ZOU Yong-wei, WU Bin
（National Key Laboratory of Precision Testing Techniques and Instrument， Tianjin University， Tianjin 300072, China）
Abstract： Mobile robot navigation in unknown environment is an advanced research hotspot. Simultaneous localization and mapping (SLAM) is the key requirement for mobile robot to accomplish navigation. Recently, many researchers study SLAM by using laser scanners, sonar, camera, etc. This paper proposes a method that consists of a Kinect sensor along with a normal laptop to control a small mobile robot for collecting information and building a global map of an unknown environment on a remote workstation. The information (depth data) is communicated wirelessly. Gmapping (a highly efficient Rao-Blackwellized particle filer to learn grid maps from laser range data) parameters have been optimized to improve the accuracy of the map generation and the laser scan. Experiment is performed on Turtlebot to verify the effectiveness of the proposed method.
Key words： Kinect sensor; mobile robot; autonomous navigation; simultaneous localization and mapping (SLAM); Turtlebot
CLD number： TP242.6 Document code： A
Article ID： 1674-8042（2018）01-0025-07 doi： 10.3969/j.issn.1674-8042.2018.01.004
References
［1］ Gao Y, Zou D, Hu D W, et al. Path planning for mobile robot in unknown environment and exploration algorithm. Electronic Design Engineering, 2014, 22(3): 1-3.
［2］ Siegwart R, Nourbakhsh I R. Introduction to autonomous mobile robots. Massachusetts: MIT Press, 2004.
［3］ Smith R, Self M, Cheeseman P. Estimating uncertain spatial relationships in robotics. In: Proceedings of IEEE International Conference on Robotics and Automation Proceedings, 2003: 435-461.
［4］ Choi J, Ahn S, Wan K C. Robust sonar feature detection for the SLAM of mobile robot. In: Proceedings of IEEE International Conference on Intelligent Robots and Systems, IEEE, 2005: 3415-3420.
［5］ Mallios A, Ridao P, Ribas D, et al. EKF-SLAM for AUV navigation under probabilistic sonar scan-matching. In: Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010), Taipei, 2010: 4404-4411.
［6］ Mei C, Sibley G, Cummins M, et al. RSLAM: a system for large-scale mapping in constant-time using stereo. International Journal of Computer Vision, 2011, 94(2): 198-214.
［7］ Marks T K, Howard A, Bajracharya M, et al. Gamma-SLAM: using stereo vision and variance grid maps for SLAM in unstructured environments. In: Proceedings of IEEE International Conference on Robotics and Automation, IEEE, 2008: 3717-3724.
［8］ Mcdonald J, Kaess M, Cadena C, et al. Real-time 6-DOF multi-session visual SLAM over large-scale environments. Robotics & Autonomous Systems, 2013, 61(10): 1144-1158.
［9］ Grisetti G, Stachniss C, Burgard W. Improved techniques for grid mapping with Rao-Blackwellized particle filters. IEEE Transactions on Robotics, 2007, 23(1): 34-46.
［10］ Biswal S. Tech News. 2011-02-12［2017-12-23］. http:∥www.techibuzz.com/xbox-360-kinect-review/.
［11］ ROS.org. ROS Wiki. 2012-09-21［2017-12-30］. http:∥www.ros.org/wiki.
［12］ Xia H J. Research of robot SLAM technology based on ROS. Shenyang: Northeasten University, 2013.
［13］ Kamarudin K, Mamduh S M, Shakaff A Y M, et al. Method to convert Kinects 3D depth data to a 2D map for indoor SLAM. In: Proceedings of the 9th IEEE Colloquium on Signal Processing and Its Applications, Kuala Lumpur, Malaysia, 2013.
［14］ Murphy K P. Bayesian map learning in dynamic environments. In: Proceedings of the Neural Information Processing Systems, Denver, CO, 1999: 1015-1021.
［15］ Doucet A, Freitas N D, Murphy K P, et al. Rao-blackwellised particle filtering for dynamic bayesian networks. In: Proceedings of the 16th Conference on Uncertainty in Artificial Intelligence, Stanford, CA, 2000: 176-183.
基于Kinect传感器的移动机器人自主导航研究
邹永卫， 吴 斌
（天津大学 精密测试技术及仪器国家重点实验室， 天津 300072）
摘 要: 未知环境移动机器人自主导航是当前的研究热点。 同步定位与建图是实现移动机器人自主导航的关键环节。 目前， 许多学者通过激光测距仪、 声呐、 相机等设备来研究同步定位与建图。 本文提出了由一台笔记本控制携带Kinect传感器移动机器人获取信息， 再通过独立工作站创建未知环境全局地图的方案。 其中， 笔记本获取的信息（深度数据）通过无线传输给工作站。 Gmapping（一种高效的Rao-Blackwellized粒子滤波器将激光扫描数据生成栅格地图）参数被优化来提高地图创建质量和激光扫描数据精度。 通过Turtlebot进行实验验证了本文所提方案的有效性。
关键词: Kinect传感器； 移动机器人； 自主导航； 同步定位与建图； Turtlebot
引用格式： ZOU Yong-wei, WU Bin. Investigation on mobile robot navigation based on Kinect sensor. Journal of Measurement Science and Instrumentation, 2018, 9(1)： 25-31. ［doi： 10.3969／j.issn.1674-8042.2018.01.004］
[full text view]
