An optimal controller for a nonlinear robotic-system

A method of designing an optimal controller for a two-link robot is developed. Robot motion is constrained and is given in terms of nonlinear dynamical equations. Each robot link is subjected to motion from an initial angular displacement and velocity to a set of final angular displacement and velocity, during which a preselected quadratic performance index of the robot is optimized. The proposed dynamic modeling method and closed-loop control algorithm for decentralized control are applied to an example two-link robot.<
>