An adaptive link position tracking controller for rigid-link flexible-joint robots without velocity measurements

This paper presents an adaptive partial state feedback controller for a rigid-link flexible-joint (RLFJ) robot. The controller compensates for parametric uncertainty throughout the entire mechanical system while only requiring measurement of link position and actuator position. To eliminate the requirements for measuring link velocity and actuator velocity, a set of filters is utilized as a surrogate for the unmeasurable quantities. Based on this set of filters, an adaptive integrator back-stepping procedure is used to develop a torque input controller which guarantees semi-global asymptotic link position tracking while also ensuring that all signals remain bounded during closed-loop operation. Simulation results for a two-link RLFJ robot are utilized to validate the performance of the proposed controller. >