Control of Polynomial Nonlinear Systems Using Higher Degree Lyapunov Functions

Shuowei Yang,Fen Wu

DOI: https://doi.org/10.1115/1.4026172

2014-01-01

Abstract:In this paper, we propose a new control design approach for polynomial nonlinear systems based on higher degree Lyapunov functions. To derive higher degree Lyapunov functions and polynomial nonlinear controllers effectively, the original nonlinear systems are augmented under the rule of power transformation. The augmented systems have more state variables and the additional variables represent higher order combinations of the original ones. As a result, the stabilization and L-2 gain control problems with higher degree Lyapunov functions can be recast to the search of quadratic Lyapunov functions for augmented nonlinear systems. The sum-of-squares (SOS) programming is then used to solve the quadratic Lyapunov function of augmented state variables (higher degree in terms of original states) and its associated nonlinear controllers through convex optimization problems. The proposed control approach has also been extended to polynomial nonlinear systems subject to actuator saturations for better performance including domain of attraction (DOA) expansion and regional L-2 gain minimization. Several examples are used to illustrate the advantages and benefits of the proposed approach for unsaturated and saturated polynomial nonlinear systems.