Master's thesis: Intelligent tools for generating test scenarios of the future for Autonomous Vehicles

Background
The development and deployment of autonomous vehicles (AVs) require extensive testing and validation to ensure safety and reliability. Traditional testing methods, such as real-world driving tests, are costly, time-consuming, and cannot cover all possible edge cases for modern AI based solutions. Scenario-based testing has emerged as a promising approach, supported by standards such as ISO 21448 (SOTIF) and ASAM OpenSCENARIO, which advocate systematic validation through predefined driving scenarios.

Description
One of the critical challenges in scenario-based testing is the generation and collection of valid, representative, and challenging scenarios that accurately reflect real-world driving conditions. The Synergies project (an EU funded research project) aims to contribute to this problem by creating intelligent tools and methods for data collection and scenario generation.

A challenge is when considering scenarios that involve complex human behavior, as human drivers exhibit unpredictable and varied responses to different traffic situations. Nordic driving conditions present unique challenges including weather-related scenarios, pedestrian behavior in winter conditions, and specific traffic patterns that differ from those in other regions where most AV testing occurs. Current scenario generation methods may not capture the full spectrum of human driving behaviors or may be limited in scope. There is a need for comprehensive data collection that combines human behavior modeling with systematic scenario generation to create a robust testing framework for AVs.

Key Responsibilities
The primary purpose of this thesis is to develop tools for collecting and analyzing driving data that incorporates human behavior models into scenario-based testing for autonomous vehicles. The specific goals include:

• Identify and implement representative Nordic driving scenarios
• Improve the environment using the CARLA simulator integrated with a driving rig for human-in-the-loop tests based on identified requirements
• Import AI based human behaviour models into CARLA 
• Create a systematic approach for generating test cases possibly using state-of-the-art AI techniques
• Collect driving data from multiple human drivers for scenario identification and validation

Qualifications
Candidates are expected to be enrolled in a master's program in a field related to computer science and engineering, control and mechatronics, or complex systems. Additionally, experience with Linux and Python is a merit.

Terms
As a master's thesis candidate in this project, you will work with researchers from the Dependable Transport Systems at RISE. We will provide you with the infrastructure and support to perform your thesis work. This thesis is located in Borås, physical presence is expected to some degree. Start is beginning of 2025. For an approved thesis project worth 30 credits, we pay a compensation of 39,990 SEK if one student and 30,000 SEK per student if more than one student.We encourage applications from two students who want to work together on the project.

Welcome with your application!
Last day of application; 10^th of December
Contact; Ashfaq Farooqui (ashfaq.farooqui@ri.se) and Åsa Olsson (asa.olsson@ri.se, +46102284643).