A Direction for Swarm Robotic Path Planning Technique Using Potential Field Concepts and Particle Swarm Optimization.

Swarm robotics has emerged as a prominent research field in recent years. Managing a group of small robots with limited capabilities to achieve their goals poses a significant challenge for developers. Path planning, a crucial process for robotic systems, is essential for ensuring that all robots can reach their destinations while avoiding obstacles. However, path planning techniques commonly face a trade-off between speed and path quality. Achieving higher path quality often demands more iterations to find shorter and smoother paths, consequently increasing the processing time. This trade-off becomes even more intricate in multi-robot systems, where an escalation in the number of robots naturally amplifies the number of processes. This can lead to system failures due to the constrained processing capabilities of such systems. This paper presents a novel path planning approach for swarm robotic systems. The proposed system aims to effectively address the trade-off dilemma by integrating two techniques recognized for their swift execution: the potential field method and Particle Swarm Optimization (PSO). The potential field, generated using a unique objective function modeling method, enables robots to navigate toward the goal while avoiding obstacles. The objective function is composed using two terms, the Euclidean and the Gaussian term. The Euclidean term is responsible for finding the shortest path possible to the goal point while the Gaussian term is responsible for assisting the robot to avoid the obstacles. This objective function is later optimized using PSO algorithm. Simulation results involving three follower robots and a leader robot demonstrate the superior performance of the proposed technique compared to the widely used Probabilistic RoadMap (PRM) path planning technique. The proposed approach excels in terms of processing speed and path quality.