Multi-UAV cluster-based cooperative navigation with fault detection and exclusion capability

Abstract

As a strategy for multiple Unmanned Aerial Vehicle (UAV) systems to commit tasks in complex application environments, cooperative navigation has attracted extensive research interest in recent years. With the increase of the formation scale, centralized cooperative navigation with fully-connected sensors would be inefficient and unreliable. To reduce the computation and communication load, this paper divides the whole system into several groups and designs two categories of cluster-based fusion architectures: locally-centralized structure and distributed structure. The UAV positions are estimated by the least-squares method. Considering the fault modes in multi-UAV systems, a fault detection and exclusion scheme is developed for improving the reliability of the cluster-based cooperative navigation system. MATLAB simulation and Spirent simulator test are carried out to validate the proposed algorithm. To simulate complex environments, the industry-recognized Spirent simulator and Spirent Sim3D software are adopted. The proposed scheme can be applied to both open areas and challenging environments, which has a wide range of applications.