Font Size: a A A

Multi-Sensor Based Robot Formation And Tracking Control

Posted on:2020-12-21Degree:MasterType:Thesis
Country:ChinaCandidate:Y YuFull Text:PDF
GTID:2428330596982644Subject:Control engineering
Abstract/Summary:PDF Full Text Request
In recent years,with the continuous development of robotics technology,the research of multi-robot collaboration has entered a new stage,and more and more applications have been obtained.Multi-robot cooperation can accomplish tasks that are difficult for a single robot.The multi-robot formation problem is a typical problem in multi-robot cooperation.Based on the comprehensive application of multi-sensors,this paper studies the composition of different formations in the "Leader-Follower" mode,the transformation between different formations,and the processing of formations through the corners.The main research content of this paper can be divided into the following parts:Firstly,based on two-dimensional laser radar,a multi-point sequential navigation method is proposed.The leader robot uses the two-dimensional laser radar to construct a high-precision experimental scene map,design the travel path of the multi-robot formation,mark a series of target points on the travel path,and enable the leader robot to achieve multi-point sequential navigation,and then provide the entire multi-robot formation a path of travel.Secondly,based on binocular vision camera,a method of target recognition and localization is proposed.The follower robot uses the image processing technology to identify and locate the target robot using the carried binocular vision camera,and sets the relative positional relationship between the two to form different formations.Then,based on the wireless communication module,a communication strategy between multiple robots is proposed.At the corresponding position,the leader robot sends a control command for formation maintenance and formation transformation to follower robots.Finally,by setting up an experimental platform to achieve the transformation between different formations,when different formations through the corners,the corresponding strategy is adopted to make the robot formation smoothly pass through the corner.The feasibility of four key techniques is verified by experiments,which are inline formation to triangle formation,triangle formation to inline formation,inline formation through a corner,and triangle formation through a corner.
Keywords/Search Tags:Robot Formation, Tracking Control, Multi-point Sequential Navigation, Computer Vision
PDF Full Text Request
Related items