Design and Application of Robust Reduced-Order Hybrid Position and ForceControllers for a Two-Link Flexible Manipulator

摘要

This research focuses on the design and application of robust reduced-orderhybrid position and force controllers on a two-link flexible manipulator. The first part of the research examines the design aspect of the hybrid controllers. Traditionally, full-order controllers are used in common techniques such as Linear-Quadratic-Gaussian (LQG) control. If the plant is of large order, then the resulting controller is also of large order. Full-order controllers have the disadvantages of requiring large amounts of memory as well as using excessive computational time. In addition, LQG-class controllers are only guaranteed to work at the design condition. With the use of the direct-optimization tool SANDY, reduced-order controllers are designed directly for several plant conditions (robustness). The experimental results show that the robust reduced-order SANDY controllers provide comparable performance to the full-order LQG controllers in the applications of position control, force control, and hybrid position and force control.