Modeling and control of micropositioning systems using Stewart platforms

The modeling and control of a 6-DOF Stewart micropositioning system with each leg actuated by a respective piezoelectric actuator are considered in this paper. The 12 multi-DOF passive joints are assumed to be well designed and fabricated so that guaranteed guiding precision and lack of backlash can be obtained. The dynamics model of the micropositioning system is derived first, and then a composite control strategy consisting of moving platform model-based feedback linearization and two sets of simple SISO fuzzy systems is proposed. By considering the internal axial forces of the six legs on the six spherical joints at the moving end as the virtual control inputs of the moving platform, feedback linearization can be easily used to derive the desired control forces for the moving platform. The corresponding desired linear displacement of each piezoelectric actuator can then be computed based on the derived leg model, and each piezoelectric actuator's control voltage can be generated by the first set of independent leg fuzzy controls. The second set of fuzzy controls is suggested for the further enhancement of robustness with respect to uncertainty. Computer simulations are presented to illustrate the effectiveness of the suggested micropositioning control strategy.