On-road Mixed Reality Autonomous Driving Simulator and its Validation

Abstract

As self-driving car technology is fully grown, autonomous vehicle research is moving toward building more human-centric interfaces and accountable experiences. Driving simulators avoid many of the ethical and regulatory concerns about self-driving cars and thus play a key role in testing new interfaces or autonomous driving scenarios. However, apart from validity studies for manual driving simulation, the capabilities of driving simulators in imitating the experience of self-driving cars have not been widely investigated. In this paper, we introduce new methods and tools for simulating driving via virtual vehicle in real-world environments. The proposed system fully exploits the benefits of vivid sensory information in real road environments without sacrificing the unlimited potential of environment construction for it. Additionally, we build six self-driving simulation platforms with varying levels of visual and motion fidelities, ranging from a screen-based in-lab simulator to the novel mixed-reality on-road simulator we propose. With a user study, we compare the sense of presence, simulator sickness for each simulator composition as well as its perceptual visual and motion fidelities. Our novel in-vehicle mixed reality simulator showed the highest fidelity and presence. Our findings suggest how visual and motion configurations affect experience in autonomous driving simulators.