In this article, a new nonlinear robust contouring control scheme is proposed for a planar manipulator. Compared with the end effector tracking error, contouring error is a more reasonable description of the minimum distance between actual position and desired contour when tracking a complex trajectory. Thus, the task coordinate contour error is selected to evaluate the tracking performance. However, most of the contouring control schemes are designed for biaxial gantry, the planar manipulator system is subjected to payload change, and the conventional contouring controller cannot solve this problem faultlessly. Here, the continuous nonsingular terminal sliding mode control and time delay estimation are integrated to develop a new contouring controller for large curvature trajectory high-speed tracking. The time delay estimation is adopted to estimate the manipulator dynamic and payload change. The tracking precision can be guaranteed even parameter uncertainty exsits. Furthermore, continuous nonsingular terminal sliding mode control is integrated with the contouring control to obtain faster convergence performance and robustness of the overall system. The proposed controller possesses obvious advantages, such as robust to payload change and a better performance when tracking a large curvature ellipse in high speed. The effectiveness of the proposed method is verified through simulation and experiment on a planar manipulator.

The study of flexible intelligent hand prosthesis has important practical significance for handicapped people. Aiming at the grasping force control of flexible hand prosthesis, this paper draws lessons from the anatomical structure design of human hand, and proposes a fingertip grasping force control method based on linear tension feedback. The control strategy can realize the stable control of fingertip grasping force. The static mechanical model of finger is established, and the mathematical relationship of fingertip contact force calculation in static state is obtained. Then a friction compensation method is proposed. The experimental results show that the fitting value of the friction moment of the lasso matches well with the test value, and the error after friction compensation is obviously reduced.