Multi-model target state prediction method and system based on deep reinforcement learning

Abstract

The invention discloses a multi-model target state prediction method based on a deep neural network and reinforcement learning, effectively solves the problems of complex strong nonlinear environmentexpression of multi-model fusion and long-time prediction of a target state, facilitates the improvement of the prediction precision and prediction duration of the target state, and achieves the online parameter self-correction of multi-model fusion. The calibration method provided by the invention has the advantages: 1) adapting to environmental scene changes and self-adjusting multi-model targetstate prediction parameters; 2) improving the target state prediction precision; 3) improving the target state prediction time length; 4) performing online self-learning the multi-model prediction parameters and improving the target state prediction precision in a dynamic scene.

Description

technical field [0001] The present invention relates to an intelligent driving system-oriented target state prediction method and system, in particular to a multi-model target state prediction method and system based on deep reinforcement learning. Background technique [0002] Intelligent driving vehicles have a positive effect on traffic safety, traffic efficiency, environmental protection and energy saving. Intelligent driving vehicles use the perception system to sense the parameters of the driving environment and identify the type of target; through the cognitive system, the understanding of the driving environment, such as the understanding of driving behavior intentions, is improved, and future environmental changes are estimated and predicted, and other road users such as The decision-making mechanism of vehicles and pedestrians makes a correct understanding of the environment; completes driving behavior and path planning through the decision-making system and execut...

AI Technical Summary

Problems solved by technology

However, prediction methods based on behavioral cognition have large prediction errors in a short period of time

[0004] At this stage, for intelligent driving technology, the target trajectory prediction has the following problems: 1) the prediction accuracy is low, and it cannot cope with the prediction accuracy in complex scenes; 2) the predictable time is short; 3) it cannot adapt to the respective scene for self-learning Fusion of Multiple Forecasting Methods

However, no related work has been carried out

Images