Proposed neural network structures

Abstract

In this chapter, four neural network-based control schemes were proposed for tip position tracking of a flexible manipulator. The first two schemes were developed by using a modified version of the “feedback-error learning” approach to learn the inverse dynamics of the flexible manipulator. Both schemes assume some a priori knowledge of the linear model of the system. This assumption was relaxed in the third and fourth schemes. In the third scheme, the controller was designed based on tracking the hub position while controlling the elastic deflection at the tip. The fourth scheme employed two neural networks, one of the neural networks defines an appropriate output for feedback and the other neural network acts as an inverse dynamics controller. Simulation results for two single flexible-link manipulators and a two-link manipulator were presented to illustrate the advantages and improved performance of the proposed tip position tracking controllers over the conventional PD-type controllers.