The variable-inertia modified computed-torque control of robot manipulators

Abstract

This paper describes a modification of the computed-torque method of motion control of robot manipulators, which utilizes inner feedback to partially decouple the inertial effects of individual links, but at the same time it minimizes the influence of the inner feedback component of the control input. In this way, it is possible to increase the control efficiency for a given magnitude of the control signals, which are also more easily kept within saturation limits. It is shown that uniform asymptotical stability of the control error during reference trajectory tracking can be ensured by choosing sufficiently high controller gains, and that the region of attraction can be made arbitrarily large. For the situations when some robot model parameters are not known precisely, an adaptive extension of the algorithm preserving asymptotical stability of the control error is proposed.