A mechatronic analysis of the classical position-force controller based on bounded environment passivity

Abstract

Bounded Environment Passivity, presented in this paper, allows one to design teleoperation systems that behave passively provided that the environment with which interaction takes place belongs to an a priori defined range of environments. The use of such a priori knowledge on the environment reduces conservativeness with respect to classical design approaches. An additional advantage lies in its capability to get a clearer insight on which type of environments are problematic for the specific controller under investigation. On the basis of a case study, i.e. the well-known Position-Force controller, this paper describes and compares different passivity-based methods. First, the traditional methods of two-port passivity and absolute stability are applied. The restrictions of these methods to come up with useful design rules are explicitly demonstrated. Second, the Bounded Environment Passivity conditions of the Position-Force controller are derived. These conditions describe the relation between the specific controller implementation, the teleoperator dynamics and the environment characteristics. In addition, the effects of structural resonance frequencies and low-pass filters, often present in realistic teleoperator setups, are described. This analysis reveals fundamental mechatronic rules of thumb for the design of a teleoperator system with a Position-Force control architecture. The theoretical results are verified experimentally on a one-degree-of-freedom teleoperation system.