Feedback linearization control of a two-link robot using a multi-crossover genetic algorithm

PD control with gravity compensation for hydraulic 6-DOF parallel manipulator

“…Do and Yang [8] used the Newton-Euler approach to solve the inverse dynamics for Stewart platform assuming the joints as frictionless and legs asymmetrical. The control strategies for parallel manipulator may be largely divided into two schemes, joint-space control developed in joint space coordinates [9][10][11], and workspace control designed based on the workspace coordinates [12][13][14]. The joint-space control scheme can be readily implemented as a collection of multiple, independent single-input single-output control system using data on each actuator length only.…”

On the design of serial manipulators with decoupled dynamics

“…It should be noted that a number of procedures for the synthesis of control systems ensuring high-quality control of manipulators have been elaborated on the basis of the general form of nonlinear dynamic equations [23][24][25][26][27][28][29][30][31][32]. However, regardless of the permanent tendency to decrease the price of microcomputer systems, the price of implementation of a complete dynamic control in the case of high-speed and precise dynamical tasks on the industrial practice is still high.…”

Preliminary investigations on the applicability of the fixed point transformations-based adaptive control for time-delayed systems

“…In the sixties of the past century, his work was translated to English [29], and by the nineties it appeared as the main design tool for adaptive controllers (e.g., [30], [31]), Model Reference Adaptive Controllers (MRAC) [32], [33], [34]. In spite of its great virtues, this method has some drawbacks such as mathematical complexity, requiring satisfactory conditions that frequently restrict much more than the necessary and satisfactory conditions, allowing the use of a great number of arbitrary control parameters that may be optimized by evolutionary methods (e.g., [35], [36]). A fresh and emerging modeling approach is the Tensor Product Model (TP) (e.g., [37], [38]) in which the Lyapunov function technique can be combined with Linear Matrix Inequalities for controller design purposes.…”