vehicle model characteristics, the collision frequency, the collision average intensity and the road condition characteristics.

Description

Driving behavior model self-adaptive updating method based on cloud platform Technical Field The application relates to the field of intelligent automobiles, in particular to a driving behavior model self-adaptive updating method based on a cloud platform. Background With the rapid development of vehicle network technology, intelligent driving has been widely applied to the field of intelligent automobiles. While the core support for achieving intelligent driving is a driving model. The driving model not only can assist the user in driving and reduce the operation difficulty, but also can remind the user to avoid risks in real time in driving and reduce the occurrence of accidents. At present, the traditional intelligent driving mostly adopts a 'deployment after training' mode. Specifically, the cloud platform uniformly trains the driving model based on massive user data, and deploys the trained driving model to each vehicle device for calling. However, the sensors of the same type of different vehicle devices often have significant individual differences, so that the accuracy of collected data is inconsistent, so that a driving model is difficult to make accurate judgment, further, false triggering of an event is caused, and the driving experience of a user is seriously affected. Disclosure of Invention The embodiment of the application provides a cloud platform-based driving behavior model self-adaptive updating method, which aims to solve the technical problem that a driving model is difficult to adapt to the precision difference of different sensors to cause false triggering of an event. In a first aspect, an embodiment of the present application provides a driving behavior model adaptive updating method based on a cloud platform, where the method is applied to a vehicle device, and the vehicle device includes at least one sensor, which includes: acquiring parameter information and sampling data of each sensor in the at least one sensor; Extracting statistical features from the sampled data of each of the at least one sensor; sending parameter information and statistical characteristics of each sensor in the at least one sensor to a cloud platform; receiving a download link of a target preset driving model returned from the cloud platform, wherein the target preset driving model is determined by the cloud platform based on parameter information and statistical characteristics of each sensor in the at least one sensor; Downloading the target preset driving model based on the downloading link; and updating a local driving model of the vehicle equipment according to the target preset driving model. Optionally, the method further comprises: Reporting historical driving data to the cloud platform, so that the cloud platform determines a correction value of a target parameter in the target preset driving model based on the historical driving data; receiving a corrected value of the target parameter returned from the cloud platform; And adjusting the target parameters of the target preset driving model based on the corrected values of the target parameters. Optionally, the historical driving data comprises vehicle state data, driving behavior data and environment perception data, and the target parameters comprise a sudden deceleration threshold value and a collision threshold value. In a second aspect, an embodiment of the present application provides a driving behavior model adaptive updating method based on a cloud platform, which includes: the method is applied to a cloud platform, and comprises the following steps: receiving parameter information and statistical features from each of at least one sensor of the vehicle device; extracting features from the parameter information and the statistical features of each sensor of the at least one sensor to obtain a plurality of feature information; Screening a target preset driving model from a plurality of preset driving models according to the plurality of characteristic information; And returning a download link of the target preset driving model to the vehicle equipment so that the vehicle equipment downloads the target preset driving model based on the download link. Optionally, each preset driving model in the plurality of preset driving models corresponds to a plurality of tag information, the plurality of tag information corresponds to the plurality of feature information one to one, and the method for screening the target preset driving model from the plurality of preset driving models according to the plurality of feature information includes: Screening a preset driving model meeting a first condition from the plurality of preset driving models, and taking the preset driving model as the target preset driving model, wherein the first condition comprises that each piece of label information in the plurality of pieces of label information corresponding to the preset driving model is matched with the characteristic information corresponding to the label in