Method and device for determining driving strategy based on reinforcement learning and rule

Abstract

The purpose of the present application is to provide a method or a device for determining a driving strategy based on reinforcement learning and rule fusion. The method comprises: based on the drivingparameter information of a vehicle, determining first driving strategy information of the vehicle through a reinforcement learning algorithm; based on the driving parameter information and the driving rule information of the vehicle, performing rationality detection on the first driving strategy information; and determining target driving strategy information of the vehicle based on the detectionresult of the rationality detection. Compared with the prior art, according to the technical scheme of the present application, the first driving strategy information calculated and determined by thereinforcement learning algorithm is constrained by the rule, so that the method for determining the driving strategy provided by the present application is more intelligent than the existing method for implementing vehicle control by using the rule algorithm, or the method for implementing vehicle control by using the reinforcement learning algorithm, and the rationality and stability of the finally determined driving strategy are improved.

Description

technical field [0001] This application relates to the field of automatic driving, in particular to a technology for determining driving strategies based on reinforcement learning and rules. Background technique [0002] In the existing vehicle driving process, the vehicle control of the vehicle, especially the automatic driving vehicle, is mainly realized through the following methods: Rule-based automatic driving technology, that is, the vehicle control is realized by using the rule algorithm. According to the logic formula, the state input The output control value is directly obtained from the output control value. This type of algorithm is simple to implement, does not require training, and the output result of the control algorithm is predictable and relatively stable. However, the algorithm is not intelligent, and it is easy to be robbed of the right of way in the complex scene of real driving. , so the algorithm cannot effectively cope with the complex scene of real d...

AI Technical Summary

Problems solved by technology

[0002] In the existing vehicle driving process, the vehicle control of the vehicle, especially the automatic driving vehicle, is mainly realized through the following methods: Rule-based automatic driving technology, that is, the vehicle control is realized by using the rule algorithm. According to the logic formula, the state input The output control value is directly obtained from the output control value. This type of algorithm is simple to implement, does not require training, and the output result of the control algorithm is predictable and relatively stable. However, the algorithm is not intelligent, and it is easy to be robbed of the right of way in the complex scene of real driving. , so the algorithm cannot effectively cope with the complex scene of real driving; the use of reinforcement learning algorithm to realize vehicle control can make the driving strategy more intelligent, but the time cost of reinforcement learning model training is high, which cannot be applied to actual automatic driving, and the algorithm The output results are unpredictable; while the existing algorithms that integrate rules and reinforcement learning can only linearly add the results determined by rule algorithms and reinforcement learning algorithms, the time cost of model training is still high, and continuous trial and error is required , resulting in the inability to be applied to actual automatic driving

Images