Optimal state and sensitivity feedback control of a second order linear regulator

In this thesis the utility of using an optimal state and sensitivity feedback control policy to drive a second order linear system with uncertain damping ratio to a desired equilibrium state is examined. A comparison is made between the sensitivity feedback approach for reducing trajectory dispersions in the second order system and the conventional state feedback approach. The sensitivity differential equations are solved using a control driving function to replace a computationally unrealizable term in the sensitivity dynamics. It is shown that the effectiveness in reducing state error of any of three possible methods is the same. The three methods are increasing the state weighting coefficients, increasing the sensitivity weighting coefficients or increasing the sensitivity control weighting coefficient in the performance index.Electrical and Computer Engineering, Department o