Functional adaptive controller for multivariable stochastic systems with dynamic structure of neural network

Abstract

The article deals with a challenging problem of adaptive control design for multivariable stochastic systems with a functional uncertainty. Model of the system is based on multi-layered perceptron neural networks where both the unknown parameters and the structure are found in real time without a necessity of any off-line training process. The unknown parameters are estimated by a global estimation method, the Gaussian sum filter, and the structure of the neural network model is optimized by a proposed pruning method. The control law is based on a bicriterial approach to the suboptimal dual control. Two individual criteria are designed and used to introduce conflicting efforts between the estimation and control; probing and caution. A comparison of the proposed dual control and its alternative with an implementation of the pruning algorithm is shown in a numerical example.

Functional Dual Adaptive Control with Recursive Gaussian Process Model

“…In recent years, several sub-optimal dual controller methods have been applied successfully in the functional approach for their positive qualities (superior control quality and admissible computational demands) and subsequently extended in several directions [7][8][9][10][11]. Although a partial progress was achieved in a functional adaptive control, one of the open and challenging issues remains; namely, finding of a suitable model of the controlled non-linear system.…”

Gaussian process based dual adaptive control of nonlinear stochastic systems

“…nonholonomic wheeled mobile robot [11] or an extension to a more general MIMO class of nonlinear systems [12], [13].…”

“…Hence, the general goal of the paper is to design a novel dual adaptive control for nonlinear stochastic systems approximated by a non-parametric GP model. The paper differs from [9] and [12] in two main aspects: (a) a parametric neural network model is replaced by a nonparametric GP model (b) The bicriterial control is modified to respect the uncertainty of the prediction provided by the GP model. From another point of view, the paper presents an extension of the caution adaptive controller based on GP models developed in [3] by the property of duality.…”