Multilayer neural-net robot controller with guaranteed tracking performance
- 1 March 1996
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks
- Vol. 7 (2) , 388-399
- https://doi.org/10.1109/72.485674
Abstract
A multilayer neural-net (NN) controller for a general serial-link rigid robot arm is developed. The structure of the NN controller is derived using a filtered error/passivity approach. No off-line learning phase is needed for the proposed NN controller and the weights are easily initialized. The nonlinear nature of the NN, plus NN functional reconstruction inaccuracies and robot disturbances, mean that the standard delta rule using backpropagation tuning does not suffice for closed-loop dynamic control. Novel online weight tuning algorithms, including correction terms to the delta rule plus an added robust signal, guarantee bounded tracking errors as well as bounded NN weights. Specific bounds are determined, and the tracking error bound can be made arbitrarily small by increasing a certain feedback gain. The correction terms involve a second-order forward-propagated wave in the backpropagation network. New NN properties including the notions of a passive NN, a dissipative NN, and a robust NN are introduced.Keywords
This publication has 24 references indexed in Scilit:
- Adaptive control of unknown plants using dynamical neural networksIEEE Transactions on Systems, Man, and Cybernetics, 1994
- Adaptive control of non-linear continuous-time systems using neural networks—general relative degree and MIMO casesInternational Journal of Control, 1993
- Direct control and coordination using neural networksIEEE Transactions on Systems, Man, and Cybernetics, 1993
- Uniqueness of the weights for minimal feedforward nets with a given input-output mapNeural Networks, 1992
- Adaptive control of nonlinear systems using neural networksInternational Journal of Control, 1992
- Neural network controller characteristics with regard to adaptive controlIEEE Transactions on Systems, Man, and Cybernetics, 1992
- A nonlinear regulator design in the presence of system uncertainties using multilayered neural networkIEEE Transactions on Neural Networks, 1991
- Universal Approximation Using Radial-Basis-Function NetworksNeural Computation, 1991
- Adaptive manipulator control: A case studyIEEE Transactions on Automatic Control, 1988
- A new adaptive law for robust adaptation without persistent excitationIEEE Transactions on Automatic Control, 1987