Vehicle manufacturers strive for an increasingly efficient and faster development process. Although computer-aided engineering has made significant progress toward a fully virtual development process, a challenge remains in integrating human subjective feedback to fully close the virtual development loop. Subjective assessment of ride and driving characteristics are still very important traits of a passenger car. Moving-base driving simulators have the ability to introduce the human into the virtual development loop, thus enabling subjective assessment of virtual vehicle models. Such an introduction has the potential to significantly speed up the development process and at the same time save resources by avoiding physical testing and providing informed decisions in the early phase of vehicle development cycles. The challenge to do so lies in the possibility to evaluate a vehicle in a driving simulator, which is highly dependent on the motion cueing.

Motion cueing algorithms are used to map the vehicle motion into the confined workspace of a driving simulator. As of today, these algorithms are still often tuned and evaluated subjectively. The challenge with this approach is that it does not guarantee the fidelity of the cueing and it needs physical vehicles to be compared with. This work thus focuses on the objective development and evaluation of motion cueing, which potentially could enable high fidelity motion cueing in the early stages of the vehicle development process, when prototypes are not available. This is very important for winter testing since the testing is challenging with regards to ambient conditions, the limited testing season and the increasing need to speed up the development process.

The goal of this work is to move towards an objective approach to cueing evaluation based on physical models combining vehicle model, simulator, and human. Therefore, this thesis presents an objective methodology to motion cueing evaluation and development. Based on the state-of-the-art review, this work addresses the need for simple linear models to evaluate the fidelity of motion cueing algorithms. The linear model is applied to the problem of positioning the longitudinal axis of rotation of the simulator cabin and shows promising results when compared to time series-based optimisation and subjective assessment. Furthermore, using the same model to improve the motion cueing by introducing tilt coordination shows that even though the immersion is improved, the tilt coordination changes the perceived vehicle characteristics. To objectively evaluate different yaw cueing strategies in winter conditions, a more detailed human model is introduced that extends the state-of-the-art vestibular organ models by introducing gaze stabilisation using a model of the vestibulo-collic reflex. The cueing evaluation indicates the potential of separating slip angle feedback from the high-pass filtering of motion cues, as well as the advantage of using the vehicle’s motion as a target for cueing optimisation rather than the human vestibular response in winter handling evaluation.

By addressing the inherent skewing of vehicle characteristics in motion cueing and suggesting improvements to the evaluation and cueing strategies, this work contributes to the possibility of virtually evaluating the vehicle dynamic characteristics in driving simulators under winter conditions.