Tour de Tech: Iterating Toward a VR CAVE Test Environment for Investigating Cyclists Intentions

This paper introduces an innovative VR CAVE test environment designed to investigate cyclists' intentions in virtual scenarios. While intention prediction is the long-term application, this study focuses on the technical feasibility and user perception of the system. The evaluation reveals that the system already delivers an engaging and enjoyable virtual cycling experience, supported by the use of a real bicycle and head tracking. Results indicate that additional sensory stimuli significantly enhance the realism and Ecological Validity of the environment, although perceived control and overall presence still require improvement. To address the identified technical bottlenecks, planned hard-and software are discussed. These developments will enable systematic, reproducible, and safe investigations of cyclists' behavior and intention. This data will help improve trajectory planning in autonomous vehicles and advance the mobility transition.