Neural network-based adaptive hybrid impedance control for electrically driven flexible-joint robotic manipulators with input saturation.

In this paper, a neural network (NN)-based adaptive hybrid impedance control (AHIC) scheme is proposed for the electrically driven flexible-joint robotic manipulators (EDFJRM) with input saturation, where the hybrid impedance is designed to obtain the reference trajectory by combining the position/force control with the impedance control. The proposed scheme integrates hybrid impedance control into backstepping technique, where a smooth saturation function is employed to facilitate input saturation. Moreover, the reference input trajectory of the robot is obtained by different target dynamics in the free space and the constrained space, and NNs are employed to approximate the saturation errors term, uncertain parts and external disturbances. In this way, the control objective of the position and force tracking can be realized in both free space and constrained space. Lyapunov stability analysis shows that all the signals in closed-loop system are guaranteed to be uniformly ultimately bounded, and the steady-state tracking errors of the position in the free space and the force in the constrained space are converged to zero. Finally, simulation results of two-rigid-link EDFJRM demonstrate the effectiveness of the proposed NN-based AHIC scheme. [ABSTRACT FROM AUTHOR]