Research And Implementation Of Target Tracking Method For Mobile Robot Based On TLD Framework

In recent years,with the development of science and technology and the in-depth study of computer vision and robots,the technology of intelligent mobile robots has made tremendous progress.Target tracking technology of mobile robots is one of the key technologies in the field of intelligent mobile robots,and has a wide application prospect in many fields.The target tracking system of mobile robots mainly consists of two parts: 1)a stable and robust machine vision target tracking algorithm suitable for use scenarios;2)vision-guided motion control of mobile robots based on target tracking results.Target tracking and vision-guided motion control are the core components of mobile robot target tracking system.In this paper,Turtlebot3 mobile robot is used as the hardware carrier,and a target tracking algorithm based on TLD(Tracking-learning-Detection)framework for mobile robot is studied and proposed,which can realize the tracking of mobile target on Turtlebot3 mobile robot platform.The main work of this paper is as follows:(1)Investigate and analyze the research background of the subject and the development status of mobile robot target tracking.On this basis,we focus on the research and analysis of TLD video target tracking algorithm,which is used for long-time real-time tracking of single target in video image sequence.TLD target tracking algorithm is a fusion algorithm of detection and tracking.The whole system framework is decomposed into detection module,tracking module and learning module.The detection module detects the target globally.The tracking module tracks the continuously visible target.The learning module can realize the learning of target characteristics.Each module runs independently and cooperates with each other to realize TLD algorithm with real-time target tracking.(2)In this paper,based on TLD target tracking algorithm,a tracking point selection method is proposed for tracking module,which combines uniform point selection with corner extraction.Experiments show that the robustness of tracking module is improved.For detection module,a fusion method of Kalman filter position prediction and TLD algorithm detection module is proposed.Experiments show that dynamic updating can effectively improve detection and real-time performance of modules.(3)Based on the improved TLD target tracking algorithm,this paper proposes a vision-guided target tracking system for mobile robots,builds the hardware and software platform of the tracking system,and proposes a vision-guided method to control the motion of mobile robots to achieve target tracking.Experiments show that in the case of low accuracy and speed requirements,the proposed tracking system can meet the requirements.The tracking system can track the target.