Joint Task Offloading and Resource Allocation for UAV Swarm Networks
This paper investigates the task offloading and resource allocation problem in unmanned aerial vehicle (UAV) swarm networks, with the objective of minimizing a weighted sum of task completion latency and energy consumption. Considering the autonomous decision‐making characteristics of individual UAVs in the swarm, each UAV is modeled as an intelligent agent and classified into heterogeneous types according to its computational capability. Based on this modeling framework, a mixed‐integer nonlinear programming (MINLP) problem is formulated to jointly optimize task offloading decisions and UAV transmission power. Owing to the high computational complexity of the original problem, it is decomposed into a transmission power allocation subproblem and a task offloading subproblem, where the optimal transmission power allocation strategy is obtained via a bisection‐based method. Furthermore, to enable efficient and rational task offloading within the UAV swarm, a matching game‐based task offloading algorithm is proposed, and its stability and convergence are theoretically proven. Finally, extensive simulation results and comparisons with multiple baseline schemes demonstrate the effectiveness and superiority of the proposed approach in terms of system latency and energy efficiency.