Fuzzy Control for Robot Manipulators with Artificial Rubber Muscles

Abstract

When controlling a robot manipulator by aplying the well-known computed torque control law, we usually need the mathematically accurate model. However, it is not necessarily easy to make a rigorous model, because the dynamic modeling of the robot manipulator essentially include some uncertainties such as system pa- rameters, disturbance inputs and nonlinear elements. Thus we can be faced with the problem of designing a robust controller that achieves a stable control, even through we use inaccurate modeling information. Some different approaches to the problem of robust control design for uncertain systems have beem proposed for a case when the bounds on the uncertainties are known (Hui and Ẓak, 1992); two major approaches to the deterministic control of uncertain systems are the deterministic control using Lyapunov functions (Corless and Leitmann, 1981; Coreless, 1989) and the variable structure control (or sliding mode control) methods (Utkins, 1977; De- Carlo et al. The approached without using the bound information on the uncertainties can also be found in Chen (1990) or Imura et al. (1991).

A fuzzy control is attracted as as practical control strategy for controlling a robot manipulator, because we need to mathematical models and can easily apply it to nonlinear systems as well as linear systems. it should be noted, that, when designing a fuzzy logic controller, we must tune some controller parameters such as scalers (or gains) for the input data and the output of the fuzzy logic controller. For this problem, some self-organizing fuzzy controllers (SOFCs) have been already examined by several authors (Procyk and Mamdani, 1979; Yamazaki and Sugeno, 1984; Daley and Gill; 1986; Linkens and Hasnain, 1991; Tanji and Ki- noshita, 1987; Maeda and Murakami, 1988). On the other hand, some (iterative or repeated) learning-type fuzzy controller (LFCs) that incorporated a neural network have been also reported recently (Watanabe and Ichihashi, 1990; Hayashi et al., 1990; Horikawa, et al., 1991; Watanabe et al., 1992; Watanabe and Tang, 1992).