Investigation of road planning appoach in swarm robotics

In this thesis, a comparison was made with three different algorithms for path planning and obstacle avoidance behavior in swarm robotics. A path planning approach has been proposed so that the swarm robots can reach the target location from the starting location in the determined arenas. In addition, algorithms have been developed to enable herd robots to avoid obstacles in the path planning process. In order to evaluate the developed path planning approach in terms of scalability and flexibility and to measure its performance, systematic simulations were applied with VFH, RRT and PRM algorithms in three different obstacle arenas with different robot numbers and different looking distances. With the developed algorithms, each swarm robot exhibits swarm behavior in three different arenas in an organized manner depending on the distance to look forward to the path to be followed to reach the target location. The developed VFH-Pure Pursuit-based collaboration algorithm directs the swarm robots to the direction where there are no obstacles or robots with the VFH method to avoid obstacles and collisions with robots.When obstacles and robots are not encountered, it follows the predetermined path with Pure Pursuit. In the RRT-Pure Pursuit-based collaborative algorithm and the PRM-Pure Pursuit-based collaborative algorithms, an unobstructed path is created in the arenas between the predetermined starting and target location.The path created is intended to be followed with the Pure Pursuit algorithm. Four different criteria used in the literature were used to measure and compare the performance of the path planning and obstacle avoidance approach. As a result of the encounters, it was seen that the PRM-Pure Pursuit-based collaborative obstacle avoidance and path planning approach for swarm robots was more successful than other algorithms.