F2010D071
Rider's Behaviour Modelling in Motorcycle Dynamic Simulations
The interdisciplinary nature of the motorcycle-rider system makes the modelling and simulation process complex. Moreover, pre-crash and crash situations make its behaviour even more complicated. Developing a well-defined methodology linking multi-body modelling, finite elements models, control theory, models based on experimental data from real tests/scenarios is one of the ways to solve this problem, and also one of the aims of the E.C. Marie Curie Project, MYMOSA (MotorcYcle and MOtorcyclist SAfety MRTN-CT-2006-035965). A part of the conducted research is developing motorcycle-rider model, which plays a key role in the system level motorcycle road behaviour simulations. In contrast to the driver model of a four wheels vehicle, which are often modelled as constant mass or even omitted without significant influence on vehicle dynamics, the rider's behaviour in motorcycle or more generally, powered two wheelers (PTW) vehicle analysis, has to be taken into account. Proper motorcycle steering and stabilizing requires not only changes in handlebar's orientation (changes of the steering angle) but also changes in the rider's position, (changes of the rider's roll angle). Many PTW models currently used in simulations define riders as one degree of freedom (DOF) inverse pendulum (IP). The rider's body parts are modelled mainly for visualisation purposes. In some cases (i.e.: simulations of the heavy motorcycles) such models allow for the simulations of standard road manoeuvres but in pre-/crash situations or small PTWs' simulations, obtained results can be inaccurate. In presented paper, two multibody models are used: motorcycle model consisting of 6 rigid bodies and 11 DOF and rider model which consists of 16 bodies and 14 joints. To control motorcycle-rider system, control algorithms were developed and implemented in Matlab/Simulink environment. The created algorithms were divided into two main parts: one responsible for motorcycle control and second responsible for rider's position control. Based on signals from multibody models, motorcycle controller calculates required rider's roll angle, orientation of the handlebar and speed profile of the motorcycle. Next based on motorcycle's and rider's properties (actual positions and orientations of the elements, masses of the rider's body parts), calculated handlebar and rider's orientation, the position of each rider's body is calculated. The paper presents a process of developing human body model for motorcycle dynamics simulations, which focuses on solving inverse kinematic problem for such system. Next implementation process to Matlab/Simulink environment and the results of the dynamic simulations motorcycle-rider system are presented.
This abstract is supplemented by a PDF, which can be viewed here.
Session: Mixed Topics in Safety