A Multistate Network Approach for Resilience Analysis of UAV Swarm considering Information Exchange Capacity

Unmanned aerial vehicle (UAV) swarms can perform tasks in a self-organized and self-adaptive manner, and are thus appropriate for resilience research. An efficient and resilient information exchange (IE) network among drones is essential for a UAV swarm to accomplish its mission. In this study, we incorporated conditional probability into a multistate network to model the IE of a UAV swarm. Subsequently, we modeled two resilient behaviors based on the actions of the drones: formation transformation and redeployment. For the former, a semi-Markov-based model was adopted to represent the changes in the state probability distribution of the IE link during this resilient behavior. A resilience-based swarm IE topology reconstruction optimal approach for UAV redeployment strategy was presented. Finally, an application case of a UAV swarm was evaluated, for which real experiments were conducted to obtain the state distributions of the IE capacity among UAVs at different distances. The simulation results show that the proposed model and method can help gain understanding of the resilience process of a UAV swarm and can be used to select appropriate recovery strategies, further supporting mission planning and improving the resilience of UAV swarms.

Automated summary

This study incorporates the characteristics of self-organized and self-adaptive task execution of UAV swarms into resilience research. It proposes an efficient and resilient information exchange network model and simulates two resilient behaviors. Experimental evaluation demonstrates that the proposed model and method contribute to understanding the resilience process of UAV swarms and provide insights for selecting appropriate recovery strategies to support mission planning and enhance the resilience of UAV swarms.