Work

Analysis of the Relationship between Road Geometric Design and Disengagements in SAE2 level Vehicles

Year: 2025

Type: article

Abstract: Vehicles with an automation level 2, according to the Society of Automotive Engineers (SAE) classification, are currently travelling on the roads. These vehicles incorporate systems that assist driving, such as longitudinal and lateral control of the vehicle. However, their operation is limited to zones and/or variables for which they are designed, known as the Operational Design Domain (ODD). Outside these zones, system disengagements can occur, leading to safety concerns and driver distrust. Previous research has attempted to analyze which factors cause these disengagements, but the relationship between system disengagements and road geometric design is still unclear. A large part of the road network has complex designs that may not even comply with the minimum thresholds established in the current design guidelines, as the case of two-lane rural roads. This research aims to analyze the relationship between road geometric design and the disengagements experienced by SAE2 level vehicles. To this end, the compliance with the guidelines -in terms of horizontal and vertical alignment- of a segment of two-lane rural road has been characterized and assessed. Using an experimental methodology, field data were collected through two SAE2 level vehicles equipped with high-definition cameras and GPS. The vehicles travelled at the posted speed limit of the road, identifying the locations where the vehicles operated autonomously and where the system underwent disengagements. Results showed that most of the disengagements occur in horizontal curves that do not meet the minimum radius criterion. These findings can serve as a basis for establishing SAE2 level failure zones, thus contributing to safer driving and increased system reliability.

Source: Transportation research procedia

DOI: 10.1016/j.trpro.2025.06.014

Authors: Griselda López, David Llopis-Castelló, Sara Moll, Víctor M. Ramos

Institution: Universitat Politècnica de València

Language: English

Topic: Human-Automation Interaction and Safety

Subfield: Social Psychology

Field: Psychology

Domain: Social Sciences

SDGs: Industry, innovation and infrastructure, Sustainable cities and communities

Open Access status: diamond

Funders: Ministerio de Ciencia e Innovación, Universitat Politècnica de València, European Regional Development Fund, Agencia Estatal de Investigación