LQR-based starting control method for motorcycle using front-wheel drive

抄録

This study discusses self-driving motorcycles that use front-wheel drive. The ability of the motorcycle to maintain stability at high speeds is owing to the gyroscopic effect. However, at lower speeds, when the gyroscopic effect is reduced, human riders need to shift their weight or steer the motorcycle to prevent it from falling. To ensure the safety of self-driving motorcycles, we simulated an autonomous motorcycle accelerating from 0 to 1.5 m/s without falling. During the acceleration, we controlled only the front-wheel drive and kept the steering angle fixed at 0.52 rad (30 deg). By fixing the steering angle to 0.52 radians, the centrifugal force was generated as the restorative force of the motorcycle, and enabled precise control of the acceleration response of the motorcycle and enhanced its stability against disturbances. The primary goal of this study is to achieve stability while the motorcycle accelerated from 0 to 1.5 m/s. To accomplish this, we introduced a two-stage control design method using Linear Quadratic Regulator (LQR) control. This method enables us to achieve the desired acceleration response and improved stability in the face of disturbances. For our simulations, we employ SPACAR, a program based on the finite element method, to derive the linearly approximated state-space representation. The simulation results demonstrated controlled motorcycle realizing good accelerating performance and stability against disturbances.