A Collaborative Navigation with Optimal Clustering for Multi-layer Swarm Systems

This paper presents a cluster-optimised collaborative navigation method for large-scale multilayer swarm systems, addressing challenges in GNSS-denied environments. The proposed approach divides the swarm into adaptive clusters, selecting cluster heads based on optimised metrics to reduce communication load while maintaining navigation accuracy; it integrates a TDOA/FDOA-based collaborative navigation model for heterogeneous node coordination and employs a multidimensional scaling algorithm for intracluster relative positioning and coordinate alignment. The system architecture features a three-layer structure—reference nodes, intermediary nodes, and clustered nodes—enabling efficient resource utilisation and robust navigation support. Experimental results demonstrate that the method substantially lowers the load while keeping a satisfactory navigation accuracy, facilitating reliable navigation in complex scenarios such as disaster response and intelligent logistics. This work provides theoretical and technical support for enhancing the adaptability and operational efficiency of large-scale multi-layer swarm systems.