Adaptive Neural Networks Control of Flexible-Joint Robots with Full-state Constraint: Dynamic Nonlinear Mapping Technique

Abstract

In this paper, the problem of tracking control for a class of flexible joint robot systems with asymmetric time-varying state constraints is studied. To ensure that the states are within the constraint range, a dynamic nonlinear function is applied to the controller design, and an adaptive neural network command filter control strategy is proposed. Using the approximation ability of radial basis function (RBF) neural network, the difficulty of unknown nonlinear function is overcome successfully. At the same time, in order to avoid the “complexity explosion” and “singularity” problems in the backstepping process, the command filtering technology is applied to the backstepping design. It is shown that the proposed control scheme can ensure that all the signals of the closed-loop system are bounded, the output can track the reference signal within a certain error range, and all the states are ensured to remain in the predefined compact set. Finally, in order to verify the effectiveness of the scheme, a single-link FJ robot system model is used to simulate the scheme.