Decision-making method, device, device and readable storage medium for automatic driving equipment

Abstract

The invention relates to the technical field of automatic driving, and discloses a decision-making method, device, equipment and readable storage medium of automatic driving equipment. Wherein, the method includes: obtaining the local environmental state information corresponding to each automatic driving device, the driving action of each automatic driving device, the return parameter value of the driving action, and locally updating the environmental state information; based on the local environmental state corresponding to each automatic driving device information and driving actions to determine the learning action value of the vehicle group corresponding to each automatic driving device; based on the return parameter value of each automatic driving device after performing the driving action and the local update environment state information, determine the target action value of the vehicle group; based on the learning action value and the value of the target action to update the driving decision-making strategy of the car group. By implementing the present invention, the update of the driving decision-making strategy of the vehicle group is realized, the problem of explosion in the training dimension is avoided, and the safety and efficiency of the driving decision-making strategy are improved.

Description

technical field [0001] The present invention relates to the technical field of automatic driving, in particular to a decision-making method, device, device and readable storage medium of automatic driving equipment. Background technique [0002] With the rapid development of artificial intelligence technology, deep learning has been widely used in task scenarios such as environment perception, feature extraction, obstacle recognition, path planning and intelligent decision-making of self-driving cars, and has been used in simulation simulators and simple real traffic scenarios. Better to achieve. Among them, in the technical route represented by deep reinforcement learning, self-driving vehicles, as agents, do not need the massive environmental state-driving action data provided by human drivers to train driving strategies, but use the Markov decision-making process to continuously interact with the traffic environment , to explore and improve autonomous driving policies fr...

AI Technical Summary

Problems solved by technology

[0003] In view of this, the embodiments of the present invention provide a decision-making method, device, device and readable storage medium for automatic driving equipment, so as to solve the problem that it is difficult to balance the safety of automatic driving vehicle group decision-making and the cost of driving strategy training at the same time

Images