Multi-robot Path Planning using Modified Particle Swarm Optimization

Multi-robot path planning in a shared environment is a critical research area in robotics. The objective is to find a collision-free path for each robot to reach its destination while optimizing various performance metrics. This paper proposes a novel multi-robot path planning algorithm based on Modified Particle Swarm Optimization (MPSO). The proposed MPSO algorithm introduces a new path planning scheme for determining the robot’s waypoints. Unlike the standard PSO algorithm, which initializes the particle swarm at the robot’s starting position and iteratively determines each waypoint until a completed path is generated, the MPSO algorithm initializes the particle swarm within a predefined search space and searches for the global best position within it to determine a specific robot waypoint through iteration updates. Moreover, a combination of global and local path planning methods is introduced to cope with a dynamic environment. The PSO algorithm functions as a global path planner, determining the complete path for each robot, whereas an obstacle avoidance algorithm serves as a local planner to avoid collision with dynamic obstacles during navigation. This study compares the performance of MPSO with normal PSO using MATLAB simulations in two scenarios. The results demonstrate MPSO’s superiority in terms of total path length and execution time.