A traffic organization scheme optimization method based on multi-agent reinforcement learning

Abstract

The invention discloses a traffic organization scheme optimization method based on multi-agent reinforcement learning, improves the Actor network in MADDPG, improves the experience base in the Critic network based on the birth and death process, and uses the maximum traffic flow setting in the morning peak as the return of the intelligent body index, using the trajectory data to train the maximum entropy inverse reinforcement learning model as the reward mechanism of the multi-agent, and designing the reward function of reinforcement learning based on this; the method of the invention realizes the optimization of the current urban traffic organization scheme, Analyze and find out the cause of traffic congestion. This method can adapt well and quickly find the optimal solution, provide traffic police experts with traffic counseling opinions, and lay the foundation for smart cities.

Description

technical field [0001] The invention belongs to the technical field of traffic flow optimization, and in particular relates to a traffic organization scheme optimization method based on multi-agent reinforcement learning. Background technique [0002] With the rapid development of cities, the urban road network has become more and more complex, and the problem of traffic congestion has become more and more serious. The control of traffic flow in urban areas is not single. If the turning of a certain intersection is restricted alone, the traffic flow of the surrounding sections is difficult to control and more prone to congestion. Using multi-agent reinforcement learning to control road traffic flow will also appear inappropriate. For example, in a complex urban road network, multi-agent team cooperation and communication information is insufficient, and integrated learning efficiency is low; it is difficult to design a standard for traffic flow returns. ; The global linkage...

AI Technical Summary

Problems solved by technology

The control of vehicle flow in urban areas is not single. If you limit the turning of a certain intersection separately, the flow of the surrounding road sections will be difficult to control and more likely to be congested.

The use of multi-agent reinforcement learning to control road traffic flow will also appear unsuitable. For example, in a complex urban road network, multi-agent teamwork and communication information is insufficient, and the efficiency of integrated learning is low; it is difficult to design a standard for traffic flow returns. ; Global linkage changes in traffic flow lead to multi-agents generating a lot of experience in the process of learning, but the traditional multi-agent algorithm is relatively simple in processing experience

Images