Adaptive Trajectory Neural Network Tracking Control for Industrial Robot Manipulators with Deadzone Robust Compensator
- PDF / 3,334,864 Bytes
- 12 Pages / 594.77 x 793.026 pts Page_size
- 89 Downloads / 231 Views
ISSN:1598-6446 eISSN:2005-4092 http://www.springer.com/12555
Adaptive Trajectory Neural Network Tracking Control for Industrial Robot Manipulators with Deadzone Robust Compensator La Van Truong, Shou Dao Huang, Vu Thi Yen, and Pham Van Cuong* Abstract: This paper proposed a novel adaptive tracking neural network with deadzone robust compensator for Industrial Robot Manipulators (IRMs) to achieve the high precision position tracking performance. In order, to deal the uncertainty, the unknown deadzone effect, the unknown dynamics, and disturbances of robot system, the Radial Basis function neural networks (RBFNNs) control is presented to control the joint position and approximate the unknown dynamics of an n-link robot manipulator. The online adaptive control training laws and estimation of the dead-zone are determined by Lyapunov stability and the approximation theory, so that the stability of the entire system and the convergence of the weight adaptation are guaranteed. In this controller, a robust compensator is constructed as an auxiliary controller to guarantee the stability and robustness under various environments such as the mass variation, the external disturbances and modeling uncertainties. The proposed control is the verified on a three-joint robot manipulators via simulations and experiments in comparison with PID and Neural networks (NNs) control. Keywords: Adaptive control, RBF network, robot manipulator, unknown deadzone.
1.
INTRODUCTION
Recently, Robot manipulators have been widely applied in the industrial. In fact, Industrial Robot manipulators are multivariable nonlinear systems and they suffer from various uncertainties in their dynamics, which deteriorate the system performance and stability, such as external disturbance, nonlinear friction, highly time-varying, and payload variation. Therefore, achieving high performance in trajectory tracking is a very challenging task. So, many researchers were proposed adaptive controller, robust adaptive controller, fuzzy logic control, neural network control, etc. [1–8]. In [9], the vector control method of induction motor using an MRAS-fuzzy logic observer was presented. This type of observer combines the Model Reference Adaptive Systems (MRAS) technique with the fuzzy logic to design an MRAS-fuzzy logic observer which can at first estimate the rotor speed and second the rotor resistance. In [10], adaptive model control and neural network based trajectory planner were designed for dynamic balance and motion tracking of desired trajectories. However, this control needed the knowledge of dynamics. In [11], an adaptive controller based neural networks was proposed to deal uncertainties and input saturation of robotic ma-
nipulators. The RBFNNs controller was used to approximate the unknown dynamic and an auxiliary system was designed to solve the input saturation. In [12–16], adaptive neural network controllers were presented by using output feedback methods, and in [17–26], artificial neural networks were widely used for the control design of nonlinear
Data Loading...