Adaptive tracking control for manipulators with only position feedback

One of the difficulties in implementing robot control algorithms is the requirement of joint velocity signal. The velocity signal is often contaminated by measurement noise and occasionally deviated from the true value. In particular, use of noisy velocity signal in the adaptation mechanism can cause system instability due to parameter drift. Hence, we propose an adaptive tracking control algorithm for rigid-link robot arm systems by using only position feedback. Via Lyapunov stability analysis, it is shown that the overall system can be made semi-globally asymptotically stable as long as some control parameter is chosen large enough. Simulation results also demonstrate the effectiveness of the proposed algorithm.