Extension of linear quadratic regulator theory and its applications.

Abstract

Linear Quadratic Regulatory theory (L.Q.R.) has received widespread application due to its simplicity and also due to the fact that the control provided this by theory is linear in form. These features make the implication of feedback control and easy task. In contrast, nonlinear regulation lack those attractive features enjoyed by the linear regulator. Moreover, in order to obtain the feedback control, one has to solve Hamilton-Jacobi-Bellman equation which is not an easy task. Also, if solution can be obtained, implementation is not always practical. In this work, we extend the Linear Quadratic Regulatory theory to the following; (I) LQR theory is modified for the case when there is no control contribution to the cost functional. (II) LQR is used to regulate or fine-tune a nonlinear system around a nominal trajectory through linearization of nonlinear systems. (III) Applying the LQR theory for the regulation of angular velocities of a three-axes satellite around a nominal point. (IV) Applying the LQR for the regulation of the movement of a robot around a time-optional trajectory. (V) The limitation of the control obtained through linearization is indicated.