# Fuzzy Q-Learning Based Multi-Agent System for Intelligent Traffic Control by a Game Theory Approach

## TL;DR

This paper presents a multi-agent fuzzy Q-learning approach combined with game theory to optimize traffic light control, reducing delays in dynamic traffic networks through adaptive learning and neighbor decision considerations.

## Contribution

It introduces a novel multi-agent system integrating fuzzy Q-learning and game theory for adaptive traffic signal control in non-stationary environments.

## Key findings

- Proposed method outperforms fixed time and other control methods.
- Fuzzy Q-learning with game theory improves traffic delay reduction.
- Simulation results validate the effectiveness of the approach.

## Abstract

This paper introduces a multi-agent approach to adjust traffic lights based on traffic situation in order to reduce average delay time. In the traffic model, lights of each intersection are controlled by an autonomous agent. Since decision of each agent affects neighbor agents, this approach creates a classical non-stationary environment. Thus, each agent not only needs to learn from the past experience but also has to consider decision of neighbors to overcome dynamic changes of the traffic network. Fuzzy Q-learning and Game theory are employed to make policy based on previous experiences and decision of neighbor agents. Simulation results illustrate the advantage of the proposed method over fixed time, fuzzy, Q-learning and fuzzy Q-learning control methods.