Autonomous Multi-Task Reliability Allocation Method for Unmanned Swarms

The technology of autonomous unmanned swarm systems, as a disruptive innovation poised to transform future technological development, is gaining increasing attention. It has the capability to autonomously complete critical tasks such as mapping, search and rescue, and strikes in harsh, dangerous, or otherwise inaccessible environments. This technology is playing an increasingly significant role across various fields, including national defense, meteorology, and exploration. Consequently, the demands for its reliability in executing complex tasks are also rising. However, due to the autonomy of individual units within the swarm and the coordination required among them, it is challenging to effectively allocate reliability metrics to individual unmanned units. Traditional reliability allocation methods, which consider factors like complexity and environmental conditions, are not well-suited for these systems. This paper, drawing on the multi-task behavioral patterns of autonomous unmanned swarms and considering the fault-recovery characteristics during mission execution, proposes a reliability allocation method based on dynamic trade-off strategies. This method aims to provide a reference for the allocation of reliability metrics in autonomous unmanned swarms.