Robust control of nonlinear systems with parametric uncertainty (Q1614434)

Probabilistic robustness analysis and synthesis for nonlinear systems with uncertain parameters are presented. Monte Carlo simulation is used to estimate the likelihood of system instability and violation of performance requirements subject to variations of the probabilistic system parameters. Stochastic robust control synthesis searches the controller design parameter space to minimize a cost that is a function of the probabilities that the design criterion will not be satisfied. The robust control design approach is illustrated by a simple nonlinear example. A modified feedback linearization control is chosen as a controller structure, and the design parameters are searched by a genetic algorithm to achieve the tradeoff between stability and performance robustness. The proposed method is not restricted by the number of uncertainties or the degree of nonlinearity of the system, and it provides a logical extension to prior results for robust linear control.