From simulations to reality: enhancing multi-robot exploration for urban search and rescue

In this paper, we propose a novel hybrid algorithm that combines Levy Flight (LF) and Particle Swarm Optimization (PSO) (LF-PSO), designed for efficient multi-robot exploration in unknown environments without Global Positioning Systems (GPS) or detailed sensor data. The focus of the study is to address challenges in autonomous multi-robot exploration for general-purpose tasks rather than specifically targeting Urban Search and Rescue (USAR) scenarios. Although multi-robot systems offer advantages in terms of scalability, flexibility, and area coverage, many existing approaches rely on assumptions that may not translate well to real-world applications, such as perfect communication or target information. Our hybrid algorithm leverages LF for large-area exploration while integrating PSO-based inter-robot repulsion to prevent clustering and ensure diverse coverage. Experimental results in a controlled obstacle-free environment demonstrate the algorithm's potential for general exploration tasks, though further studies are required to adapt this approach for complex, obstacle-rich environments relevant to USAR scenarios. These simulations highlight improved area coverage compared to traditional methods but do not yet address challenges posed by sensor-based navigation or real-world USAR conditions.