Research And Design Of Multiple UAV System For Cooperatively Tracking Moving Targets

With the rapid development of unmanned aerial vehicle(UAV)technology,UAV has gradually entered the civilian field.UAV equipped with cameras can be used to search,monitor and track ground targets,so it is widely used in disaster rescue,regional monitoring,urban anti-terrorism and other fields.However,in the actual environment,the flight environment and tasks performed by the UAV are often complicated,and single UAV has limited capability,so it may not be able to complete the task successfully.Therefore,multiple UAVs need to perform tasks cooperatively.This thesis mainly focuses on the research and design of multiple UAV system for cooperatively tracking ground moving targets.The details are as follows:Aiming at the task requirement of cooperatively tracking moving targets by UAVs,a cooperative tracking system platform is built through hardware and software design.Then the robot operating system(ROS)is used to complete the communication of the system.In order to obtain real-time position of ground moving targets,the Tiny-YOLOv3 algorithm is introduced to detect targets,and the semi-global block matching(SGBM)algorithm is adopted to obtain target depth,Then the target position is obtained by position calculation.Finally,the reliability of the method is verified by using the platform.Aiming at the trajectory planning problem of tracking target by the UAV formation,the leader-follower method is used to complete UAV formation.Then the improved rapidly-exploring random tree(RRT)algorithm based on the artificial potential field method is proposed.By improving the generation of random points in the RRT algorithm,the smoothness of the trajectory is improved,and the safety of UAVs is ensured.Finally,the virtual simulation and outdoor experiment of tracking target by UAV formation are successfully realized,and the effectiveness of the proposed methods is verified.Aiming at the trajectory planning problem of cooperatively tracking target by UAVs,and considering that the tracking is a multi-objective and multi-constraint problem,a fuzzy trajectory planning method is proposed.Target coverage degree and control input cost are set as the objective functions.Then the fuzzy optimization is introduced to fuzzify all the objectives,and the UAVs fuzzy trajectory model is established.The cooperatively tracking trajectory is obtained by solving the model.Finally,the virtual simulation and outdoor experiment of cooperatively tracking target by UAVs are successfully realized,and the effectiveness of the proposed methods is verified.To solve the problem of cooperatively tracking multiple targets by UAVs,the task allocation method is used to assign targets for the UAVs.First a color detection algorithm is introduced to distinguish multiple targets.Then according to the actual requirements,the task allocation model is established.The model is solved by branch and bound method and ruled-based heuristic algorithm to assign targets to UAVs.And the proposed fuzzy trajectory planning method is used to plan the tracking trajectory for the UAVs.Finally,the virtual simulation and outdoor experiment of cooperatively tracking multiple targets by UAVs are successfully realized,and the effectiveness of the proposed methods is verified.