Two hybrid model-based control strategies for a flexible parallel planar manipulator

Designers can reduce the inertia of moving parts of a robotic manipulator aiming for energy efficiency. Nevertheless, this design option could yield undesired vibrations due to its lighter components, requiring control strategies for their mitigation. In addition, other challenges arise when designing parallel manipulators since the control strategies cannot use a direct measurement of the manipulator end-effector's pose and their coupling dynamics. This work proposes two hybrid model-based control design strategies for a 3RRR parallel manipulator with flexible links in this work. Both strategies have two feedback loops, one of which exploits the encoder's measurements and the inverse kinematic model of the manipulator considering rigid links. The difference between the strategies is related to the feedback signal used in the second feedback loop that exploits an LQR/LQG strategy. The first strategy uses the end-effector's pose measured by a limited frame camera, while the second uses the end-effector's pose derived from a reduced-order finite element model. These strategies are experimentally compared with a PID strategy for assessing their assets and limitations. The steady-state error was lower when using the model-based control strategies. Moreover, the second strategy responds faster than the first for the evaluated tasks.