First Results on Stable Adaptive Robot Control with RBF Networks
We present a method to use RBF networks for stable adaptive robot control, with the task to track prescribed joint trajectories. The controller architecture is based on the concept of feedback linearization. The RBF networks serve as the model-based controller part. The learning rule is derived from a Lyapunov-based stability criterion to assure global stability at all times. Bounds on network approximation errors, which are essential to validate the stability proof, are established with the help of multidimensional sampling theory. Training the networks with point-to-point movements between randomly generated locations leads to a trajectory independent controller over the complete working range. Simulation results for a planar 4-joint manipulator (4JM) are given to demonstrate the performance of this control method.
KeywordsGlobal Stability MIMO System Radial Basis Function Network Feedback Linearization Stability Proof
Unable to display preview. Download preview PDF.
- M. Kawato. Optimization and learning in neural networks for formation and control of coordinated movement. ATR Technical Report, 1990.Google Scholar
- M. Jordan, D. Rumelhart. Forward models: supervised learning with a distal teacher. Center for Cognitive Science, Technical Report (Occ. paper) 40, 1990.Google Scholar
- R.M. Sanner, J.J. Slotine. Direct adaptive control using gaussian networks. Technical Report NSL-910303, 1991.Google Scholar
- E. Tzirkel-Hancock, F. Fallside. Stable control of nonlinear systems using neural networks. Technical Report 81, Cambridge University, Cambridge, 1991.Google Scholar