A decision-making method for intelligent vehicle traffic at unsignalized intersections based on constraint trees

Abstract

The present invention provides a decision-making method for intelligent vehicles at intersections without signal lights based on constraint trees. Step 1. Receive vehicle information through the vehicle infrastructure interconnection system (V2I). Whenever an intelligent networked vehicle enters the communication area, set The roadside agent near the intersection establishes a connection with the vehicle to receive the motion state information of the vehicle when it enters the communication area, including its position x i (t), speed v i (t); Step 2, determine the traffic sequence of all N vehicles in the communication area; Step 3, determine the objective equation of each vehicle; Step 4: compare the node distribution obtained in step 2 with the objective equation obtained in step 3 Combination, that is, the constraint tree that constitutes the traffic decision result of the unsignalized intersection. The intelligent vehicle communication decision-making method at a non-signal-light intersection based on the constraint tree of the present invention can be adapted to the scene of time-varying traffic flow, avoids complex optimization solution methods, and ensures the high efficiency of the decision-making process.

Description

technical field [0001] The invention relates to a passing method of an intelligent vehicle, in particular to a decision-making method for passing an intelligent vehicle at an intersection without signal lights based on a constraint tree. Background technique [0002] Unsignalized intersections represented by township road interchanges are prone to traffic accidents. The difficulty lies in how to make vehicles cooperate with each other to achieve safe passage. With the development and implementation of intelligent network technology, smart vehicles can use the vehicle-to-vehicle and vehicle infrastructure interconnection system communication technology to provide the possibility to realize multi-vehicle coordination at intersections without signal lights. [0003] Traffic decision-making is the main problem of multi-vehicle coordination at unsignalized intersections. The purpose of traffic decision-making is to obtain the control target for subsequent vehicle control. Existi...

AI Technical Summary

Problems solved by technology

Existing research can be divided into two categories. One is to regard all the vehicles heading to the intersection as the same virtual queue, and the decision result obtained is a unified expected inter-vehicle distance. This kind of method is not suitable for the scene of traffic flow change. For example, when the initial distance between vehicles is too large, setting a uniform expected distance between vehicles will cause unnecessary chasing by the following vehicles; the second is to optimize and adjust the expected arrival time of all vehicles without conflict, so as to avoid the time delay of vehicles occupying conflict areas. Overlapping, such methods rely on centralized optimization, have a large communication burden and computational burden, and are difficult to deploy in actual scenarios

Images