Applying Optimization Algorithms for Robot Path Planning

Abstract

This thesis presents a hybrid optimization algorithm combining Genetic Algorithm and Particle Swarm Optimization to improve robot path planning in static and dynamic environments. The algorithm was tested in ROS 2 and Gazebo simulations using a two-wheeled robot with a LiDAR sensor. Results showed that the hybrid approach outperformed individual GA and PSO by generating shorter paths, better obstacle avoidance, and smoother motion. While effective, future work could enhance the algorithm by incorporating machine learning for real-time decision-making and obstacle prediction in complex environments.