||RBPF and ICP based SLAM and Q-learning based Obstacle Avoidance Strategy for Home Service Robots
||Department of Electrical Engineering
Rao-Blackwellized Particle Filter
本論文主要在探討居家服務型機器人之同步定位與地圖建立(SLAM) 及避障策略的建立及實現。首先，本篇論文介紹結合Rao-Blackwellised粒子濾波器與ICP演算法以實現SLAM之系統，機器人在陌生的環境下可透過雷射測距儀所回傳的距離資訊以學習環境地圖，其中ICP演算法藉由疊代的過程以修正前一時刻地圖與下一時刻觀測資訊之間的轉換式，而Rao-Blackwellised粒子濾波器則為一穩健的解決SLAM的方法，用於處理非線性及非高斯的狀態空間模型上。Q-學習法則應用於四輪獨立轉向四輪獨立驅動系統(4WIS4WID)移動平台上，以實現機器人導航時之避障功能。藉由此學習過程，機器人得以平順且安全的方式在環境中移動。最後將融合以上所提出的方法，透過在實驗室的實驗結果以及 2014 RoboCup日本公開賽居家組的Restaurant項目的比賽成績以驗證SLAM系統和策略系統的可行性與效益。
This thesis mainly discusses the design and implementation of simultaneous localization and mapping (SLAM) and obstacle avoidance strategies for home service robots. The SLAM system is first built using the Rao-Blackwellized Particle Filter (RBPF) method and Iterative Closest Point (ICP) algorithm. The robot learns the map for an unknown environment through information on the distance received by a laser range finder. The ICP algorithm estimates the pose of the robot by iteratively revising the transformation from the prior map to the posterior observation. The RBPF method is a robust way to solve the SLAM problem, which can deal with both the nonlinear and non-Gaussian state space model. Secondly, Q-learning is applied to the four wheel independent steering and four wheel independent driven (4WIS4WID) platform for obstacle avoidance during navigation. After the learning step, the robot navigates smoothly through the environment away from dangers. In the end, the methods mentioned above are implemented in the experimental results in the laboratory and in the competition, Restaurant Mission, in robot@home league at RoboCup Japan Open 2014. The validity and efficiency of the SLAM system and strategy system for the home service robot are demonstrated.
List of Figures Ⅵ
List of Tables Ⅷ
Chapter 1. Introduction 1
1.1 Motivation 1
1.2 Hardware and Software Structure 3
1.3 Thesis Organization 9
Chapter 2. Simutaneous Localization and Mapping 11
2.1 Introduction 11
2.2 The Formulation of the SLAM Problem 12
2.3 Motion Model of the Home Service Robot 15
2.4 Measurement Model of the Home Service Robot 17
2.5 SLAM Module 21
2.5.1 Rao-Blackwellised Particle Filter 23
2.5.2 Iterative Closest Point 25
2.5.3 SLAM Module Using RBPF and ICP 30
2.6 Summary 33
Chapter 3. Q-learning based Obstacle Avoidance 34
3.1 Introduction 34
3.2 Q-learning Algorithm 35
3.3 The Definition of States, Actions and Rewards 38
3.4 Summary 42
Chapter 4. Control Strategy for Restaurant Mission 44
4.1 Introduction 44
4.2 The Rule of the Restaurant Mission 45
4.3 Common Functions 48
4.3.1 Human Following 49
4.3.2 Speech Recognition System 50
4.4 The Strategy System for the Restaurant Mission 52
4.5 Summary 55
Chapter 5. Experimental Results 56
5.1 Introduction 56
5.2 Experimental Result of the Simultaneous Localization and Mapping 57
5.3 Experimental Result of the Obstacle Avoidance 62
5.4 Experimental Result of the Restaurant Mission 63
Chapter 6. Conclusions and Future Work 67
6.1 Conclusions 67
6.2 Future Work 68
 M. W. M. G. Dissanayake, P. Newman, H. Durrant-Whyte, S. Clark, and M. Csorba, “A solution to the simultaneous localization and map building (SLAM) problem,” IEEE Transactions on Robotics and Automation, vol. 17, no. 3, pp. 229-241, 2001.
 H. Durrant-Whyte and T. Bailey, “Simultaneous localization and mapping: Part I,” IEEE Robotics and Automation Magazine, vol. 13, no. 2, pp. 99-110, 2006.
