A Self-Organizing Neural Model for Fault-Tolerant Control of Redundant Robots

This paper describes a self-organizing neural model that is capable of controlling the kinematics of robots with redundant degrees of freedom. The self-organized learning process is based on action perception cycles where the robot is perturbed minimally about a given joint configuration and learns to map these perturbations to changes in sensor readings corresponding to these minimal perturbations. This motor babbling phase provides self-generated movement commands that activate correlated sensory, spatial and motor information that are used to learn an internal coordinate transformation between sensory and motor systems. This idea was tested on two different tasks: reaching targets in 2-D space with a three degree of freedom robot and saccading to targets in 3-D with a twelve degree of freedom head-neck-eye system. Computer simulations show that the resulting controller is highly fault-tolerant and robust to previously unseen disturbances much like biological systems.

[1]  Andrew G. Barto,et al.  Reinforcement learning , 1998 .

[2]  P. L. Adams.  THE ORIGINS OF INTELLIGENCE IN CHILDREN , 1976 .

[3]  Stephen Grossberg,et al.  Adaptive neural networks for control of movement trajectories invariant under speed and force rescaling , 1991 .

[4]  Narayan Srinivasa,et al.  Execution of saccades for active vision using a neurocontroller , 1997 .

[5]  J. C. Fiala.  A network for learning kinematics with application to human reaching models , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[6]  David E. Orin,et al.  Efficient Dynamic Computer Simulation of Robotic Mechanisms , 1982 .

[7]  Andrew G. Barto,et al.  Reinforcement Learning and Dynamic Programming , 1995 .

[8]  S. Grossberg,et al.  Neural representations for sensory-motor control, I: Head-centered 3-D target positions from opponent eye commands. , 1993, Acta psychologica.

[9]  Stephen Grossberg,et al.  A Neural Theory of Punishment and Avoidance , II : Quantitative Theory , 2003 .

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

[11]  Stephen Grossberg,et al.  Vector associative maps: Unsupervised real-time error-based learning and control of movement trajectories , 1991, Neural Networks.

[12]  Stephen Grossberg,et al.  Neural representations for sensory-motor control, II: Learning a head-centered visuomotor representation of 3-D target position , 1993, Neural Networks.

[13]  Narayan Srinivasa,et al.  Efficient Learning of VAM-Based Representation of 3D Targets and its Active Vision Applications , 1998, Neural Networks.

[14]  Narendra Ahuja,et al.  A Learning Approach to Fixating on 3D Targets with Active Cameras , 1998, ACCV.

[15]  Daniel Bullock,et al.  Neural Representations for Sensory-Motor Control, III: Learning a Body-Centered Representation of 3-D Target Position , 1993 .

[16]  S. Grossberg,et al.  A Self-Organizing Neural Model of Motor Equivalent Reaching and Tool Use by a Multijoint Arm , 1993, Journal of Cognitive Neuroscience.

[17]  Bruno Siciliano,et al.  Kinematic control of redundant robot manipulators: A tutorial , 1990, J. Intell. Robotic Syst..

[18]  S. Grossberg,et al.  Neural Representations for Sensorimotor Control. III. Learning a Body-Centered Representation of a Three-Dimensional Target Position , 1994, Journal of Cognitive Neuroscience.