A reduced-order controller design for Nonlinear systems with uncertainties and disturbances

This paper addresses a problem of reduced-order controller design for nonlinear systems. The design objectives are disturbance attenuation and global stabilization in the presence of nonlinear dynamic and static uncertainties arising from various locations in the system. In the situation not to have access to full information of the plant state, we need to embed appropriate dynamics in the controller to reconstruct the state information from available measurement. The reduced-order dynamic controller employed in this paper does not have the mechanism to rebuild the full state vector. By constructing an observer-like mechanism only for the unknown part of the state vector, we can reduce the complexity of the controller. This paper demonstrates that the reduced order design problem for uncertain nonlinear systems can be solved by the pair of state-dependent scaling and change of coordinates which are newly tailored to the reduced order feedback problem.