Abstract
This paper introduces MR-HVIL, a novel mixed-reality-based platform designed for on-road human-vehicle-in-the-loop validation in vehicle research. MR-HVIL leverages an optical see-through mixed reality framework to seamlessly integrate real-world driving environments with virtual traffic scenarios. Unlike traditional vehicle-in-the-loop and driving simulators, our platform enhances driver experience by combining real-world physical vehicle dynamic motions with optimized virtual content overlay, ensuring a realistic and immersive validation environment. To address the challenges of synchronization and localization in outdoor environments, we develop a real-time virtual object localization and correction algorithm that integrates vehicle sensor data, enhancing mixed reality accuracy and stability. Experimental validation demonstrates the platform’s efficacy in comparing human responses to real and virtual autonomous vehicles, offering new opportunities for human-centric validation in mixed traffic scenarios. MR-HVIL paves the way for safer, more versatile, and naturalistic testing of mixed traffic systems.
Funding Information
  • U.S. Department of Energy (DOE) Vehicle Technologies Office (VTO) under the Energy Efficient Mobility Systems (EEMS) Program

This publication has 31 references indexed in Scilit: