Dynamic Obstacle Avoidance in a Complex Environment using Meta-heuristic Algorithms for 4-wheel Robot

Dynamic obstacle avoidance is one of the most important tasks for autonomous navigation of robots. Accordingly, several algorithms have been proposed to determine the velocity at which the robot can drive while avoiding obstacles. Meta-heuristic algorithms are a class of optimization methods that are inspired by natural processes and social behavior. This paper uses meta-heuristic algorithms such as Genetic Algorithms (GA) and Particle Swarm Optimization (PSO) to determine the optimal velocity. The algorithm contains formulas and methods that help control the 4-wheel robot. Also, this work are experimented in a dynamic and complex virtual environment. The experimental environment is set up as dynamic 45 obstacles. The experiments are consists of about two methods and six parameter sets for considering various situation. When implementing to experiment about obstacle avoidance with proposed methods, collision counts and travel times are decreased 0.74 times and about 9.88 on average with PSO.