Position tracking in delayed bilateral teleoperators without velocity measurements

Abstract

SUMMARYThis work considers the control of nonlinear bilateral teleoperators with variable time-delays without the need of velocity measurements. The recently proposed Immersion and Invariance (I&I) observer is used to obtain an exponentially convergent estimate of the unmeasured velocities. Under the classical assumption that the human operator and the environment define passive, velocity to force, maps it is proved that with this observer together with a Proportional plus damping (P+d) controller, velocities and position error are globally bounded. Finally, in the case that the human operator and the environment do not exert forces on the local and remote manipulators, respectively, global asymptotic convergence of velocities and of position error to zero is achieved. The theoretical results are sustained with simulations using a couple of two degreesof-freedom nonlinear manipulators.

Control of bilateral teleoperators with time delays using only position measurements

“…iii. the P+d controller with the I&I velocity observer reported in the work of Sarras et al 20 and that requires the complete teleoperator dynamics to be implemented. The P+d controller is given as follows:…”

“…Few remarkable works deal with the control of teleoperators without relying on velocity measurements. () In the work of Garcia‐Valdovinos et al, a sliding control technique was used to control a linearized version of the local and remote manipulators, making use of measurements of the force of human and environment. An adaptive control scheme, to render the teleoperator input‐to‐state stable, was provided in the work of Polushin et al The work of Hua and Liu proves boundedness of the position error using a high‐gain velocity observer.…”

“…Furthermore, Hua et al proposed a terminal‐sliding‐mode scheme to obtain an estimation of the manipulators' velocities in finite time, and this estimation was used in a bounded proportional‐derivative‐like controller to establish, after the solution of an LMI, asymptotic convergence of position error and velocities to zero. In the work of Sarras et al, the immersion and invariance (I&I) velocity observer() was ported to the bilateral teleoperator control. The main practical drawback of the works of Sarras et al and Hua et al is the fact that their observer design requires exact knowledge of the complete system dynamics.…”

“…In the work of Sarras et al, the immersion and invariance (I&I) velocity observer() was ported to the bilateral teleoperator control. The main practical drawback of the works of Sarras et al and Hua et al is the fact that their observer design requires exact knowledge of the complete system dynamics.…”

Nonlinear bilateral teleoperators under event‐driven communication with constant time delays

“…In practice, however, the velocity signals could be either difficult to obtain or too noisy to be included in the control input. [38][39][40] Therefore, in the last part of this paper, we extend the developed bilateral teleoperators with event-driven communication to the system without velocity measurement. By assuming that the norm of robotic velocity is bounded by the estimated velocity with a positive bias term, triggering conditions are proposed so that the bilateral teleoperation is stable with ultimately bounded tracking errors.…”