Adaptive robust control for ball-screw drives with flexible transmission and nonlinear friction via dynamic surface control approach

Flexible deformation and nonlinear friction in ball-screw drive systems are important factors that restrict the improvement of tracking performance. In this paper, a high-performance adaptive controller is presented for ball screw drives to suppress vibration and improve tracking accuracy. A two-inertia model with torsional vibration state is established to fit the dynamics of the drive system while the continuously differentiable LuGre model characterizes the nonlinear friction disturbance. Based on the established nonlinear model, an adaptive robust controller (ARC) is designed by using the backstepping approach to overcome the parametric uncertainties and hard-to-model dynamics. The dual-observer is employed in the controller to observe and compensate for the nonlinear friction, which improves the low-velocity tracking performance of the ball-screw drives. Meanwhile, first-order filters are introduced by dynamic surface control (DSC) technique to eliminate the “complexity explosion” problem caused by the backstepping method. The controller theoretically guarantees that all signals of the closed-loop system are bounded, and the convergence of tracking error is also ensured via Lyapunov analysis. The effectiveness of the proposed controller is verified through simulation and experimental results.