Research On Autonomous Collaborative Search Method Of Multi-UAV Based On Swarm Intelligence

Searching for unknown areas has always been a difficult and dangerous job,but with the development of UAV technology,we get a new solution.Under normal circumstances,due to the complex environment and the large number of targets,a single UAV is often not enough to complete the coverage search of the area and the follow-up target tracking or strike work,so multiple UAVs are required to cooperate.In the process of cooperation,as an intelligent agent with independent decision-making and execution capabilities,how to make better use of swarm intelligence to achieve the effect of "1+1>2" is a problem worthy of in-depth study.Taking a typical collaborative search scenario as an example,the effectiveness of the collaborative search method in this paper is verified through simulation experiments and numerical analysis.The main work and achievements of this paper are as follows:In this paper,aiming at the cooperative search problem of multi-UAV to multi-moving target in the region,considering the constraints of UAV sensors and collision avoidance,as well as the characteristics of pan-tiltzoom sweep and target random movement,a cooperative search method of multi-UAV based on vertical line search is proposed.The method consists of task assignment and path planning.In the task assignment part,the equally-divided course vertical line search pattern is designed.In the path planning part,the improved artificial potential field method is designed.In typical scenarios,the simulation verifies the effectiveness of the method.Compared with the traditional vertical search method,this method can guide multi-UAV to capture more targets under the premise of ensuring safety and effectively improving the cooperative search effect of multiUAV.In order to further build a more realistic model,the detection probability and false alarm probability of sensors and 3D search environment are considered,and a 3D collaborative search method of multi-UAV based on information map is proposed.Based on the balance of short-term,long-term and coordination benefits of UAV search,a collaborative multi-UAV search model considering 3D scenes is constructed in this method,and a search information map is designed,including target presence probability,environmental uncertainty,revisit pheromone and search gain.Based on the rolling programming framework,the proposed pruning method was integrated and the model was solved by particle swarm optimization(PSO).Simulation results show the effectiveness of the proposed method in typical scenarios.The simulation results show that the proposed method can make 3D path planning for each UAV in seconds.The comparative simulation shows that the proposed method can capture more targets and have fewer misjudgment times,which effectively improves the cooperative search effect of multi-UAV.