The incorporation of prior knowledge into learning is essential in achieving good performance based on small noisy samples. Such knowledge is often incorporated through the availability of related data arising from domains and tasks similar to the one of current interest. Ideally one would like to allow both the data for the current task and for previous related tasks to self-organize the learn...

Growing research interest in space robotic systems capable of accurately performing autonomous manipulation tasks within an acceptable execution time has led to an increased demand for lightweight materials and mechanisms. As a result, joint flexibility effects become important and represent the main limitation to achieving satisfactory trajectory-tracking performance. This paper addresses the ...

Abstract: A new output feedback controller design approach for flexible-joint (FJ) robots via the observer dynamic surface design technique is presented. The proposed approach only requires the feedback of position states. We first design an observer to estimate the link and actuator velocity information. Then, the link position tracking controller using the observer dynamic surface design proc...

A stable neural network based observer for general multivariable nonlinear system is presented in this paper. Unlike most previous neural network observers, the proposed observer uses nonlznear m parameter neural network (NLPNN). Therefore, it can be applied to systems with higher degrees of nonlinearity without any a priori knowledge of system dynamics. The learning rule of the neural network ...

In this paper a robust composite control algorithm is proposed for flexible joint manipulators, with the emphasis on satisfying control effort limitations. An H∞ framework is used for the slow subsystem controller design, instead of robust PID synthesis introduced in the literature. Linear identification techniques are used to represent the nonlinear dynamics of the system into a linear model p...

We propose a new model for robot manipulators with rigid links and flexible joints. Our model is expressed in Cartesian dependent coordinates which represent the positions, orientations and velocities of the links and the rotors, as opposed to independent generalized coordinates in which classical “Lagrangian” models are defined. The model consists of a set of dynamics equations and holonomic c...

Flexibility at the joint of a manipulator is an intrinsic property. Even “rigid-joint” robots, in fact, possess a certain amount of flexibility. Previous experiments confirmed that joint flexibility should be explicitly included in the model when designing a high-performance controller for a manipulator because the flexibility, if not dealt with, can excite system natural frequencies and cause ...

We address the problem of (semiglobal) trajectory tracking control of rigid and exible joints robots using computed torque based controllers. The paper contains two contributions; rst, for rigid robots we show that velocity in the control algorithm can be replaced by the ltered position, thus obviating the explicit need of state observers. Second, we show that it is posible to achieve trajector...

In this paper the design of an optimal nonlinear H∞ controller for flexible joint robot (FJR) is presented. An approximate solution based on Taylor Series expansion is considered for the Hamilton-Jacobi-Isaac (HJI) inequality. A two-degree-of-freedom controller combined of optimal nonlinear H∞ controller and inverse dynamics controller is proposed to tackle the regulation as well as tracking pr...