Multi-Robot Cooperative Path Planning: Advances, Challenges, and Future Trends

Abstract

From automated warehouse robots coordinating shelf movements to drones and ground vehicles navigating disaster zones, multi-robot systems (MRS) are transforming how complex tasks are performed in dynamic environments. Unlike single-robot systems, MRS can execute tasks in parallel, cover large areas efficiently, and adapt to unexpected changes. These advantages make them ideal for applications such as smart manufacturing, autonomous logistics, and emergency response. However, as robot teams scale up, they face growing challenges—task allocation becomes harder, path conflicts increase, and communication delays can undermine real-time coordination. This paper provides a structured review of cooperative path planning approaches for MRS. First, outline the fundamental characteristics of multi-robot coordination and key challenges such as collision avoidance, computational complexity, and dynamic re-planning. Then, categorize the main planning algorithms into three types: graph-based methods, intelligent optimization techniques, and deep learning models. We also compare centralized and distributed planning frameworks in terms of scalability, robustness, and real-world feasibility. Application case studies from warehouse systems, intelligent transportation, and search-and-rescue missions are presented to demonstrate how these planning strategies are implemented in practice. Finally, discuss open challenges such as heterogeneous robot integration, safety assurance, and communication resilience, and highlight future research directions including hybrid architectures and learning-driven coordination. This review aims to support the development of scalable, adaptive, and efficient path planning in multi-robot systems.