Canonical parameterization of excess motor degrees of freedom with self-organizing maps

The problem of sensorimotor control is underdetermined due to excess (or "redundant") degrees of freedom when there are more joint variables than the minimum needed for positioning an end-effector. A method is presented for solving the nonlinear inverse kinematics problem for a redundant manipulator by learning a natural parameterization of the inverse solution manifolds with self-organizing maps. The parameterization approximates the topological structure of the joint space, which is that of a fiber bundle. The fibers represent the "self-motion manifolds" along which the manipulator can change configuration while keeping the end-effector at a fixed location. The method is demonstrated for the case of the redundant planar manipulator. Data samples along the self-motion manifolds are selected from a large set of measured input-output data. This is done by taking points in the joint space corresponding to end-effector locations near "query points", which define small neighborhoods in the end-effector work space. Self-organizing maps are used to construct an approximate parameterization of each manifold which is consistent for all of the query points. The resulting parameterization is used to augment the overall kinematics map so that it is locally invertible. Joint-angle and end-effector position data, along with the learned parameterizations, are used to train neural networks to approximate direct inverse functions.

[1]  Richard Durbin,et al.  An analogue approach to the travelling salesman problem using an elastic net method , 1987, Nature.

[2]  J. Burdick.  On the inverse kinematics of redundant manipulators: characterization of the self-motion manifolds , 1989 .

[3]  Yong-San Yoon,et al.  Practical inverse kinematics of a kinematically redundant robot using a neural network , 1991, Adv. Robotics.

[4]  Jean-Jacques E. Slotine,et al.  Robot analysis and control , 1988, Autom..

[5]  Klaus Schulten,et al.  Implementation of self-organizing neural networks for visuo-motor control of an industrial robot , 1993, IEEE Trans. Neural Networks.

[6]  Charles W. Wampler,et al.  On the Inverse Kinematics of Redundant Manipulators , 1988, Int. J. Robotics Res..

[7]  Charles A. Klein,et al.  Review of pseudoinverse control for use with kinematically redundant manipulators , 1983, IEEE Transactions on Systems, Man, and Cybernetics.

[8]  R. M. Kil,et al.  Bidirectional continuous associator based on Gaussian potential function network , 1989, International 1989 Joint Conference on Neural Networks.

[9]  Tsuneo Yoshikawa,et al.  Manipulability of Robotic Mechanisms , 1985 .

[10]  F. Pourboghrat,et al.  Neural networks for learning inverse kinematics of redundant manipulators , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.

[11]  C. Gielen,et al.  Neural computation and self-organizing maps, an introduction , 1993 .

[12]  Allon Guez,et al.  On the solution to the inverse kinematic problem , 1990, Proceedings., IEEE International Conference on Robotics and Automation.

[13]  Sukhan Lee,et al.  Applications of Neural Networks to Robotics , 1991 .

[14]  Homayoun Seraji,et al.  Configuration control of redundant manipulators: theory and implementation , 1989, IEEE Trans. Robotics Autom..

[15]  R. Eckmiller,et al.  Neural kinematics net for a redundant robot arm , 1989, International 1989 Joint Conference on Neural Networks.

[16]  R. Palmer,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[17]  Jorma Laaksonen,et al.  Variants of self-organizing maps , 1990, International 1989 Joint Conference on Neural Networks.

[18]  D. Kohli,et al.  The Jacobian analysis of workspaces of mechanical manipulators , 1987 .

[19]  Klaus Schulten,et al.  Topology-conserving maps for learning visuo-motor-coordination , 1989, Neural Networks.

[20]  Charles W. Wampler.  Inverse kinematic functions for redundant spherical wrists , 1989, IEEE Trans. Robotics Autom..

[21]  Kenneth Kreutz-Delgado,et al.  Global Regularization of Inverse Kinematics for Redundant Manipulators , 1992, NIPS.

[22]  Daniel E. Whitney,et al.  Resolved Motion Rate Control of Manipulators and Human Prostheses , 1969 .

[23]  Alexander Linden,et al.  Inversion of neural networks by gradient descent , 1990, Parallel Comput..

[24]  N. Steenrod.  The Topology of Fibre Bundles. (PMS-14) , 1951 .

[25]  Sukhan Lee,et al.  Robot kinematic control based on bidirectional mapping neural network , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[26]  Charles W. Wampler.  The Inverse Function Approach to Kinematic Control of Redundant Manipulators , 1988, 1988 American Control Conference.

[27]  Joel W. Burdick,et al.  On the inverse kinematics of redundant manipulators: characterization of the self-motion manifolds , 1989, Proceedings, 1989 International Conference on Robotics and Automation.

[28]  John Baillieul,et al.  Avoiding obstacles and resolving kinematic redundancy , 1986, Proceedings. 1986 IEEE International Conference on Robotics and Automation.

[29]  Helge J. Ritter,et al.  Three-dimensional neural net for learning visuomotor coordination of a robot arm , 1990, IEEE Trans. Neural Networks.

[30]  N. Steenrod.  Topology of Fibre Bundles , 1951 .

[31]  Siddhartha Sen,et al.  Topology and geometry for physicists , 1983 .

[32]  Franklin A. Graybill,et al.  Introduction to The theory , 1974 .

[33]  Bruno Siciliano,et al.  Closed-Loop Inverse Kinematics Schemes for Constrained Redundant Manipulators with Task Space Augmentation and Task Priority Strategy , 1991, Int. J. Robotics Res..

[34]  Daniel R. Baker.  Some topological problems in robotics , 1990 .

[35]  M. Golubitsky,et al.  Stable mappings and their singularities , 1973 .

[36]  John Baillieul,et al.  Resolution of Kinematic Redundancy using Optimization Techniques , 1988, 1988 American Control Conference.

[37]  Yoshihiko Nakamura,et al.  Optimal Redundancy Control of Robot Manipulators , 1987 .

[38]  Kenneth Kreutz-Delgado,et al.  Issues in learning global properties of the robot kinematic mapping , 1993, [1993] Proceedings IEEE International Conference on Robotics and Automation.

[39]  John Baillieul,et al.  Kinematic programming alternatives for redundant manipulators , 1985, Proceedings. 1985 IEEE International Conference on Robotics and Automation.

[40]  Roman Bek,et al.  Discourse on one way in which a quantum-mechanics language on the classical logical base can be built up , 1978, Kybernetika.

[41]  Michael I. Jordan,et al.  Forward Models: Supervised Learning with a Distal Teacher , 1992, Cogn. Sci..

[42]  Kenneth Kreutz-Delgado,et al.  Learning Global Properties of Nonredundant Kinematic Mappings , 1998, Int. J. Robotics Res..