Multi-robot Cooperative Search for Radioactive Sources Based on Improved Particle Swarm Optimization

Effective management and monitoring of radioactive sources are critical to ensuring nuclear safety and minimizing risks to human health and the environment. This paper proposes a multi-robot collaborative radioactive source search algorithm based on improved particle swarm optimization (PSO). The algorithm fuses the latest observations into the particle sampling, enabling particles to converge toward high-likelihood regions and mitigating particle degradation. In addition, the multi-robot gradually narrows the search range using particle filtering algorithm to improve the accuracy of source state estimation. Finally, all robots combine DWA dynamic window obstacle avoidance with Infotaxis to realize distributed collaborative decisions. Experimental results show that when the algorithm adopts the improved particle swarm optimization algorithm, the search accuracy and average search time under different parameter combinations are significantly improved.