HMI and VR for trustable and comfortable driving on autonomous vehicles - A dynamic driving simulator for testing multimodal interfaces

Automated systems, and autonomous vehicles (AVs) in particular, promise to profoundly change everyday life. In recent years, lot of research activities and progresses have been made in this field, which led to growing expectations that fully autonomous vehicles (FAVs) will reach the market within the next few decades. However, in order to make this technology a kind of standard, it is necessary that the passengers trust the AVs’ safety and reliability. The objective is to reach a high-enough level of trust to let the user spend his or her time on non-driving-related tasks. For this reason, when referring to AVs trust is one of the main human-machine interaction issues. Starting from this premise, the goal of this thesis is to study how the possible interfaces inside the vehicle can influence the occupants’ perception of safety. To achieve this goal, study of the state of the art in the field was first conducted, summarizing and analyzing the interfaces proposed in the literature. Afterwards, in order to assess the effectiveness of the proposed interfaces, it was necessary to design and implement a synthetic 3D environment, in which the complex and realistic behavior of traffic and pedestrians was simulated. Furthermore, to make the simulation easily manageable, a graphic editor that enable the creation and modification of the paths traversed by 3D elements was created. Finally, using immersive virtual reality and an inertial platform, a simulator that can emulate a journey aboard an AV within the above environment was designed and built. The vehicle has been programmed to behave according to SAE L3, i.e., having both autonomous and manual driving capabilities. Particular attention was paid to the vehicle’s path, which includes a series of situations designed to test the users’ trust and the effectiveness of the interfaces. The outcomes of the thesis will support experimental activities that are planned as future work.