Neural network adaptive backstepping control of multi-link flexible-joint robots

被引：0

作者：

Li C. ^{[1
]}

Cui W. ^{[1
]}

You J. ^{[2
]}

Lin J. ^{[1
]}

Xie Z. ^{[1
]}

机构：

[1] College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing

[2] College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing

来源：

Shanghai Jiaotong Daxue Xuebao/Journal of Shanghai Jiaotong University | 2016年 / 50卷 / 07期

关键词：

Adaptive backstepping control; Flexible-joint; Joint motion; Multi-link; Neural network;

D O I：

10.16183/j.cnki.jsjtu.2016.07.019

中图分类号：

学科分类号：

摘要：

Aimed at the control problem of joint motion of multi-link flexible-joint robots (RFJ) with model inaccuracies, a new kind of neural network (NN) adaptive backstepping control algorithm was proposed. The control scheme proposed uses a kind of NN adaptive backstepping control scheme of single-link RFJ for reference, separates the nonlinear unknowns from the known terms, and uses radial basic function (RBF) to approximate the nonlinear unknowns in the process of backsteppping. Then, by improving the NN adaptive backstepping control scheme of single-link RFJ, the control scheme proposed constructs a new nonlinear unknown term and estimates the diagonal elements of rotor inertia matrix instead of the rotor inertia matrix. A control law and parameter update laws applying to the multi-link RFJ were designed according to the Lyapunov function. Consequently, the control of the joint trajectory tracking of multi-link RFJ can be realized. The simulation results show that the control algorithm proposed has a better tracking performance than the general proportion-differential algorithm. Meanwhile, the control algorithm proposed can guarantee the tracking precision of a certain trajectory although the number of nodes in the neural network has a small value. © 2016, Shanghai Jiao Tong University Press. All right reserved.

引用

页码：1095 / 1101

页数：6

共 11 条

[1]

Shi X., Liu S., A survey of trajectory tracking control for robot manipulators, Control Engineering of China, 18, 1, pp. 116-122, (2011)

[2]

Li J., Yang C., Culverhouse P., Et al., Stochastic neural network control of rigid robot manipulator with passive last joint, Control (CONTROL), 2012 UKACC International Conference on, pp. 662-667, (2012)

[3]

Soltanpour M.R., Khooba M.H., A particle swarm optimization approach for fuzzy sliding mode control for tracking the robot manipulator, Nonlinear Dynamics, 74, 1-2, pp. 467-478, (2013)

[4]

Sasiade J.Z., Green A., Ulrich S., Et al., Nonparametric identification and control of flexible joint robot manipulator, Methods and Models in Automation and Robotics (MMAR), 2014 19th International Conference on, pp. 221-228, (2014)

[5]

Yeon J.S., Yim J., Park J.H., Robust control using recursive design method for flexible joint robot manipulators, Journal of Mechanical Science & Technology, 25, 12, pp. 3205-3213, (2011)

[6]

Chang Y.C., Yen H.M., Design of a robust position feedback tracking controller for flexible-joint robots, Control Theory & Applications Iet, 5, 2, pp. 351-363, (2011)

[7]

Chen Z., Chen L., Adaptive backstepping control of flexible-joint space robot based on neural network, Engineering Mechanics, 30, 4, pp. 397-401, (2013)

[8]

Huang A.C., Chen Y.C., Adaptive sliding control for single-link flexible-joint robot with mismatched uncertainties, IEEE Transactions on Control Systems Technology, 12, 5, pp. 770-775, (2004)

[9]

Macnab C.J.B., D'Eleuterio G.M.T., Neuroadaptive control of elastic-joint robots using robust performance enhancement, Robotica, 19, 11, pp. 619-629, (2001)

[10]

Pai M.C., Dynamic output feedback RBF neural network sliding mode control for robust tracking and model following, Nonlinear Dynamics, 79, 2, pp. 1023-1033, (2015)

← 1 2 →