Improvement of Tracking Control of a Sliding Mode Controller for Robot Manipulators by a Neural Network

다국어 초록 (Multilingual Abstract)

This article presents a neural network control technique to improve the tracking performance of a robot manipulator controlled by the sliding mode control method in a non-model-based framework. The sliding mode controller is a typical nonlinear controller that has been well developed in theory and used in many applications due to its simplicity and practicality. Selection of the gain of the nonlinear function plays an important role in performance as well as stability. When the sliding mode controller is used for the non model-based configuration in robot control, the nonlinear gain should be selected large enough to guarantee the stability. Since the appropriate selection of the gain value is essential and difficult in the sliding mode control framework, a neural network compensator is introduced at the trajectory level to help the fixed gain deal with the stability and performance more intelligently.
Stability of the proposed control scheme is analyzed. Simulation studies of following the Cartesian trajectory for a three-link rotary robot manipulator are conducted to confirm the control improvement by the neural network.

참고문헌 (Reference)

1 이상덕, "외란관측기 설계를 위한 Q필터 시정수 영향 분석:외바퀴 로봇의 균형 제어 응용" 대한전자공학회 53 (53): 123-129, 2016

2 R. G. Rodriguez, "Tracking control of robot manipulators using second order neuro sliding mode" 39 (39): 285-294, 2009

3 정슬, "Stability Analysis of Reference Compensation Technique for Controlling Robot Manipulators by Neural Network" 제어·로봇·시스템학회 15 (15): 952-958, 2017

4 H. Zhang, "Sliding mode controller with RBF neural network for manipulator trajectory tracking" 45 (45): 2015

5 E. Sariyildiz, "On the explicit robust force control via disturbance observer" 62 (62): 1581-1589, 2015

6 W. Gao, "Nonlinear and adaptive suboptimal control of connected vehicles: A global adaptive dynamic programming approach" 85 (85): 597-611, 2017

7 공경철, "Nominal Model Manipulation for Enhancement of Stability Robustness for Disturbance Observer-Based Control Systems" 제어·로봇·시스템학회 11 (11): 12-20, 2013

8 S. Jung, "Neural network inverse control techniques for PD controlled robot manipulator" 19 (19): 305-314, 2000

9 F. L. Lewis, "Neural Network Control of Robot Manipulators and Nonlinear Systems" Taylor & Francis 1999

10 H. Gomi, "Learning control for a closed loop system using feedback error learning" 3289-3294, 1990

11 R. J.Wai, "Fuzzy-neural network inherited sliding-mode control for robot manipulator including actuator dynamics" 24 (24): 274-287, 2013

12 H. Xin, "Full-order neural sliding model control of robotic manipulator with unknown dead-zone" 1664-1669, 2015

13 T. Liu, "An improved neural network adaptive sliding mode control used in robot trajectory tracking control" 11 (11): 1655-1666, 2015

14 W. Gao, "Adaptive dynamic programming and adaptive optimal output regulation of linear system" 61 (61): 4164-4169, 2016