Đang chuẩn bị liên kết để tải về tài liệu:
Enhanced robot pose estimation method using selective scan data in structured environments

This paper suggests enhanced robot pose estimation method by using selective scan data in structured environments. Previous pose estimation approaches, which estimate the pose by scan matching method, use scan data of constant interval. | Journal of Automation and Control Engineering Vol. 3, No. 5, October 2015 Enhanced Robot Pose Estimation Method Using Selective Scan Data in Structured Environments Wonsok Yoo and Beom H. Lee Department of Electrical and Computer Engineering, Seoul National University, Seoul, Republic of Korea Email: {yoowon0, bhlee}@snu.ac.kr object registration with the scan matching algorithm. We changed the overlapping areas between two consecutive scan data for the efficiency of the registration. This technique is Selective Data Process (SDP). This technique alters the overlapping areas according to the prior scan data distribution. In this paper, we will extend this SDP, which is used for the object registration, to the robot pose estimation problem. The rest of this paper is organized as follows: Section II describes the problem of previous constant interval scan matching algorithms. Section III gives an overview of the SDP which is our previous work. Section IV describes detailed technique of our new approach. In section V, the experimental results are shown and the results of our new approach and previous approach are compared. Finally, we conclude the paper with an outlook on future work in section VI. Abstract—This paper suggests enhanced robot pose estimation method by using selective scan data in structured environments. Previous pose estimation approaches, which estimate the pose by scan matching method, use scan data of constant interval. However, these approaches do not consider the property that scans matching result is affected by the scan data distribution in the overlapping areas between two consecutive scans. Our proposed method varies scan interval in order to adjust overlapping areas between current and next scans. Through the experiments, we compared our method to the previous approaches which use constant interval scans and verified improved performance of our new approach. Index Terms—robot, pose estimation, scan matching, map building, selective data