A sliding mode control design methodology is applied to solve the position tracking problem of an uncertain very lightweight single-link flexible robot arm. Throughout this work we seek a control scheme robust with regards to payload and actuator friction changes, without the establishment of any bounding limits on these parameter variations. Based on the decomposition of the robot dynamics into the motor and flexible link submodels, two nested control loops are proposed, each one of these controlled by an independent sliding mode controller. By carrying on this model decomposition, both motor friction and payload changes become matched uncertainties in their respective models, allowing for their effective rejection using these sliding controllers. Simulations and experimental results are shown to demonstrate the performance of the control system proposed.

No abstract

No abstract

In this article an adaptive position control scheme for a DC motor based on an on-line closed-loop continuous-time identification method is proposed. A fast, non-asymptotic, algebraic identification method is used to estimate the unknown system parameters and to update the controller. The method is suitable for simultaneously identifying both the viscous friction coefficient and the inertia of the motor. The adaptive controller and the identification algorithm have been verified by simulation and experiments.

Abstract. The position of the DC motor is controlled by using a continuous sliding mode control (SMC), which is highly robust to the Coulomb friction torque and to high unknown payload variations, which involve changes in the rotational inertia of the motor shaft. The main contribution of the work is the experimentation of a SMC control which does not requires the knowledge of the payload variation range, i.e., the system is quite robust to any unknown change in the payload mass value.