RMMAC: a novel robust adaptive control scheme. Part II. Performance evaluation

We present and discuss numerous stochastic simulation results that demonstrate and validate the superior performance of the robust multiple model adaptive control (RMMAC) methodology introduced in part I (Fekri et al., 2004). The system used is akin to the two-cart benchmark problem and it has a single uncertain mass. We show that the RMMAC significantly improves disturbance-rejection, as compared with the "best" nonadaptive controller designed by mixed-μ synthesis; moreover, the RMMAC requires lower amplitude control signals. In the example considered, in addition to the uncertain mass, there are unmodeled dynamics as well as (unmeasured) stochastic disturbance inputs and noisy sensor measurements.