Published in IET Control Theory and Applications Received on 13th September 2010 Revised on 18th February 2011 doi: 10.1049/iet-cta.2010.0530 ISSN 1751-8644 Velocity control realisation for a self-balancing transporter C.-H. Huang 1 W.-J. Wang 1 C.-H. Chiu 2 1 Department of Electrical Engineering, National Central University, No. 300, Jhongda Rd., Jhongli City, Taoyuan County 32001, Taiwan 2 Department of Electrical Engineering, Yuan Ze University, No. 135, Yuan-Tung Rd., Jhongli City, Taoyuan County 32003, Taiwan E-mail: wjwang@ee.ncu.edu.tw Abstract: This study proposes the motion control design of a real self-balancing transporter (SBT) which is fabricated in the laboratory. Based on the system-on-a-programmable-chip (SoPC) developmental architecture, two controls are implemented to achieve the objectives of balanced standing and moving with a constant velocity for the SBT. Based on the Takagi–Sugeno (T–S) fuzzy model of the SBT, the parallel distributed compensation (PDC) controller is employed to achieve the mentioned controls. Then an extra Mamdani If–Then fuzzy rule base (FRB) is built to cooperate with the PDC control such that the SBT not only can stand with balance, but can also move forward with a desired constant velocity ‘quickly’. The maximum emphasis is that there is a novel idea in the design of the If–Then FRB which is the relationship between the moving velocity error and the desired inclination angle of the SBT. Owing to the aids of the extra If–Then FRB, the time of the controlled response to reach the desired state is obviously shortened. This is the significant contribution of the study. Finally, both the fuzzy PDC and the If–Then FRB are realised in the SoPC platform for the real SBT. The performance and merit of the proposed control scheme is exemplified by conducting computer simulations and practical experiments. 1 Introduction Recently, many studies have investigated the kind of self- balancing transporter (SBT) (or called two-wheel vehicle system) that has only two coaxial wheels for its balance. The mechanism conception of SBT evolves from the typical rail-cart inverted pendulum systems with underactuated dynamics [1, 2]. Usually, such an SBT system is constructed by mechatronic integration and control techniques. Previous researches focusing on the modelling and control designs for the SBT are reported as follows. Baloh and Parent [3] concentrated the autonomous mobility of SBT for an urban transportation system without the driver’s posture control. Chiu [4] implemented the balanced standing control of an SBT by using the adaptive output recurrent cerebellar model articulation controller. Fiacchini et al. [5] proposed a physical dynamic model of a personal transportation vehicle and developed the linear and non-linear controllers for balance control. Grasser et al. [6] developed a dynamic model for designing the mobile controllers using the Newtonian approach and linearisation method. Ha and Yuta [7] presented a state-space model of SBT and designed a linear state feedback and feedforward controller for its trajectory tracking control. Later, the control algorithm of SBT proposed by Ha and Yuta [7] was continuously developed to perform baggage transportation and navigation task [8]. Jeong and Takahashi [9] applied a state feedback control method for mobile controls of the SBT-type assistant robot. Jung and Kim [10] created a mobile SBT using the neural network control combined with a proportional – integral – derivative controller. Based on a non-linear dynamic model using the Lagrange technique, Karkoub [11] designed a m-synthesis controller for passenger transportation of SBT. Kim et al. [12, 13] derived a 3D dynamics of an SBT and analysed the stability of the dynamics, including longitudinal and yaw movements. Li and Luo [14] proposed the adaptive robust dynamic balance and motion controls for an SBT. Nawawi et al. [15] described a hardware design of SBT and implemented a pole-placement controller for its stabilisation. Pathak et al. [16] analysed dynamics of an SBT, and then applied the partial feedback linearisation concept for its velocity and position control design. Salerno and Angeles [17] proposed a mathematical model of an SBT along with a controllability analysis. Tsai et al. [18] presented an adaptive control using radial-basis-function neural networks for a two-wheeled scooter. Summarising the aforementioned reviews, the control problem of an SBT system is indeed an interesting research topic and is worth being studied continuously. Herein, two fuzzy control methods are suggested for an SBT control system. The first control design is the Takagi– Sugeno (T – S) model-based approach [19]. A non-linear dynamic system can be represented by a T–S fuzzy model which is described by a set of rule-type local linearisations. IET Control Theory Appl., 2011, Vol. 5, Iss. 13, pp. 1551–1560 1551 doi: 10.1049/iet-cta.2010.0530 & The Institution of Engineering and Technology 2011 www.ietdl.org