UAV-USV Cooperative Task Allocation based on Improved Grey Wolf Optimizer Algorithm
Shi-Qi Liu, Xunhong Lv, Zehui Mao · 2025
Task allocation is a prerequisite for the cooperative operation of the UAV-USV system. Its essence is a multi-objective optimization problem under different constraints, and heterogeneous platforms make the constraints more complex and increase the difficulty of solving the problem. In this paper, an improved grey wolf optimization algorithm is proposed. The chaos reverse learning strategy is introduced in population initialization to increase the diversity of initial solutions to accelerate the convergence speed and improve the search efficiency. A weighted position update method based on the competitive leader is used to avoid falling into the local optimum and speed up the fusion. The adaptive control parameters based on event-driven are introduced to take into account the different computing requirements of offline task allocation and online task reallocation. The experimental results show that compared to the traditional grey wolf algorithm, the task pre-allocation results are improved by 8.06%, the running time of task reallocation is reduced by 29.9%, and performance is comprehensively improved.