In this paper, we propose a nonlinear antiwindup scheme to guarantee stability and local performance recovery on Euler-Lagrange systems subject to input magnitude saturation. The antiwindup scheme consists in a dynamic augmentation to the baseline control system which, based on nonlinear analysis, is shown to induce global asymptotic stability and local exponential stability on the overall control scheme. To induce desirable closed-loop performance on a specific system, two parameters of the antiwindup law can be tuned for each position variable of the Euler-Lagrange system, similar to the tuning procedure for proportional-derivative gains. Simulation results are reported showing the behavior of the proposed antiwindup law on the model of a popular industrial robot. The resulting responses confirm the effectiveness of the antiwindup scheme, both for set-point regulation tasks and for trajectory tracking tasks where the saturation nonlinearity is repeatedly activated.

It is known in the practice of closed-loop identification of mechanical systems that, since the Langrangian dynamic model is linear in the constant physical parameters, a sufficient condition for parameter convergence is that the robot be made to follow reference trajectories which render the regressor matrix, persistently exciting. Slotine and Li established this many years ago for a particular adaptive controller. This result is based on adaptive linear systems theory and relies on the condition that the applied control law makes the robot asymptotically track the imposed rich reference trajectories. This paper revisits this problem and establishes a proof that this condition is not only sufficient but also necessary to establish the stronger property of uniform global asymptotic stability of the overall closed-loop system.

In this paper we summarize the anti-windup construction suggested by Teel et al. (1997) for Euler-Lagrange systems and apply it to a SCARA robot controlled by a windup-prone decoupling PID controller. It is shown through simulation that the anti-windup design guarantees the local decoupled linear performance in. addition to global set-point tracking.