Robust tracking control design and its application to balance a two-wheeled robot steering on a bumpy road

A reinforced adaptive fuzzy sliding-mode controller was developed to track the desired state for a certain class of nonlinear dynamic system. Initially, adaptive fuzzy logic was used to derive the update laws for approximating the uncertain nonlinear functions of a dynamic system. Subsequently, a reinforced adaptive mechanism and sliding-mode control were incorporated into the adaptive fuzzy control scheme. In the reinforced adaptive mechanism, both the model and controller parameters were adjusted simultaneously to attenuate the effects caused by the unmodeled dynamics and disturbances. Performance analysis demonstrated the superiority of the proposed reinforced adaptive law over the conventional direct adaptive scheme regarding faster tracking and convergence of parameters. To verify its effectiveness and extend its application, the proposed reinforced adaptive fuzzy sliding-mode controller was applied to balance a two-wheeled robot steering on a bumpy road. A number of simulations and experiments were performed. In addition, a conventional direct adaptive scheme under the same environment was also performed for comparison. These results demonstrated that the balance performance of a nonlinear dynamic two-wheeled robot improved because of embedding the proposed reinforced adaptive fuzzy sliding-mode controller into the microcontroller.