Fundamental Limits of Performance for Force Reflecting Teleoperation

Abstract

The quality of telepresence provided by a force-reflecting teleopera tor is determined, for the most part, by the fidelity of the contact-force information fed back to the operator. These fed-back forces, how ever, also directly influence system stability, and in this paper we investigate the relationship between fidelity and stability with a view toward understanding how stability considerations impose funda mental limits on system performance. The key idea of our work is to draw an explicit distinction between the information conveyed by the force signal and the energy inherent in that signal. Using known physiological properties of the operator, we argue that there exists a natural partitioning between information and energy wherein in formation is conveyed at frequencies above roughly 30 Hz, while the energetic interaction between the slave and the environment takes place at frequencies below this. We embody this distinction in a two-channel framework that we claim provides insight into the de sign of force-reflecting systems. Using a 1-DOF model, we study the effect of various system characteristics, notably mass, stiffness, and damping properties, on performance and stability. This model is used to derive expressions for the maximum force-reflection ratio that guarantees stability against pure-stiffness environments and to investigate the role of various compensation elements, including lo cal force control around the slave. Finally, a framework is developed for force-reflecting teleoperation that maximizes the force informa tion conveyed to the operator, subject to the constraints imposed by stability considerations.