Modelling and Intelligent Control of an Elastic Link Robot Manipulator

Abstract

In this paper, precise control of the end-point position of a planar single-link elastic manipulator robot is discussed. The Timoshenko beam theory (TBT) has been used to characterize the structural link elasticity including important damping mechanisms. A suitable nonlinear model is derived based on the Lagrangian assumed modes method. Elastic link manipulators are classified as systems possessing highly complex dynamics. In addition, the environment in which they operate may have a lot of disturbances. These give rise to special problems that may be solved using intelligent control techniques. The application of two advanced control strategies based on fuzzy set theory is investigated. The first closed-loop control scheme to be applied is the standard Proportional-Derivative (PD) type fuzzy logic controller (FLC), also known as PD-type Mamdani's FLC (MPDFLC). Then, a genetic algorithm (GA) is used to optimize the MPDFLC parameters with innovative tuning procedures. Both the MPDFLC and the GA optimized FLC (GAOFLC) are implemented and tested to achieve a precise control of the manipulator end-point. The performances of the adopted closed-loop intelligent control strategies are examined via simulation experiments.