Abstract
A multilayer neural net (NN) controller for a general serial-link robot arm is developed. The structure of the NN controller is derived using a filtered error approach. It is argued that standard backpropagation tuning, when used for real-time closed-loop control, can yield unbounded NN weights if: (1) the net can not exactly reconstruct a certain required control function, (2) there are bounded unknown disturbances in the robot dynamics, or (3) the robot arm has more than one link (i.e. nonlinear case). On-line weight tuning algorithms including correction terms to backpropagation, plus an added robustifying signal, guarantee tracking as well as bounded weights. The correction terms involve a second-orderforward-propagated wave in the backprop network.
Similar content being viewed by others
References
Alcorta, E. and Sánchez, E. N.: On-line Function Approximation via Neural Networks, Preprint, 1992.
Bartle, R. G.:The Elements of Real Analysis, Wiley, New York, 1964.
Chu, S. R. and Shoureshi, R.: Neural-based adaptive nonlinear system identification,Intelligent Control Systems DSC-45, ASME Winter Annual Meeting, 1992.
Craig, J. J.:Adaptive Control of Robot Manipulators, Addison-Wesley, Reading, MA, 1988.
Cybenko, G.: Approximation by superpositions of a sigmoidal functions,Math. Control, Signals, Systems 2(4) (1989), 303–314.
Dawson, D. M., Qu, Z., Lewis, F. L. and Dorsey, J. F.: Robust control for the tracking of robot motion,Int. J. Control 52(3) 19??, 581–595.
Goodwin, G. C.: Can we identify adaptive control?Proc. Eur. Control Conf. July 1991, pp. 1714–1725.
Goodwin, G. C. and Sin, K. S.:Adaptive Filtering, Prediction, and Control, Prentice-Hall, New Jersey, 1984.
Horn, B., Hush, D. and Abdallah, C.: The state space recurrent neural network for robot identification,Adv. Control Issues for Robot Manipulators, DSC-39, ASME Winter Annual Meeting, 1992.
Hornik, K., Stinchombe, M. and White, H.: Multilayer feedforward networks are universal approximators,Neural Networks 2 (1989), 359–366.
Iiguni, Y., Sakai, H. and Tokumaru, H.: A nonlinear regulator design in the presence of system uncertaities using multilayer neural networks,IEEE Trans. Neural Networks 2(4) (1991), 410–417.
Kraft, L. G. and Campagna, D. P.: A summary comparison of CMAC neural network and traditional adaptive control systems, in W. T. Miller, R. S. Sutton and P. J. Werbos (eds),Neural Networks for Control, MIT Press, Cambridge, 1991, pp. 143–169.
Landau, Y. D.:Adaptive Control:The Model Reference Approach, Marcel Dekker, New York, 1979.
Lewis, F. L., Abdallah, C. T. and Dawson, D. M.:Control of Robot Manipulators, Macmillan, New York, 1993.
Lewis, F. L., Liu, K. and Yesildirek, A.: Neural net robot controller with guaranteed performance, 1993, in press.
Liu, K. and Lewis, F. L.: Robust control techniques for general dynamic systems,J. Intel. Robotic Syst. 6 (1992), 33–49.
Miller, W. T., Sutton, R. S. and Werbos, P. J. (eds):Neural Networks for Control, MIT Press, Cambridge, 1991.
Narendra, K. S.: Adaptive control using neural networks, in W. T. Miller, R. S. Sutton and P. J. Werbos (eds),Neural Networks for Control, MIT Press, Cambridge, 1991, pp. 115–142.
Narendra, K. S. and Annaswamy, A. M.: A new adaptive law for robust adaptation without persistent excitation,IEEE Trans. Automat. Control AC-32(2) (1987), 134–145.
Narendra, K. S. and Parthasarathy, K.: Identification and control of dynamical systems using neural networks,IEEE Trans. Neural Networks 1 (Mar. 1990), 4–27.
Olvera, J., Guan, X. and Manry, M. T.: Theory of Monomial Networks,Proc. Symp. Implicit and Nonlinear Systems, Dec. 1992, pp. 96–101.
Ozaki, T., Suzuki, T., Furuhashi, T., Okuma, S. and Uchikawa, Y.: Trajectory control of robotic manipulators,IEEE Trans. Ind. Elec. 38 (June 1991), 195–202.
Pao, Y. H.:Adaptive Pattern Recognition and Neural Networks, Addison-Wesley, Reading, MA, 1989.
Park, J. and Sandberg, I. W.: Universal approximation using radial-basis-function networks,Neural Comp. 3 (1991), 246–257.
Polycarpou, M. M. and Ioannu, P. A.: Identification and control using neural network models: design and stability analysis, Tech. Report 91-09-01, Dept. Elect. Eng. Sys., Univ. S. Cal., Sep. 1991.
Polycarpou, M. M. and Ioannu, P. A.: Neural networks as on-line approximators of nonlinear systems,Proc. IEEE Conf. Decision and Control, Tuscon, Dec. 1992, pp. 7–12.
Poggio, T. and Girosi, F.: A theory of networks with for approximation and learning, Tech. Report Memo No 1140, Artif. Intel. Lab, MIT, 1989.
Sadegh, N.: Nonlinear identification and control via neural networks,Control Systems with Inexact Dynamics Models,DSC-33, ASME Winter Annual Meeting, 1991.
Sanner, R. M. and Slotine, J.-J. E.: Stable adaptive control and recursive identification using radial gaussian networks,Proc. IEEE Conf. Decision and Control, Brighton, 1991.
Sastry, S. S. and Bodson, M.:Adaptive Control, Prentice-Hall, New Jersey, 1989.
Slotine, J.-J. E. and Li, W.: Adaptive manipulator control: a case study,IEEE Trans. Automat. Control 33(11) (1988), 995–1003.
Slotine, J.-J. E. and Li, W.:Applied Nonlinear Control, Prentice-Hall, New Jersey, 1991.
Yabuta, T. and Yamada, T.: Neural network controller characteristics with regard to adaptive control,IEEE Trans. Syst., Man. Cybern. 22(1) (1992), 170–176.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Lewis, F.L., Yeşildirek, A. & Liu, K. Neural net robot controller: Structure and stability proofs. J Intell Robot Syst 12, 277–299 (1995). https://doi.org/10.1007/BF01262965
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF01262965