Abstract
This work aims to demonstrate and benchmark different model-based nonlinear control design techniques for elementary controls education. A standard, two-link planar robot arm (Fig.1) is chosen to be the plant due to its nonlinear system dynamics and relatively simple inverse kinematics considering the math level of complexity of a typical senior-level mechanical engineering class. To facilitate model-based control design, the system dynamic model is derived through Lagrange’s equations in non-conservative form. An inverse kinematic analysis is also performed to extract the target joint variables to match for a circular motion executed in the task space. Finally, the system dynamic model is incorporated with feedback control designs based on the computed torque and sliding mode control methods. The tracking performance of the proposed nonlinear controllers are validated in closed-loop numerical simulations based on the same target trajectories to compare the performances achieved by the different algorithms.
