Stable training of computationally intelligent systems by using variable structure systems technique

This paper presents a novel training algorithm for computationally intelligent architectures, whose outputs are differentiable with respect to the adjustable design parameters. The algorithm combines the gradient descent technique with the variable-structure-systems approach. The combination is performed by expressing the conventional weight update rule in continuous time and application of sliding-mode control method to the gradient-based training procedure. The proposed combination therefore exhibits a degree of robustness with respect to the unmodeled multivariable internal dynamics of gradient descent and to the environmental disturbances. With conventional training procedures, the excitation of this dynamics during a training cycle can lead to instability, which may be difficult to alleviate due to the multidimensionality of the solution space and the ambiguities on the free design parameters, such as learning rate or momentum coefficient. This paper develops a heuristic that a computationally intelligent system, which may be a neural network architecture or a fuzzy inference mechanism, can be trained such that the adjustable parameter values are forced to settle down (parameter stabilization) while minimizing an appropriate cost function (cost optimization). The proposed approach is applied to the control of a robotic arm in two different ways. In one, a standard fuzzy system architecture is used, whereas in the second, the arm is controlled by the use of a multilayer perceptron. In order to demonstrate the robustness of the approach presented, a considerable amount of observation noise and varying payload conditions are also studied.

[1]  Madan M. Gupta,et al.  Dynamic Neural Units with Applications to the Control of Unknown Nonlinear Systems , 1993, J. Intell. Fuzzy Syst..

[2]  Li-Xin Wang,et al.  A Course In Fuzzy Systems and Control , 1996 .

[3]  Okyay Kaynak,et al.  A comparative study of soft-computing methodologies in identification of robotic manipulators , 2000, Robotics Auton. Syst..

[4]  Chai Tianyou,et al.  Adaptive fuzzy sliding mode control for nonlinear systems , 1996, Proceedings of IEEE 5th International Fuzzy Systems.

[5]  Ken-ichi Funahashi,et al.  On the approximate realization of continuous mappings by neural networks , 1989, Neural Networks.

[6]  John Y. Hung,et al.  Variable structure control: a survey , 1993, IEEE Trans. Ind. Electron..

[7]  Okyay Kaynak,et al.  Fuzzy adaptive sliding mode control of a direct drive robot , 1996, Robotics Auton. Syst..

[8]  Okyay Kaynak,et al.  A comparative study of neural network structures in identification of nonlinear systems , 1999 .

[9]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[10]  Simon Parsons,et al.  Soft computing: fuzzy logic, neural networks and distributed artificial intelligence by F. Aminzadeh and M. Jamshidi (Eds.), PTR Prentice Hall, Englewood Cliffs, NJ, pp 301, ISBN 0-13-146234-2 , 1996, Knowl. Eng. Rev..

[11]  Weibing Gao,et al.  Variable structure control of nonlinear systems: a new approach , 1993, IEEE Trans. Ind. Electron..

[12]  Stanislav V. Emelyanov,et al.  Variable-Structure Control Systems , 1995 .

[13]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[14]  José Santos-Victor Vision-based remote control of cellular robots , 1998, Robotics Auton. Syst..

[15]  Richard D. Braatz,et al.  On the "Identification and control of dynamical systems using neural networks" , 1997, IEEE Trans. Neural Networks.

[16]  Woonchul Ham,et al.  Adaptive fuzzy sliding mode control of nonlinear system , 1998, IEEE Trans. Fuzzy Syst..

[17]  Li-Xin Wang,et al.  Adaptive fuzzy systems and control - design and stability analysis , 1994 .

[18]  Li-Xin Wang,et al.  Adaptive fuzzy systems and control , 1994 .

[19]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[20]  Hideki Hashimoto,et al.  Variable structure systems theory applied to sub-time optimal position control with an invariant trajectory. , 1984 .