TASK ALLOCATION FOR UAV SWARM INSPECTION IN WIND FARMS BASED ON IMPROVED CBBA ALGORITHM

Wind turbines are widely distributed and located in complex environments，which are more suitable for autonomous inspection by UAV swarms，where the multi-UAV task allocation problem needs to be solved. In this paper，we propose a multi-UAV wind farm inspection task allocation method based on improved CBBA algorithm. Firstly，based on the basic information of wind turbines，the K-means algorithm is used to determine the number and placement of UAVs，and the value assessment function of wind turbine inspection is constructed to more reasonably evaluate the inspection revenue of each wind turbine；then，a two-way construction strategy of task package based on CBBA algorithm is proposed for the homing and returning problem of UAVs；Finally，to overcome the issue of individual greed reducing overall benefits in the traditional CBBA algorithm，a category-based reward system is designed to enhance the rationality of task allocation. The experiments show that the proposed method can develop a corresponding inspection task allocation scheme according to the changes of wind speed and wind direction，and provide a referenceable scheme for the task allocation of autonomous inspection of wind farms by UAV swarms. © 2024 Science Press. All rights reserved.