Dual-Mode Model Predictive Control of an Omnidirectional Wheeled Inverted Pendulum

• Published in: IEEE/ASME transactions on mechatronics Vol. 24; no. 6; pp. 2964 - 2975
• Main Authors: Watson, Matthew T., Gladwin, Daniel T., Prescott, Tony J., Conran, Sebastian O.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-12-2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Complexity; Computer simulation; Controllers; Dynamics; Pendulums; Predictive control; Predictive models; Prototypes; Symmetric matrices; System dynamics; Systematics; Torque; Trajectory; underactuated robots; wheeled robots; Wheels
• ISSN: 1083-4435; 1941-014X
• DOI: 10.1109/TMECH.2019.2943708

This article describes the position and heading control of a novel form of omnidirectional wheeled inverted pendulum platform known as a collinear Mecanum drive. This concept uses four collinear Mecanum wheels to balance in a similar manner to a typical two-wheeled inverted pendulum while also being able to simultaneously translate directly along its balance axis. Control is performed using a constrained time-optimal infinite horizon model predictive controller, with feasibility maintained across the full reference input set. Explored in this article is the derivation of the system dynamics model and controller, a systematic approach to selection of controller parameters and analysis of their effect on control performance and complexity, and an evaluation of the controller's efficacy in both simulation and on a real-world experimental prototype for simple and complex trajectories.