Skip to Main Content
The purpose of this paper is to demonstrate the advantages of hybrid, model-matching adaptive control schemes over similar discrete schemes for control of robotic manipulators. In the hybrid approach, the controller itself is a continuously adjustable one, but the adaptation mechanism uses discrete algorithms. Thus, it retains the simplicity and efficiency of discrete parameter estimation techniques and requires estimation of fewer parameters. Furthermore, it retains the minimum phase property of the original continuous time plant and thus avoids any potential problem of instability.