Cooperative Path Planning Method Of A Vehicle And A Vehicle-carried UAV For Transmission Tower Inspection

With the UAV inspection gradually becoming the normalization means for the inspection of transmission towers by provincial power companies,the inspection efficiency of power towers has been greatly improved compared with the period of manual inspection.However,the current UAV inspection mainly depends on manual operation,with low degree of automation,and the efficiency is difficult to further improve.In this case,the way of vehicle and UAV cooperating to inspect transmission tower independently is proposed,and the path planning of UAV and vehicle,as the key to directly affect the inspection efficiency,is of great significance.The heterogeneity of transmission tower inspection task,the limited endurance of UAV and the relationship between vehicle path and UAV path bring new changes and challenges to the modeling and solution of the path planning problem.In this paper,the cooperative path planning method of UAV and vehicle for transmission tower inspection is studied.In this paper,we analyze the key elements of the cooperative path planning of UAV and vehicle,and construct a model of TSP-D variant problem.First of all,the functions of UAV and vehicle in the inspection should be defined,and their flight and driving rules should also be defined according to the characteristics of the functions.On this basis,the cooperative mode between them and the path characteristics under the cooperative constraints should be further considered.The time consumption is reduced through collaborative and parallel,the he total time needed to complete the inspection is taken as the objective function,that is,the TSP-D model for transmission tower inspection(TTITSP-D)is constructed.Then we use genetic algorithm to solve the above model.By analyzing the characteristics of the cooperative path planning problem and model,we propose the odd and even layered coding method to realize the transformation from problem space to chromosome space,and design the matching cross method,variation method,check method of solution and correction method of infeasible solution.Finally,the performance of the algorithm is analyzed in several aspects through simulation experiments.The correctness of the algorithm is verified by comparing with enumeration algorithm in small-scale test cases.The effectiveness of the algorithm is shown by comparing with conventional strategies in different scale tests.In addition,in order to improve the efficiency of the algorithm,the sensitivity of the relevant parameters of the algorithm is analyzed through experiments,and the optimal parameter configuration of the algorithm is given.