Yuxin Su

Genetic design of kinematically optimal fine tuning Stewart platform for large spherical radio telescope

Abstract This paper presents a methodology for the design of optimal kinematical characteristics of Stewart platform using genetic algorithms (GAs). The optimal kinematics index which expressed by Jacobian matrix of Stewart platform is first deduced, and then the minimum of condition numbers of Jacobian matrix in the whole trajectory tracing workspace is used as the objective function. The constrained optimal design problem is transformed to unrestrained optimal one which is suitable to GAs by penalizing strategy, and the corresponding real-encoding scheme is used. This genetic methodology for optimal kinematical Stewart platform is illustrated by applying to design the fine tuning Stewart platform for large spherical radio telescope, and its validity is verified with the kinematic accuracy comparison of genetic-designed fine tuning platform with that of conventional Quasi-Newton optimal method.

The mechanical design and kinematics accuracy analysis of a fine tuning stable platform for the large spherical radio telescope

A large Stewart platform for fine tuning of the feed source tracing is presented in this paper. The model of kinematics control is developed with coordinate transformation, and a quasi-static load analysis is made by virtual work principle with Jacobian matrix because the tracing speed is slow. The kinematics accuracy model is derived by position vector analysis, and the kinematics accuracy results are provided. It has shown that the designed Stewart platform can satisfy the requirement of feed source trajectory control of large spherical radio telescope.

A Simple Nonlinear Observer for a Class of Uncertain Mechanical Systems

A simple nonlinear observer is proposed for a class of uncertain nonlinear multiple-input-multiple-output (MIMO) mechanical systems whose dynamics are first-order differentiable. The global asymptotic observation of the proposed observer is proved. Thus, the observer can be designed independently of the controller. Furthermore, the proposed observer is formulated without any detailed model knowledge of the system. These advantages make it easy to implement. Numerical simulations are included to illustrate the effectiveness of the proposed observer.

A Global Asymptotic Stable Output Feedback PID Regulator for Robot Manipulators

In this paper, we provide an answer to the long-standing question of designing global asymptotically stable proportional-integral-derivative (PID) regulators with only position feedback for uncertain robots. Our main contribution is to establish the global asymptotic stability of the controlled system by using Lyapunov direct method and LaSalles invariance principle. We provide explicit conditions on the regulator gains to ensure global asymptotic stability. The proposed controller does not utilize the modeling information in the control formulation, and thus permits easy implementation. Simulations performed on a planar two degrees-of-freedom robot manipulator demonstrate the effectiveness of the proposed approach.

Auto disturbance rejection motion control for direct-drive motors

A simple and high performance auto-disturbance rejection controller (ADRC) only based on the rotor position measurement is proposed for motion control of direct-drive motors. The motion controller is consisted of a nonlinear tracking differentiator in the feedforward path, an extended-state observer (ESO) and a nonlinear proportional derivative synthesizer in the feedback path. The ESO obtains the estimated state of the plant and the unknown disturbance action at the plant input with simple computation using only the measurement output of the plant. The proposed method is implemented to control an AC direct-drive servomotor system to verify its high-precise and robustness performance.

Global continuous finite-time output feedback regulation of robot manipulators

A simple continuous output feedback proportional-derivative (PD) plus desired gravity compensation controller is proposed to solve the global finite-time regulation of robot manipulators with position measurements only. The global finite-time convergence is proved by using Lyapunov theory and finite-time stability theory. Simulations preformed on a two degrees-of-freedom (DOF) manipulator demonstrate the expected properties of the proposed approach.

Correction to “Global Asymptotic Saturated PID Control for Robot Manipulators” [Nov 10 1280-1288]

Manuscript received March 25, 2014; accepted April 13, 2014. Date of publication May 8, 2014; date of current version December 15, 2014. Manuscript received in final form April 14, 2014. Recommended by Associate Editor T. Parisini. Y. Su is with the School of Electro-Mechanical Engineering, Xidian University, Xi’an 710071, China (e-mail: yxsu@mail.xidian.edu.cn). P. C. Müller is with the School of Safety Control Engineering, University of Wuppertal, Wuppertal D-42097, Germany (e-mail: mueller@srm.uni-wuppertal.de). C. Zheng is with the School of Electronic Engineering, Xidian University, Xi’an 710071, China (e-mail: chzheng@xidian.edu.cn). Digital Object Identifier 10.1109/TCST.2014.2320698 It can be shown that (1) is positive definite with respect to φ because Ki and Cosh 2(·) are diagonal positive-definite matrices, σi |σi =0 = 0, and σi is an increasing function with respect to σi . In addition, the time derivative of (1) is given by

A simple nonlinear PID control for global finite-time regulation of robot manipulators without velocity measurements

A simple nonlinear proportional-integral-derivative (PID) controller is proposed for global finite-time regulation of robot manipulators without velocity measurements. A Lyapunov-based stability argument is employ to prove global finite-time stabilization. The proposed control algorithm does not involve the model parameters in the control law formulation and the control gains can be explicitly determined based on some well-known bounds extracted from the robot dynamics, and thus permits easy implementation. Simulations are included to demonstrate the expected properties of the proposed approach.

A Simple Linear Velocity Estimator for High-Precision Motion Control

High-quality low speed motion control calls for precise position and velocity signals. However, velocity estimation based on simple numerical differentiation from the only position measurement may be very erroneous, especially in the low-speed regions. A simple linear velocity estimator is proposed based for high-quality instantaneous velocity estimation, on position measurements only. The proposed estimator is constructed based on the fact that velocity belongs to the chain of kinematic quantities: position and velocity, and numerical integration can provide more stable and accurate results than numerical differentiation in the presence of noise. The main attraction of the new algorithm is that it is very effective as well as in low-speed ranges, high robustness against noise, and easy to implement with simple computation. Both extensive simulations and experimental tests have been performed to verify the effectiveness and efficiency of the proposed approach.

A Simple Repetitive Learning Control for Asymptotic Tracking of Robot Manipulators with Actuator Saturation

Abstract In this paper, we provide a simple decentralized saturated repetitive learning controller for asymptotic tracking of robot manipulators under actuator saturation. The proposed control consists of a saturated nonlinear proportional plus derivative action and a saturated learning-based feedforward compensation term. A Lyapunov-like stability argument is employed to show semiglobal asymptotic tracking. Advantages of the proposed controller include an absence of modeling parameter in the control law formulation and an ability to ensure actuator constraints are not breached. This is accomplished by selecting control gains a priori, removing the possibility of actuator failure due to excessive torque input levels. The effectiveness of the proposed approach is illustrated via simulations.