Approach to a holistic rider input determination for a dynamic motorcycle riding simulator.

In order to shorten development cycles, reduce prototype costs and prevent test-persons from being harmed in experiments, it is more and more common, to use riding simulators in the vehicle development process. A key issue working with such simulators is to provide the rider with the best possible presence – the feeling of being inside a certain (virtual) scenario – so that this rider will act and react in the simulation environment as he would do it in a real life situation. While throttle, clutch, brakes and gearshift are trivial to implement in a motorcycle riding simulator, two other rider inputs – steering and leaning – are part of closed loop systems which necessitate the rider both as sensor and actuator. Thus it is of utmost importance, that the mechatronical systems that determine the rider’s inputs and provide the motion/force-feedback comply with the rider’s expectation. This paper presents an approach on how to design a dynamic motorcycle riding simulator that enables the rider to utilize both steering and leaning controls in order to maneuver a virtual motorcycle that is not limited to e.g. force measurements at the foot pegs or other rider-specific behavior. Therefore a method to determine the rider-induced roll torque is being presented. It is shown, that this method is able to differentiate between quasi static riding styles like lean-in or lean-out, as well as to determine higher frequency influences of the rider to the vehicle’s roll motion. Through the used real time multi body simulation, a coupling to the steering behavior is implemented.