Robust hierarchical sliding mode control of a two-wheeled self-balancing vehicle using perturbation estimation

•First, the mathematical modelling of a manned two-wheeled vehicle is derived by using state space equation and the included perturbations are described in detail. Perturbation estimation technique is further adopted to provide with online estimations for perturbations.•Second, a hierarchical sliding mode control (HSMC) is used to handle the coupling effects and uncertainties of the underactuated two-wheeled vehicle, thus both balancing and velocity control purposes are achieved. Also, since most of the underactuated control systems only discuss the regulation not tracking problems, it is the first time that the HSMC is developed for the vehicle to achieve velocity tracking while balancing the chassis with driver around the upright position. The stability proof of the proposed scheme with PE is presented in detail.•Third, a group of experimental results for different driving situation (regulation task with slight and vigorous push, velocity tracking and balancing control on flat and inclined surfaces) are given to show the robust control performance of the proposed scheme, with convincing comparisons with the HSMC, adaptive HSMC and conventional linear feedback control. The readers can clearly see the significant improvements on balancing and velocity tracking performances by the proposed control. This paper presents the design and implementation of hierarchical sliding mode control (HSMC) with perturbation estimation (PE) technique on a two-wheeled self-balancing vehicle (TWSBV), to simultaneously realize real-time balancing and velocity tracking control purposes. Considering the fact that the TWSBV system is a typical second-order underactuated system with one controlled actuator and two required control objectives, two sliding surfaces constructed by the velocity and tilt angle information are first designed and an HSMC is proposed to simultaneously achieve both balancing control and velocity control. In order to further enhance the ability of disturbance rejection of the HSMC control, the PE is used to assist with the proposed control for estimating the perturbations online such that the uncertainty bound information is not required in the control design. The excellent balancing and velocity tracking performance can be well achieved even under external disturbances. The effectiveness of the proposed control is verified by a group of comparative experimental investigations on a real TWSBV.