Study Of Multiple UAV Cooperative Path Plan For Area Search Under Various Area Information Conditions

Cooperative search of multiple unmanned aerial vehicles(multi-UAV)is the main approach to obtain the battlefield information in the future,which is one of the most important techniques for future combat.Path planning is a critical technology to guide the UAVs to implement autonomous flight and complete the combat mission.Cooperative search path planning is becoming a hot research direction in the mission planning field,which has important application values and practical significance for military and civil practices.In this paper,aiming at the requirements of the multi-UAV cooperative search task,multi-UAV cooperative search path planning method with different priori information is researched.The main contents of this paper include the following aspects.(1)The research status and key technologies of multi-UAV cooperative search technology are summarized,and the advantages and the existing problems of the overall research are summarized.(2)From the task requirement of multi – UAV cooperative search,the research issues and definition of the search task are clarified firstly.Then,the components of multi-UAV cooperative search are analyzed.Finally,the environment model,the UAV kinematics model,the sensor search model,the target model and the data communication model are established respectively,which lays the foundation for multi-UAV cooperative search path planning problem.(3)For the task region information is completely unknown,the scan line method is used to achieve full coverage of the task region.Firstly,according to the shape of the task area,we focus on the full-coverage multi-UAV cooperative region searching path planning in the rectangular area(barrier-free)and the convex polygon area(obstacle).In this paper,two kinds of multi-UAV cooperative search path planning methods are designed respectively for these two regions.The simulation results show that the planned path based on the scan line method can satisfy the requirement of full coverage of the two regions.(4)In the case that the information for the task region is partially known,the objects in the task region are abstracted into multiple dispersed key targets,which contain point targets and line targets,according to a priori information or target natural attribute.Then,a key type of multi-UAV cooperative area search model is established and is transformed into an extended multi-traveling salesman problem.And an opposition-based genetic algorithm with double chromosome integer encoding and multiple mutation operators is proposed to solve the problem whose objective function is to minimize the task completion time and task consumption for multiple UAVs.Finally,the simulation results show that the proposed method is superior to the ordinary genetic algorithm and the random search algorithm in terms of optimality and robustness.(5)Aiming at the situation that the target probability distribution in the uncertain environment is known,we first establish the search probability graph under uncertain environment and the Bayesian criterion is used to update the search probability map.And maximizing the environmental search benefits is chosen to be the objective function which is inspired by the search information.A multi-UAV cooperative search receding path planning method inspired by the target probability information is proposed.In this method,the multi-UAV cooperative area search path planning under uncertain environment is realized using the Receding Horizon Optimization strategy.Finally,the simulation experiment proves that the heuristic search method is better than the greedy search and the random search.