A robust nonlinear control approach for flexible-link robots using Kalman Filtering
Gerasimos Rigatos
A robust control approach for flexible-link robots is developed that comprises sliding-mode control theory and Kalman Filtering. There are two issues associated to the control of flexible structures, such as flexible-link robots: (i) simultaneous position control and suppression of the flexible structure vibrations. Assuming a known model of the robot dynamics, this can be succeeded with the use of robust model-based control schemes, such as sliding-mode control, (ii) obtaining measurements of the complete state vector of the
vibrating structure, so as to implement state-feedback control. To solve the latter problem, in this paper, state estimation for the flexible-link robot is implemented with the use of Kalman Filtering. The fast recursion of the Kalman Filter provides real-time estimates of the robot’s state vector through the processing of measurements coming from a limited number of sensors. The obtained state estimates are optimal with respect to the effects of measurement noise. The efficiency of the proposed state estimation-based sliding-mode controller is evaluated through simulation experiments in
the case of a 2-link flexible manipulator.
CYBERNETICS AND PHYSICS, VOL. 1, No. 2, 2012, 134-143.