A RISE-based asymptotic prescribed performance trajectory tracking control of two-wheeled self-balancing mobile robot

This paper proposes a robust integral of sign error (RISE) based tracking control method with asymptotic prescribed performance for two-wheeled self-balancing mobile robot (TSBR) in presence of exogenous disturbances and modeling uncertainties. First, a velocity planner is designed to provide the desirable longitudinal speed and yaw rate based on the TSBR's kinematics model. Afterwards, a modified prescribed performance function (MPPF) is devised to restrain all tracking errors of the TSBR within the predefined range without requiring the accurate initial values of tracking errors. Besides, the radial basis function neural network (RBFNN) based on minimum parameter learning approximator is utilized to attenuate the impact of exogenous disturbances and modeling uncertainties of the TSBR. Then, the MPPF and RBFNN are implanted into the RISE scheme to form an expected trajectory tracking controller for the TSBR, which can guarantee the control continuity and system asymptotic stability. Finally, comparative simulations are conducted to verify the feasibility and effectiveness of the proposed MPPF-RISE controller.