Development of a vehicle dynamics model including a steering model to support a car-driver interaction PhD

Abstract

During the tire development process, its handling properties are first assessed objectively by tests on specific test benches as on track. Then, professional test-drivers perform some subjective tests on track to evaluate its performances based on their own feeling.

Here comes the importance of the vehicle-driver interactions understanding, i.e. the subject of the PhD supported by this Master Thesis, to answer such questions and adapt the tire to customer needs : "Which tire characteristic does influence this handling properties?", "Why is the driver not performing exactly the same maneuver when that tire parameter is changed?", "What does alter the driver judgment?".

Some studies ongoing in the frame of this PhD had already shown correlations between subjective and objective assessments. As a results, a qualitative model was needed to corroborate these correlations, to link them to tire characteristics and specifications, and to bring a physical explanation of the phenomena observed.

A theoretical tool, split in the three main blocks listed below, has then been implemented in MATLAB© during this internship to bring an analytical approach.

1. A body dynamics model consisting in a modified yaw-slip-roll single track model. The kinematics and compliance terms will also be integrated in this model.

2. A steering model, developed in four sub-models of increasing complexity.

3. A non linear tire model, defined following the Pacejka ’94 Magic Formula.

All together, these blocks built a robust global model able to simulate a wide range of both open and close-loop maneuvers in the linear and non linear domains. Moreover, this model helped the study of tire characteristics influence on these maneuvers by using sensitivity analyses, which leaded to highlight two tire characteristics having a big influence on subjective assessments: the rear cornering stiffness and the pneumatic trail.

Having been validated, this model will now be mostly used in a large-scale study within the Company, in the frame of Nico Pagliarecci PhD.

Lastly, note that none of the parameters value or tire constructions used in this work to obtain the results presented will be provided explicitly, to respect the Goodyear confidentiality policy.