Font Size: a A A

Design And Implementation Of Localization And Mapping System Based On The Fusion Of 2D Lidar And RGBD Camera

Posted on:2021-02-10Degree:MasterType:Thesis
Country:ChinaCandidate:Y F WangFull Text:PDF
GTID:2428330632453240Subject:Computer technology
Abstract/Summary:PDF Full Text Request
At present,the development of computer has greatly influenced on robots'development.At the same time,the AGV robot is used more and more,but there are still several problems to be solved,such as the poor positioning accuracy of a single sensor,while traditional laser data and In the visual data fusion scheme,when using low-cost lidar,the accuracy of positioning and map construction is not ideal.Moreover,the utilization efficiency of the map is low.In indoor scenes,although the traditional grid map has good positioning accuracy,it sometimes does not provide good assistance for navigation and obstacle avoidance.To solve the above problems and improve robot system's stability,the following efforts are made in this paper:1.The process and algorithm of location and map construction using single sensor lidar and rgbd camera are studied respectively.The traditional methods of lidar and visual data fusion are analyzed.Finally,a data tightly coupled fusion way is suggested in this essay.According to the test,the way suggested has a good performance than traditional way.2.In order to solve the problem that the time of robot relocation is too long and improve the navigation accuracy and the success rate of obstacle avoidance,this paper modifies the data structure of traditional raster map,and adds a dimension of feature information to the position information of two-dimensional information,which can not only improve the speed and success rate of robot relocation,but also provide more useful feature information for obstacle avoidance and navigation.Finally,in engineering practice,the proposed 2D lidar and RGBD camera data fusion method and the new raster map representation method are successfully applied to the factory AGV robot,and experiments show that the method in this paper can meet the factory environment The demand for AGV robot positioning and mapping ensures the stability of factory cargo logistics and transportation,and it greatly improves the efficiency of factory logistics and transportation,which can save a lot of labor costs.
Keywords/Search Tags:AGV'robot, SLAM, Laser-Vision multi-sensor fusion, nonlinear optimization
PDF Full Text Request
Related items