Research And Implementation Of Multi-UAV Cooperative Search Algorithm

UAVs play an important role in the military and civilian fields,and are increasingly used.In the search and rescue mission scenario,a single UAV is not well qualified for this task due to the limited sensing range of the airborne sensor.Therefore,multiple uavs need to be used for cooperative search.It has important practical significance to control multiple uavs to cooperate with each other to efficiently complete the search of the mission area.This thesis aims at the safe and efficient search of the mission area by Multi Unmanned Aerial Vehicle.First,in order to ensure the safety of the uav,the obstacle avoidance trajectory planning algorithm of the uav in different environments is designed;and then in the case where the target location information is unknown,a multi-UAV coverage search algorithm based on area division is designed;finally,for the situation with prior information about the target location,Aiming at the situation with a priori information of the target position,a multi-UAV cooperative area target search algorithm inspired by the target probability map is designed.This thesis is mainly reflected in the following three aspects:(1)In order to ensure the safety of UAV flight,the obstacle avoidance trajectory planning algorithm of UAV in different environments is designed.In an unknown environment,an artificial potential field method based on the Tangent Bug algorithm is designed to complete the UAV’s obstacle avoidance;in an environment with known obstacles,the A* algorithm is used to quickly generate a pre-planned path.When the obstacle is unknown,the improved artificial potential field method is used to guide,and when passing through the sub-path points of the pre-planned path,it continues to fly along the pre-planned path.The simulation experiment results show the effectiveness of the improved artificial potential field method and the efficiency of the pre-planned path obstacle avoidance algorithm based on the A* algorithm.(2)Aiming at the situation that the target location information in the mission area is unknown,a coverage search algorithm based on area division is designed to convert the area coverage search problem of multiple uavs to the area coverage search problem of single uavs.The area is divided based on the initial position of the uavs and the number of uavs,and then uses the spanning tree algorithm to cover,and the virtual node is used to improve the problem that the master node with obstacles in the traditional spanning tree algorithm area cannot be covered.Simulation experiments show that the multi-UAV coverage search algorithm designed in this thesis has a higher coverage rate and a lower repetitive search rate.(3)Aiming at the situation that the target location information in the mission area is partially known,a multi-UAV cooperative area target search method inspired by the target probability map is designed.First,establish the target probability graph model,and initialize the target probability graph according to the prior information,then use the Bayes criterion to update the target probability graph,and set the objective function of maximizing environmental search revenue as the target search trajectory planning,and then use distributed model predictive control to solve the objective function,and finally realizes the cooperative search of multiple UAVs.In addition,considering the existence of obstacles in the mission environment,an improved artificial potential field method is used to avoid obstacles.Simulation experiments show that compared to coverage search and random search,the search algorithm can find targets faster and can avoid obstacles in the task area well.