 T. Bailey and H. Durrant-Whyte, “Simultaneous localization and mapping (SLAM): Part II,” IEEE Robotics and Automation Magazine, vol. 13, no. 3, pp. 108-117, 2006.
 R. Leenen, J. J. Ploeg, H. H. Nijmeijer, L. Moreau, and F. Veldpaus, Motion control design for a 4ws and 4wd overactuated vehicle, Master Thesis, Department Mechanical Engineering Dynamics and Control Technology Group, Eindhoven University of Technology, Eindhoven, January 2004.
 F. Yili, H. Xu, W. Shuguo, and M. Yulin, “A navigation robot with reconfigurable chassis and bionic wheel,” in Proceedings of 2004 IEEE International Conference Robotics and Biomimetics, ROBIO 2004, pp. 485-489, 2004.
 S. Thrun, W. Burgard, and D. Fox, “Robot perception,” in Probabilistic Robotics. Cambridge, MA: MIT press, pp. 171-172, 2005.
 R. Smith, M. Self, and P. Cheeseman, “Estimation uncertain spatial relationships in robotics,” Autonomous Robot Vehicles, pp. 167-193, 1990.
 H. Durrant-Whyte, D. Rye, and E. Nebot, “Localization of autonomous guided vehicles,” Robotics Research, pp. 613-625, 1996.
 R. E. Kalman, “A new approach to linear filtering and prediction problems 1,” Transactions of the ASME-Journal of Basic Engineering, vol. 82, no. Series D, pp. 35-45, 1960.
 N. Gordon, D. Salmond, and A. Smith, “Novel approach to nonlinear non-Gaussian Bayesian state estimation,” IEE Proceedings F: Radar and Signal Processing, pp. 107-113, 1993.
 J. Maccormick, “A probabilistic exclusion principle for tracking multiple objects,” International Journal of Computer Vision, vol. 39, no. 1, pp. 57-71, 2000.
 K. Kanazawa, D. Koller, and S. Russell, “Stochastic simulation algorithms for dynamic probabilistic networks,” in Proceedings of the Eleventh Annual Conference on Uncertainty AI., pp. 346-351, 1995.
 M. West, “Mixture models, Monte Carlo, Bayesian updating, and dynamic models,” Computing Science and Statistics, vol.24, pp. 325-333, 1993.
 A. Doucet, S. J. Godsill, and C. Andrieu, On sequential simulation-based methods for Bayesian filtering, UK: Department of Engineering, University of Cambridge, 1998.
 A. Doucet, J. de Freitas, K. Murphy, and S. Russell, “Rao-Blackwellised Particle Filtering for Dynamic Bayesian Networks,” in Proceedings of Conference on Uncertainty in Artificial Intelligence, Stanford, CA, pp. 176-183, 2000.
 K. Murphy and S. Russell, “Rao-blackwellized particle filtering for dynamic bayesian networks,” in Sequential Monte Carlo Methods in Practice, 2001.
 C. Stachniss, G. Grisetti, and W. Burgard, “Recovering particle diversity in a Rao-Blackwellised particle filter for SLAM after actively closing loops,” in Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pp. 667-672, 2005.
 X. Xu and B. Li, “Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance,” IEEE Transactions on Image Processing, vol. 16, no. 3, pp. 838-849, 2007.
 P. J. Besl and N. D. McKay, “Method for registration of 3-D shapes,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 239-256, 1992.
 Z. Zhang, “Iterative point matching for registration of free-form curves and surfaces,” International Journal of Computer Vision, vol. 13, pp. 119-152, 1994.
 P. E. Debevec, C. J. Taylor, and J. Malik, “Modeling and rendering architecture from photographs: A hybrid geometry-and image-based approach,” in Proceedings of the 23rd Annual Conference on Computer Graphics and Interactive Techniques. ACM, pp. 11-20, 1996.
 C. J. Watkins and P. Dayan, “Q-learning,” Machine Learning, vol. 8, no. 3-4, pp. 279-292, 1992.
 L. P. Kaelbling, M. L. Littman, and A. W. Moore, “Reinforcement learning: A survey,” J. Artif. Intell. Res., vol. 4, pp. 237-285, 1996.
 R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction. Cambridge, MA: MIT Press, 1998.
 E. Alpaydin, Introduction to Machine Learning. Cambridge, MA: MIT Press, 2004.