Abstract
The task assignment strategy of Unmanned Aerial Vehicle (UAV) swarm under the improved Successive Convex Approximation (SCA) framework is studied in this paper. A mathematical model with binary decision variables is established via considering the heterogeneous characteristics of Unmanned Aerial Vehicles (UAVs) and multiple constraints such as completion rate, execution synchronization and timeliness of task implementation. To address the deficiencies of traditional SCA, a single-loop structure where convex approximation functions and Lagrange multipliers are updated in parallel is introduced. Alternating Direction Method of Multipliers (ADMM) algorithm is used to reduce the number of iterations for solving multivariate optimization problems. The convergence of the algorithm and the stable solution of the optimization problem are obtained through applying the difference method and the Lagrange multiplier method. The results are also applied to the case of UAV failure. Finally, the efficacy of the proposed method is validated through the utilization of UAV swarm to attack enemy positions.
Funding Information
  • National Natural Science Foundation of China (62333011)
  • National Natural Science Foundation of China (W2341048)
  • Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX25_0588)

This publication has 35 references indexed in Scilit: