Trajectory Tracking and Balance Control for Unmanned Motorcycles with a Hierarchical Framework

Abstract

Unmanned motorcycles indicate outstanding maneuverability and adaptability under complex and narrow terrains due to their compact structure and platform pattern. However, autonomous driving for such non-coaxial two-wheeled platform faces two core challenges needed to be urgently dealt with: balance and trajectory tracking capacity. To address such a problem, this paper attempts to design an advanced control scheme for coordinating balance and trajectory tracking control without additional mechanical auxiliary devices. To be specific, a simplified the unmanned motorcycle dynamic model based on front-wheel steering is constructed via the moment of momentum theorem as well as linearization processing. An MPC-based trajectory-tracking controller is developed at the kinematic layer, with an SF-based trajectory generator introduced as a benchmark; an NTSM balance controller featuring finite-time convergence is designed at the dynamic layer, with a conventional SMC controller adopted for comparison. Finally, simulation comparisons under two operating conditions verified the feasibility and effectiveness of the proposed MPC+NTSM control scheme in trajectory tracking and balance-related performance. Besides, the effectiveness of the proposed balance control method was confirmed via the physical prototype experiment.