fortiss is the research institute of the Free State of Bavaria for the development of software-intensive systems with headquarters in Munich. The scientists at the institute cooperate in research, development and transfer projects with universities and technology companies in Bavaria, Germany and Europe. The focus is on research into state-of-the-art methods, techniques and tools for the development of software- and AI-based technologies for dependable, secure cyber-physical systems such as the Internet of Things (IoT). fortiss is organized in the legal form of a non-profit limited liability company. Shareholders are the Free State of Bavaria (majority shareholder) and the Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.

For a project in the Autonomous Driving domain, the Software Dependability group offers the position of a

Student Assistant – Scenario-Based testing of Autonomous Driving Systems (M/F/D)

Testing Advanced Driver-Assistance Systems (ADAS) is crucial to avoid the harm to humans but testing ADAS on real roads is expensive and time-intensive, why simulation-based testing is often a complementary strategy. One approach to reveal faults of such systems is to perform a search-based testing approach, where the behavior of the ADAS is simulated in an artificial environment for different driving scenarios. The result of a simulation run is evaluated using a criticality metric, to decide how good a scenario reveals a misfunction of the ADAS, and an optimizer is used to iteratively modify the scenarios to gather more critical scenarios.

To determine critical scenarios, we follow an evolutionary-based approach where we perform evolutionary operations to retrieve different and “better “scenario candidates. The goal is to apply the testing approach to the “Automated Valet Parking” use case, where a car is fully autonomous parked in a closed parking area, after the driver has left the car for parking.

Your tasks:
  • Support the implementation of Driving Scenarios for the Automated Valet Parking use case
  • Support the implementation and evaluation of evolutionary search-based approaches for identification of critical test cases
Your profile:
  • Completed Bachelor's degree and currently enrolled in Computer Science or a related field
  • Profound knowledge and practical experience in Object Oriented Programming, Software Engineering, Python, git
  • Ideally Knowledge in MATLAB, Simulink
  • Experience in (Multi-Objective) Optimization, Evolutionary Algorithms
  • Interest to acquire new techniques, frameworks and tools.
  • Communication skills in English, basic skills in German is a plus.
Our offer:
  • Position available for three months with a possibility for extensions.
  • An exciting and inspiring open research environment.
  • An international and dynamic work environment with highly qualified and motivated colleagues.
  • Collaboration with industrial leaders in the automated driving systems sector.
  • Varied and future-oriented field of activity.
  • Possibility to do Master thesis.
  • Space for new ideas, for assuming responsibility, as well as for personal and professional development. 
Did we catch your interest?

Please submit your application with a motivational statement, a detailed CV and a current transcript of records.

Job-ID: SD-FOCETA-SA-01-2022
Contact: Lev Sorokin