Efficient Collision-Free Motion Planning in Distributed Multi-Robot Systems
Om Sharma, Nilotpal Chakraborty · 2025
In distributed multi-robot systems, ensuring collision-free motion planning is a complex challenge, especially in dynamic environments where multiple robots are operating simultaneously. Traditional path-planning algorithms are effective for individual robot navigation but struggle when dealing with the unpredictable movements of other robots in the system. To overcome these limitations, this research focuses on advanced motion planning techniques that ensure safe and efficient robot coordination. We model the overall motion planning problem into a two-dimensional grid and our objective is to minimize both the time and the energy taken by the robots to reach their destinations from their respective sources. We enhance standard path-finding algorithms such as A* search, along with Deep Reinforcement Learning (DRL) with suitable modifications to adapt to real-time changes. DRL allows robots to learn from their environment and make optimal decisions, improving their ability to avoid collisions and navigate efficiently in uncertain conditions. By integrating these algorithms and conducting a comparative analysis, we propose a framework that enables robots to dynamically adjust their paths, ensuring collision-free movement and optimal coordination.