Position drift compensation in time domain passivity based teleoperation

Abstract

The Time Domain Passivity Control Approach is gathering interest in the robotics field. Simplicity and flexibility and the fact that system design emerges from ideal cases make it a powerful stability tool for teleoperation systems. Communication time delay is an inherent attribute of nearly every realistic teleoperation system. Unless the communication channel guarantees transmission delays of less than the system sampling time, the delay must be considered in the design in order to guarantee stability and satisfy a desired degree of performance. In previous work it has been shown how passivity can be considered in the time domain and how control rules are derived from it in order to dissipate the energy produced by the delayed communication. However, a weakness of these approaches is the impossibility of observing the exact amount of energy stored in the communication channel due to its delayed nature. A passive estimation is therefore needed which outcomes in an over-dissipation and in turn impacts on transparency. In constrained communications over-dissipation may become apparent in the form of a non-neglectful position drift between master and slave. This paper tackles the over-dissipative behavior of the Passivity Controller by resembling the energetic behavior of an ideal communication, i.e. where no delay is present and the transmission is lossless. Thus, the communication channel is not just controlled to be passive, as has been the case up to now, but also lossless. Energy can be dissipated to prevent activity, but activity can be also produced to prevent dissipative behaviors. The approach is sustained with experimental results.