Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Robotics > Robot

The SLAM Research of Indoor Robot Based on Laser Range Finders

Author PengZuoYuan
Tutor WuHuaiYu
School Wuhan University of Science and Technology
Course Control Theory and Control Engineering
Keywords Mobile Robot Simultaneous Localization and Mapping (SLAM) Particle Filter Scan Matching
CLC TP242
Type Master's thesis
Year 2012
Downloads 249
Quotes 0
Download Dissertation

These years the SLAM problem has been a hot issue in research on intelligent robot. It isa basic and important function to use the information captured from sensors to locate the robotreliably. It is also a challenging topic of much concern in the research of robot. SimultaneousLocalization and Mapping (SLAM) is an essential capability for mobile robots to move inunknown environments. The two key questions of navigation are building map in real time andobtaining robot’s pose precisely.Firstly this paper introduces the two key techniques of SLAM, one of which isrepresentative expression means of environmental map, and another is typical localizationmethods of mobile robots. To avoid the inborn problem of the odometry whose angle error willaccumulate ceaselessly with time growing, a mixed localization approach which fuses theodometry and laser information is developed. An agglomerative hierarchical clusteringalgorithm is used to extract line features whose parameters are extracted from range scans,while using least squares algorithm to fit straight lines. ICP is used to match the consecutivescan data to correct the angle error of the odometry.In addition, to reduce the complexity of data association and enhance the real-timeperformance of the algorithm, an improved the re-sampling part of Rao-BlackwellizedFastSLAM algorithm is put forward using Bayesian gird map and partial filter. We called itRPF, regular particle filter. It approximately could restore the discrete posterior probabilitydensity into a series of distribution, so RPF could avoid the problem of lacking variety.Simulation results demonstrate that the estimate errors of the pose and the landmarkscalculated by RPF are far less than EKF-based and normal particle filter approaches under thesame noise condition.Finally, many experiments are performed based on mobile robot MT-R with proposedSLAM method. The experimental results demonstrate that the proposed method cansuccessfully create a consistent map while locate itself using the map.

Related Dissertations
More Dissertations