Research On Path Planning For UAV Fixed-point Tasks

Thanks to the rapid development of science and technology,the functions of all kinds of unmanned intelligent machines are becoming richer and richer.More and more industries put unmanned intelligent machines into social production and daily use.UAV is widely used in military,special industries,consumer and civil fields because of its strong performance and minimal restriction by terrain and road.With the substantial reduction of hardware cost and the increasing use of UAVs,it is difficult to meet the needs of practical application scenarios only relying on manpower for task allocation and flight control of UAVs.The scenario of cooperative mission planning of UAV cluster has been paid more and more attention.According to the actual situation of UAV group and the information of mission points,the cooperative mission planning of UAV group needs to formulate a mission implementation scheme for UAV group in order to achieve the corresponding purpose.At present,there are two main problems in the cooperative task planning of UAV cluster.The first is the uneven task allocation of UAV cluster.The second is that the task planning takes too long to meet the real-time requirements in the dynamic redistribution of UAV tasks.Aiming at the above two problems,the key work of this paper mainly includes the following three parts:(1)a task target model based on UAV range and UAV task balance is proposed,and based on the above model,a task pre allocation method of UAV cluster based on improved genetic algorithm is designed The population grouping strategy and the metropolis function acceptance criterion of simulated annealing control the direction of population search and accelerate the convergence speed in the early stage of the algorithm.Secondly,the adaptive crossover mutation strategy is used to improve the shortcomings of the traditional genetic algorithm,which is easy to fall into the local optimal solution and poor optimal solution in the later stage,and strengthen the search performance of the genetic algorithm;(2)A dynamic task redistribution method for UAV is proposed.Firstly,the UAV tasks are grouped by using the task division algorithm based on task point cluster and route double verification,and the multi UAV path planning problem is transformed into a single UAV path planning problem.Secondly,the performance of tabu algorithm is improved by three optimization strategies: preset initial solution space,multiple initial solutions and adaptive tabu table length,and the task set of a single UAV is planned by this method.(3)Combined with the control system of unmanned intelligent equipment,the task collaborative planning method of UAV group is implanted into it.The experimental results show that the mission target model and UAV task allocation method based on UAV range and UAV task balance can effectively reduce the total flight distance of UAV group and the range between UAVs is relatively balanced;The UAV task redistribution method proposed in this paper can quickly respond to emergencies,and the response time is seconds,which meets the needs of the practical application scenario of UAV task allocation system.