Research On Vision-based Robot Target Tracking System

With the development of science and technology,the service robot,as the representative of various robot technology integration,is developing towards the direction of intelligence and autonomy.As a kind of service robot,the following-type robot can independently complete a continuous follow-up action without a physical contact with the target,and keep a certain distance from it.It has important applications in the fields of special population(disability,elderly)assistance,logistics handling,military transportation and so on.Visualbased tracking robot can adapt to complex working environment,and the recognition accuracy is much higher than other detection methods.Therefore,the following-type robot based on vision guidance is the development trend in the future.However,how to achieve the accuracy and real-time performance of visual tracking system at the same time under the condition of limited hardware resources is still a difficulty in the research.Taking the situation as an opportunity,based on embedded hardware platform,a kind of autonomous following-type robot with monocular camera is designed in this paper.The research work on the carrier structure,target recognition algorithm and motion control strategy of the robot is carried out.The specific research contents are as follows:Firstly,according to the actual functional requirements,the mechanical system and control system structure of the robot are designed.Then,select the related hardware,and build the development environment.Secondly,by using the AprilTags recognition method to implify target features to improve recognition rate.Aiming at the low efficiency of the algorithm on the embedded platform and the poor recognition effect under the condition of partial occlusion,the original algorithm is improved by taking the real-time performance and accuracy into consideration.The real-time performance of the algorithm is improved by appropriately reducing the image resolution and taking local search to reduce the amount of calculation.By introducing the local Haar-Like feature of the tag and combining with the cascade classifier,the recognition ability of AprilTags marker under local occlusion is improved and the accuracy of detection algorithm is enhanced.Then,based on the target location information,the motion control strategy of the robot system is designed,and the construction and debugging of the system with the robot are completed.Finally,test the following-type robot system,the improved target detection algorithm and control strategy are verified by experiments.The results show that the improved algorithm has better real-time performance and accuracy,the robot can achieve stable tracking of the guided target.The following-type robot can achieve the stable following action of the guiding target and complete the autonomous following task.