A traffic decision-making method for complex scenes at intersections of distributed intelligent networked vehicles

Abstract

The invention puts forward a distributed intelligent connected vehicle intersection complex scene passage decision method, and belongs to the technical field of intelligent traffic. Compared with a no-signal-lamp intersection decision method based on rules or reinforcement learning, the method disclosed by the invention introduces a vehicle traffic rule violation detection mechanism to realize thefusion of rules and a reinforcement learning method. Firstly, an implicit curve-family vehicle traffic rule violation detection algorithm is adopted to judge whether a vehicle in the intersection violates a traffic rule or not; when all vehicles obey the traffic rule, a rule method is adopted to make an intersection vehicle passage decision; and otherwise, a method based on the reinforcement learning is adopted to make the intersection vehicle passage decision. By use of the method, the independent passage decision of the no-signal-lamp intersection intelligent connected vehicle under a situation that a complex traffic scene with mixed traffic rules is obeyed and violated is realized, and technical support is provided for improving traffic safety and intersection passage efficiency underthe complex environment.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and in particular relates to a decision-making method for passing a complex scene at a distributed intelligent networked vehicle intersection. Background technique [0002] Relevant reports show that in 2016, the number of casualties caused by road traffic accidents in my country ranks among the top in the world, and the economic losses directly caused by them are as high as billions. Literature (Ren Shuhang. TCT-Based Research on Safety Evaluation of Urban Road Intersections [D]. Chang'an University, 2016.) recorded a survey of traffic accident data in a representative city in my country. The survey results show that urban intersections are the number one traffic accident. A high traffic section. This paper conducts field research on 15 intersections in a certain city, and the results show that the violation of traffic rules by pedestrians and car drivers is the root cause of ...

AI Technical Summary

Problems solved by technology

At present, the detection method of violating traffic rules is mainly used to provide auxiliary reference for the traffic management bureau to identify violations of traffic rules, but it is not directly applied to the autonomous decision-making of self-driving vehicles at intersections.

Images