Research On Target Tracking System Of UAV Based On Type-2 Fuzzy Logic Control Theory

With the continuous maturity of related hardware technologies of UAV,UAV target tracking systems have contributed greatly in many tasks such as navigation,military reconnaissance,field rescue,and traffic monitoring,and have therefore been widely used.However,There are also many difficulties in the research of UAV target tracking systems,such as target occlusion,uncertainty in external environmental interference,target multi-scale transformation,and deformation of non-rigid objects,etc.These are the shortcomings in the current research on UAV tracking systems.In this paper,aiming at the problem of target multi-scale transformation and uncertainty in the UAV tracking process,combined with the idea of multi-scale target detection,the classical Kernel Correlation Filter(KCF)algorithm is improved,and a multi-scale adaptive target detection and tracking algorithm is proposed.Based on the idea of type-2 fuzzy logic control,the UAV target tracking IT2 FLC is designed.First,we preliminarily set up the structural framework of the UAV target tracking system and formulate the overall design plan of the system.Then,for the purpose of solving with the problem that the KCF algorithm cannot adapt to the target at multiple scales and the model is contaminated after the target is lost,the model cannot be tracked again,the search strategy after the target is lost is further designed.Finally,a IT2 FLC is designed to control the yaw angle and speed of the UAV to realize the autonomous flight control of the UAV under uncertainty.On the premise of ensuring the safety of the experiment,in order to verify the reliability of the system designed in this paper,firstly carry out modeling and simulation on the gazebo software,and then perform the target tracking experiment on UAV in an outdoor environment.In a practical sense,the autonomous tracking of UAV on moving targets on the ground is realized.The main contributions of this article are as follows:(1)A multi-scale target detection and tracking algorithm based on KCF is proposed.Multi-scale training is performed on candidate target regions by introducing scale transformation factors,so that the algorithm can adaptively track the moving targets of multi-scale transformation,thereby improving the accuracy of the target tracking algorithm.(2)Aiming at the shortcoming of the KCF tracking algorithm that the target cannot be relocated and tracked when the target reappears after the target is lost,a target airspace search and relocation strategy is proposed,so that the target can be relocated and tracked continuously after the target is lost.(3)The IT2 FLC we designed solves various uncertainties in the tracking control process by controlling the yaw angle and speed of the UAV,and improves the efficiency of UAV target tracking.(4)Combining the idea of target tracking and fuzzy control theory,a complete UAV target tracking system is designed.Experiments are carried out on the simulation environment and the actual quadrotor UAV to prove the reliability of the target tracking system designed in this paper.