Robust design of independent joint controllers with experimentation on a high-speed parallel robot

The dynamic model of a robot manipulator is described by a set of nonlinear, highly coupled differential equations. Model-based control schemes were proposed to enhance tracking capabilities with respect to simple linear control schemes. Independent joint controllers (of PD or PID type) are usually employed in industrial robot manipulators but cannot achieve satisfactory performance due to their inherent low rejection to disturbances and parameter variations. In this paper, a new linear independent joint control scheme is proposed; the design is made robust by closing another feedback loop that uses acceleration information besides the typical position and velocity loops. Reconstruction of acceleration measurements is performed via a suitable state-variable filter. Linear feedforward compensation is used to improve tracking performance of the closed-loop scheme. The control algorithm is tested first in a discrete-time simulation on a single-joint drive system with imposed disturbance torques. Then real-time implementation on a high-speed parallel robot is presented; the experimental results demonstrate the effectiveness of the proposed technique.