Single net indirect learning architecture

Abstract

This paper presents a novel indirect learning architecture which uses a single neural network in implementation. The new architecture generates the error signal required in training the controller network by an innovative design using a memory element and few switches. The new controller needs only half the number of neurons and connection weights in comparison with the original indirect learning architecture. Also given are the simulation results in controlling a nonlinear plant.

A new approach to adaptive fuzzy control: the controller output error method

“…In the context of fuzzy control, this implies that the control engineer is required to take the usual steps of fuzzy controller design and verification (see Section III) before employing COEM to tune the parameters of the fuzzy controller. A concept similar to COEM has been described previously by Psaltis, Sideris, and Yamamura [3] and Andersen, Teng, and Tsoi [7] who applied it to neural network controllers. In this context the name, "indirect learning" was used.…”

A fuzzy controller with evolving structure

“…; TD; where TD is the number of training data samples). As it is known, however, the learning techniques for NN are iterative and, therefore training of the neuro-controller in [1] and [15] has been done offline. Additional important disadvantage of neural networks in comparison to the fuzzy rule-based models is their lack of interpretability and transparency [11,12].…”

“…In the original works on IL-based control [1,15] the controller has been realised as a neural network trained off-line using the sets of data (y k , y kþ1 , u k ; k ¼ 1; 2; . .…”

“…In the present paper, this approach is applied to on-line design of fuzzy controllers combined with the indirect learning technique, proposed in [15] and developed in [1]. Instead of a neural network trained off-line as in [1,15] we use evolving rule-based models [6] to design the controller structure on-line by adding more informative rules than those, that previously formed the model.…”

“…Instead of a neural network trained off-line as in [1,15] we use evolving rule-based models [6] to design the controller structure on-line by adding more informative rules than those, that previously formed the model. In a driven evolution of the controller structure the existing rules are replaced with new ones based on the informative potential of the data samples, measured by their spatial proximity to the potential rules centers.…”