TRAVERSE: Traffic-Responsive Autonomous Vehicle Experience & Rare-event Simulation for Enhanced safety

TL;DR

This paper introduces TRAVERSE, a VR-based traffic simulation platform designed to study human responses to rare accident scenarios, enhancing autonomous vehicle safety research by providing realistic adverse event data.

Contribution

It presents a novel VR traffic simulation framework with lightweight hardware to facilitate research on rare traffic events and human driving behaviors in simulated accidents.

Findings

Insights into human responses during simulated accidents

A scalable, portable VR traffic simulation platform

Enhanced understanding of safety in autonomous driving

Abstract

Data for training learning-enabled self-driving cars in the physical world are typically collected in a safe, normal environment. Such data distribution often engenders a strong bias towards safe driving, making self-driving cars unprepared when encountering adversarial scenarios like unexpected accidents. Due to a dearth of such adverse data that is unrealistic for drivers to collect, autonomous vehicles can perform poorly when experiencing such rare events. This work addresses much-needed research by having participants drive a VR vehicle simulator going through simulated traffic with various types of accidental scenarios. It aims to understand human responses and behaviors in simulated accidents, contributing to our understanding of driving dynamics and safety. The simulation framework adopts a robust traffic simulation and is rendered using the Unity Game Engine. Furthermore, the simulation framework is built with portable, light-weight immersive driving simulator hardware, lowering the resource barrier for studies in autonomous driving research. Keywords: Rare Events, Traffic Simulation, Autonomous Driving, Virtual Reality, User Studies