Robust adaptive control of linear systems with constant uncertain parameters

Investigates the robust adaptive control problem of linear systems with uncertain parameters in the system matrix A and control coefficient matrix B. A minimax dynamic programming approach is used to derive a measurement based control strategy in the presence of both plant and measurement disturbances. The paper generalizes previously proposed approaches, and investigates a simple numerical algorithm to solve for the robust control strategy. The algorithm is demonstrated on a lightly damped second order spring-mass system to show the feasibility of the proposed control strategy and the subsequent implementation scheme. The performance of the controller and its main features are compared to a H/sub /spl infin// design.