Identification of Critical Driving Scenarios for
Performance Assessment of Long Combination Vehicles

The Swedish National Road and Transport Research Institute (VTI) has an opening for a master thesis student in Linköping at the Vehicle Systems and Driving Simulation unit (FSK). The objective of the thesis project is to identify critical driving scenarios for performance assessment of long combination vehicles.

VTI

VTI is an independent and internationally prominent research institute in the transport sector. Its principal task is to conduct research and development relating to infrastructure, traffic and transport and its operations include all modes of transport. VTI has about 240 employees and is located in Linköping (head office), Gothenburg, Stockholm and Lund.

Background

In December 2023, the Swedish government introduced two long combination vehicles (LCVs), A-double and AB-double, on the selected road network across Sweden. LCVs, as their name implies, are heavy vehicles composed of multiple units connected in a series, with a total length exceeding 25.25 meters. Before this decision to introduce them on Swedish roads, a few LCVs had been granted exemptions in specific road networks for their trials in the last decade. These trials facilitated the collection of naturalistic driving data (NDD) in real-world traffic conditions. The data consists of GPS and IMU measurements obtained from two different units of the considered vehicles. The NDD helps to answer one of the important questions related to these vehicles, i.e., how LCVs perform in real-world traffic. Previous research on this data has assessed the LCV performance in predefined scenarios such as lane changes, roundabouts, intersections and curves. First, the intervals in NDD corresponding to the above scenarios are extracted. Subsequently, their performance is assessed in the extracted intervals using Performance-based standards (PBS) measures such as rearward amplification and offtracking. While valuable, this approach may overlook other critical scenarios in NDD which are not part of the predefined scenarios and where the performance of the vehicle is sub-optimal.

Purpose of thesis

The purpose of this thesis is to identify critical driving scenarios without explicitly targeting predefined scenarios, and thereby, evaluate the performance of the vehicle in the identified scenarios.

Work outline

The thesis work is planned to start in January 2025 and includes the following tasks:

• Familiarize with existing methodologies used for the assessment of performance of LCVs.
  
• Review literature that deals with identifying critical scenarios from the driving data.
• Develop new methods or improve existing methods to identify critical scenarios from the driving data of LCVs.
• Conduct a performance assessment in the scenarios identified by the developed method and compare the performance with previous studies.

Qualification

• Educational background in the areas of vehicle engineering, computer engineering or a similar area.
• Experience in programming and simulation.
• Ability to work independently and to take initiative.

Application instruction

Send your application with a cover letter, CV and transcripts to abhijeet.behera@vti.se