Environmental Adaptive Urban Traffic Signal Control Based on Reinforcement Learning Algorithm

Yanming Feng,Yongrong Wu

DOI: https://doi.org/10.1088/1742-6596/1650/3/032097

2020-10-01

Journal of Physics: Conference Series

Abstract:Abstract Urban traffic signal control is an important part of the construction of intelligent regional traffic. Aiming at the problem of the optimal control strategy in urban traffic signals, this paper proposes environmental adaptive urban traffic signal control based on reinforcement learning algorithms. Through the continuous perception of the traffic environment, the position and speed of the vehicles in different environments are expressed in a matrix, and the parameters are continuously iteratively optimized through the reinforcement learning method to optimize the objective function (the vehicle that can pass the most in a limited time) to achieve the purpose of effective vehicle control. According to the traffic simulation software Vissim, it can be known that the algorithm proposed in this paper performs better in terms of average waiting queue length and global average speed compared with other algorithms, and the deep learning algorithm is significantly better than other algorithms in terms of stability. The average speed of the deep learning algorithm is increased by 9% compared with the baseline, and the average waiting queue length is reduced by 13.4% compared with the baseline. The experiments studied this time are sufficient to prove that the algorithm in this paper can adapt to the dynamically changing and complex urban traffic environment and has great research value.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the optimal control strategy for urban traffic signal lights. With the continuous increase of urban traffic pressure, the traditional single - point signal control method has been difficult to meet the requirements of efficient traffic management. Therefore, this paper proposes an urban traffic signal control method based on a reinforcement - learning algorithm with environmental adaptability. By continuously sensing traffic conditions and mining hidden patterns, this method aims to find the optimal control strategy to improve traffic efficiency. Specifically, this research focuses on how to use the reinforcement - learning algorithm to dynamically adjust the control strategy of traffic signal lights according to the real - time changes in the traffic environment (such as the position and speed of vehicles), thereby maximizing traffic flow and reducing traffic congestion. The experimental results show that, compared with traditional control methods, this method performs better in terms of the average waiting queue length and the global average speed, especially in high - traffic environments, and can significantly improve the stability and efficiency of the traffic system.

Environmental Adaptive Urban Traffic Signal Control Based on Reinforcement Learning Algorithm

A Deep Reinforcement Learning Approach to Traffic Signal Control With Temporal Traffic Pattern Mining

Evaluation and Application of Urban Traffic Signal Optimizing Control Strategy Based on Reinforcement Learning

Deep Reinforcement Learning for Traffic Signal Control Model and Adaptation Study

Traffic Signal Timing via Parallel Reinforcement Learning

A Traffic Signal Control Method Based on Asynchronous Reinforcement Learning

A Stochastic Adaptive Traffic Signal Control Model Based on Fuzzy Reinforcement Learning

Adaptive urban traffic signal control based on enhanced deep reinforcement learning

Adaptive Traffic Signal Control with Deep Reinforcement Learning An Exploratory Investigation

A Method of Reinforcement Learning Based Automatic Traffic Signal Control

Adaptive Traffic Signal Control: Deep Reinforcement Learning Algorithm with Experience Replay and Target Network

Low Carbon City Traffic Signal Control Based on Deep Reinforcement Learning

Adaptive Traffic Signal Control Model on Intersections Based on Deep Reinforcement Learning

Traffic Signal Control in Mixed Traffic Environment Based on Advance Decision and Reinforcement Learning

Researches on Intelligent Traffic Signal Control Based on Deep Reinforcement Learning

An Intelligent Traffic Signal Control System Based on Deep Reinforcement Learning

Optimization Control of Adaptive Traffic Signal with Deep Reinforcement Learning

A Deep Adaptive Traffic Signal Controller with Long-Term Planning Horizon and Spatial-Temporal State Definition under Dynamic Traffic Fluctuations

Optimizing Traffic Signal Control in Mixed Traffic Scenarios: A Predictive Traffic Information-Based Deep Reinforcement Learning Approach

Adaptive Traffic Signal Control Method Based on Offline Reinforcement Learning

Adaptive Traffic Signal Control With Deep Reinforcement Learning and High Dimensional Sensory Inputs: Case Study and Comprehensive Sensitivity Analyses