Robust stabilization of high-speed oscillations in single track vehicles by feedback control of gyroscopic moments of crankshaft and engine inertia.

摘要

To analyze the problem of high-speed oscillations in motorcycles we have extended mathematical models of motorcycles to describe the motions in straight running, steady state turning, and transitory maneuvers. This model also includes the dynamics of the engine, as it is attached to the frame by elastic mounts. The model is nonlinear and has 18 states, including tire dynamics, engine dynamics, and important structural properties of the frame. Equations of motion along the generalized quasi-coordinates are derived using modified Lagrange equations. This model clearly demonstrates the experimentally observed oscillatory behaviors.; We have proposed a Multi-Input Multi-Output feedback control system to effectively control oscillatory motion of motorcycles at speeds. Measured signals can be roll rate and yaw rate or alternatively, steering angle and lateral acceleration of the vehicle. Controller outputs are signals to active engine mounts. By feedback control of engine's orientation with respect to the frame, this controller stabilizes the vehicle in a wide range of speeds well beyond the current speed record for single-track vehicles. Special challenges in design of the controller are that the engine displacement relative to frame must remain small, and controller must be robust with respect to significant variations in vehicle's forward speed. It is well known that chassis torsional rigidity is an important factor in steering stability of single-track vehicles. We have studied the issues related to structural rigidity of the chassis and conclude that the feedback configuration proposed here can also be used to design vehicles with super light flexible chassis. We present preliminary results showing that a controller can be designed so that it will compensate actively for lack of rigidity in the chassis. Robustness properties of the closed loop system to perturbations caused by variations in forward speed, acceleration/deceleration, and dynamic uncertainties are studied. Since the controller is designed based on a lateral dynamic model for a vehicle in straight running condition, it is necessary to show robustness of the controlled system to variations in the roll angle for a vehicle in cornering condition. This is accomplished by nonlinear simulation. Simulation results clearly demonstrate the advantages of this design in straight running, steady state turning, and transitory maneuvers. (Abstract shortened by UMI.)