This repository contains a Q-learning-based AI model developed for optimizing traffic flow and managing autonomous vehicle routes. It was created as part of an assignment for the course CP468 - Artificial Intelligence at Wilfrid Laurier University.
The model leverages a reinforcement learning algorithm, Q-learning, combined with a heuristic method named BNART (Best Neighbor Algorithm for Routing Traffic), designed to minimize travel time and congestion in a grid-based simulation of traffic control for autonomous vehicles.
The approach is informed by the methodologies described in the paper "BNART: A Novel Centralized Traffic Management Approach for Autonomous Vehicles" which can be found here.
- Implementation of Q-learning algorithm for path optimization.
- Integration of BNART heuristic to enhance route selection.
- Grid-based simulation environment for traffic scenarios.
- Performance metrics to evaluate efficiency and effectiveness.
- This project was developed as part of a course requirement for CP468 - Artificial Intelligence at Wilfrid Laurier University.
- Special thanks to the course instructors and teaching assistants for their guidance and support.
- Appreciation to the authors of the paper "BNART: A Novel Centralized Traffic Management Approach for Autonomous Vehicles" for their insights into traffic management strategies.