Neural network-based motion control of an underactuated wheeled inverted pendulum model

In this paper, automatic motion control is investigated for one of wheeled inverted pendulum (WIP) models, which have been widely applied for modeling of a large range of two wheeled modern vehicles. First, the underactuated WIP model is decomposed into a fully actuated second order subsystem Σ_a consisting of planar movement of vehicle forward and yaw angular motions, and a nonactuated first order subsystem Σ_b of pendulum motion. Due to the unknown dynamics of subsystem Σ_a and the universal approximation ability of neural network (NN), an adaptive NN scheme has been employed for motion control of subsystem Σ_a. The model reference approach has been used whereas the reference model is optimized by the finite time linear quadratic regulation technique. The pendulum motion in the passive subsystem Σ_b is indirectly controlled using the dynamic coupling with planar forward motion of subsystem Σ_a , such that satisfactory tracking of a set pendulum tilt angle can be guaranteed. Rigours theoretic analysis has been established, and simulation studies have been performed to demonstrate the developed method.