ROAD: The Road Event Awareness Dataset for Autonomous Driving
Year: 2025
Type: article
Abstract: The development of autonomous vehicles (AVs) requires a thorough comprehension of dynamic roadconditions in addition to object identification. By advancing from static object recognition to dynamicevent understanding, the Road Event Awareness Dataset (ROAD) presents a fresh standard for AVsituational awareness. The rich semantic triplets (Agent, Action, Location) that ROAD is labeled withacross more than 122,000 video frames allow models to understand how traffic players interact with theiractions in different spatial situations. This study provides a thorough review of the ROAD dataset, talksabout how useful it is for improving road event recognition, and investigates how it may be used for AVdecision-making, event prediction, and continuous learning. ROAD establishes the foundation for saferand more intelligent autonomous driving solutions by promoting context-aware technologies.
Author: A. Mamatha, P.Venkatesh
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6930571957
FWCI:
Cites:
Cited by:
Related to: