Learning of Obstacle Avoidance with Redundant Manipulator by Hierarchical SOM
Yuichi Kobayashi and Takahiro Nomura
Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho, Koganei, Tokyo 184-8588, Japan
This paper proposes a method of obstacle avoidance motion generation for a redundant manipulator with a Self-OrganizingMap (SOM) and reinforcement learning. To consider redundancy, two types of SOMs – a hand position map and a joint angle map – are combined. Multiple joint angles corresponding to the same hand position are memorized in the proposed map. Preserved redundant configuration information is used to generate motions based on tasks and situations, while resolving inverse kinematics problems with a redundant manipulator. The proposed map is applied to planning motion control using reinforcement learning in an unknown environment, where collision with obstacles is detected only directly by tactile sensing. The feasibility of the proposed framework was verified by simulation and experiments with an arm robot with force and a vision sensors.
-  Z. W. Luo and M. Ito, “Diffusion-based learning theory for organizing visuo-motor coordination,” Biological Cybernetics, Vol.79, pp. 279-289, 1998.
-  K. Ueno, K. Horio, T. Yamakawa, and K. Ishii, “A method to solve inverse kinematics of redundant manipulator using topology representing network,” Proc. of SICE Annual Conf. 2005, pp. 2865-2870, 2005.
-  P. Jorg and A. Walter, “PSOM network: Learning with few examples,” Proc. of Int. Conf. on Robotics and Automation, pp. 2054-2059, 1998.
-  Y. Zhang and J. Wang, “Obstacle Avoidance for Kinematically Redundant Manipulators Using a Dual Neural Network,” IEEE Trans. on Systems, Man, and Cybernetics (Part B), Vol.34, No.1, pp. 752-759, 2004.
-  S. Liu and J.Wang, “Obstacle Avoidance for Kinematically Redundant Manipulators Using the Deterministic Annealing Neural Network,” Advances in Neural Networks – ISNN 2005 Lecture Notes in Computer Science, 2005, Vol.3498, pp. 240-246, 2005,
-  A. Hayashi and B. J. Kuipers, “Path planning for highly redundant manipulators using a continuous model,” Proc. of the National Conf. on Artificial Intelligence (AAAI-91), pp. 666-672, 1991.
-  R. Sutton and A. Barto, “Reinforcement Learning,” MIT Press, 1998.
-  H. Kimura, T. Yamashita, and S. Kobayashi, “Reinforcement Learning of Walking Behavior for a Four-Legged Robot,” Proc. of IEEE Conf. on Decision and Control, pp. 411-416, 2001.
-  J. Morimoto and K. Doya, “Acquisition of stand-up behavior by a real robot using hierarchical reinforcement learning,” Robotics and Autonomous Systems Vol.36, No.1, pp. 37-51, 2001.
-  H. Miyamoto, J. Morimoto, K. Doya, and M. Kawato, “Reinforcement learning with via-point representation,” Neural Networks, Vol.17, No.3, pp. 299-305, 2004.
-  K. Ito and F. Matsuno, “Control of hyper-redundant robot using QDESEGA,” Proc. of SICE Annual Conf. 2002, pp. 1677-1682, 2002.
-  R. Miikulainen, “Script Recognition with Hierarchical Feature Maps,” Connection Science Vol.2, pp. 83-101, 1990.
-  M. Asada, K. MacDorman, H. Ishiguro, and Y. Kuniyoshi, “Cognitive developmental robotics as a new paradigm for the design of humanoid robots,” Robotics and Autonomous Systems, Vol.37, pp. 185-193, 2001.
-  A. Stoytchev, “Some basic principles of developmental robotics,” IEEE Trans. on Autonomous Mental Development, Vol.1, No.2, pp. 122-130, 2009.
-  T. Okamoto, Y. Kobayashi, and M. Onishi, “Acquisition of Body and Object Representation Based on Motion Learning and Planning Framework,” Proc. of the 9th Int. Conf. on Intelligent Systems Design and Applications, pp. 737-742, 2009.
-  T. Asamizu and Y. Kobayashi, “Acquisition of image feature on collision for robot motion generation,” Proc. 9th Int. Conf. on Intelligent Systems Design and Applications, pp. 1312-1317, 2009.
-  Y. Kobayashi, M. Shibata, S. Hosoe, and Y. Uno, “Learning of object manipulation with stick/slip mode switching,” Proc. of Int. Conf. on Intelligent Robots and Systems, pp. 373-379, 2008.