Automated Vehicles and Adverse Weather Phase 3 – Final Report

Boyapati, Rama Krishna; Eddy, Martha Morecock; Seitz, Timothy · 2023 · ROSA P / United States. Department of Transportation. Federal Highway Administration

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

The Federal Highway Administration’s Automated Vehicles and Adverse Weather Phase 3 (AVAW3) project investigates how adverse weather and road conditions impact the dynamics, operations, and sensor capabilities of automated vehicles (AVs). Motivated by the rapid advancement of AV technology and the need to understand system shortcomings in complex environments, the study focuses on work zones, signalized intersections, and lane changes. The research aims to provide data on AV performance under varying environmental stresses to inform future infrastructure and policy decisions. The methodology involved two rounds of field tests conducted at the Transportation Research Center Inc. proving grounds between 2019 and 2021. The first round (Summer FT#1) utilized production vehicles with Society of Automotive Engineers (SAE) Level 2 automation features. These vehicles were tested in baseline fair weather, wet pavement during daylight, and wet pavement at night, with crosswinds applied in specific scenarios. Scenarios included work zone lane changes with barrels, lane closures with markings, pavement markings with brake marks, and disappearing shoulders. The second round (Winter FT#2) included SAE Level 2 vehicles and a prototype SAE Level 3 automated driving system. Winter tests assessed lane keeping, right lane changes, maneuvers at green signalized intersections, and stopped car detection under baseline, snow-covered, plowed, and ice-covered conditions. Key findings reveal significant inconsistencies in AV performance across different vehicles and test runs. Perception limitations were more pronounced in winter conditions than in summer. Vehicles frequently experienced localization loss, failed to detect work zone barrels, and struggled to follow pavement markings obscured by glare or snow. Notably, the vehicle equipped with a Light Detection and Ranging (LiDAR) and high-definition map-based perception system performed better in winter conditions than the vehicle relying on a multiple-camera-based system. However, even the LiDAR-equipped vehicle had limited capabilities in perceiving certain roadway features like pavement markings. The study also observed that SAE Level 3 systems reacted to adverse winter conditions more efficiently than Level 2 systems. The significance of these findings lies in the demonstrated need for redundant sensing systems in AVs to ensure safety in adverse weather. The report highlights that current proprietary technologies may lead to driver over-confidence, as systems perform reliably in moderate weather but fail drastically in severe conditions like excessive glare or icy surfaces. The authors recommend future research into advanced testing using open-source algorithms, the integration of connected vehicle data for redundancy, and the examination of steering torque effects under critical driving conditions. The study underscores that robust AV integration requires not only improved vehicle sensors but also enhanced roadway infrastructure and data sharing mechanisms.

Key finding

Automated vehicles exhibited significant performance inconsistencies in adverse weather, with LiDAR-based systems outperforming camera-only systems in winter conditions, demonstrating a clear need for redundant sensing technologies.

Methodology

field_study

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• weather rain fog snow