An intelligent sliding mode controller of robotic manipulators with output constraints and high‐level adaptation

Abstract

Fast transient responses and excellent steady-state control performances with flexibility in control operation are core motivations to promote a vast of research in the robotic control sector nowadays. As a sequence, in this paper, we present a learning nonlinear controller for motion control of robotic manipulators with output constraints. The constrained control objectives are first transformed to new free variables using nonlinear synthetization profiles. Boundedness of the main control objectives within feasible ranges of the physical outputs is next strictly guaranteed under a simple sliding-mode control policy. To improve the control performance, dominant terms of the system dynamics are learnt by a neural network with a nonlinear adaptation law. The systematic deviation remaining is then estimated by an adaptive disturbance observer. A novel high-level adaptation rule is designed to maximize the learning ability and the control quality. Both an asymptotic control performance and estimation effectiveness are theoretically thoroughly proven by Lyapunov-based analyses. Feasibility of the overall system is carefully investigated by intensive comparative simulations.

An Intelligent Position-Tracking Controller for Constrained Robotic Manipulators Using Advanced Neural Networks

“…Validation results in previous work [46,47] confirmed the learning efficiency of the disturbance observer Eq. ( 20) for simple systems.…”

“…The controllers were employed for motion control of a 3DOF robot, as depicted in Figure 2. Detailed dynamics of the 3DOF robot were derived based on the Lagrange method [4,19,47], as formulated in Appendix D. The neural network had 9 inputs q i , _ q i , τ i ÀÁ |i¼1,2,3 and 730 neurons with the logsig activation function in the hidden layer [42,45]. All of the initial values of the weight vectors ŵi|i¼1,2,3 ÀÁ were set to be zero.…”

Impedance Force Control of Manipulator Based on Variable Universe Fuzzy Control

“…Fuzzy neural networks also find extensive applications in the field of robot control [10][11][12]. References [13][14][15] used a sliding-mode controller to compensate system input. There are also some parameter compensation control methods [16,17], such as impedance controllers that compensate for the position or impedance parameters to improve control accuracy.…”