Master Thesis “Automated Road Damage Avoidance using Reinforcement Learning Concepts”

Road damage is a problem that not only reduces driving comfort but sometimes also threatens driving safety. At present, the information about road damage (e.g., position, shape, and severity) can be transmitted to highly automated vehicles via infrastructure to vehicle (I2V) communications. Road damage has irregular shapes, which presents a challenge for conventional trajectory planning algorithms. Reinforcement Learning (RL) has obtained more attention recently as it does not rely on preprogrammed logics and has the potential to solve more general decision-making problems. As part of a dynamic and international team and within the scope of an international research project, the thesis student shall help with the development of a first-of-a kind use case for an automated driving vehicle concept for avoidance of road damages using RL concepts and will have chance to contribute to dissemination of the implementation results.

TASKS

• Understand RL concept (Reinforcement Learning) and existing benchmarks
• Define the RL problem for road damage avoidance task with specific constraints
• Test and compare state-of-art RL algorithms
• Report as dissertation and assist in publication of the results

PROFILE

• Student of Computer Science, Electrical engineering, Mechatronics, Telematics, Robotics, Vehicle Engineering, or another related field
• Interest in Reinforcement Learning (RL)
• Experience in Python programming
• Fluent in English

OFFER

• Collaboration and contribution in an engaged, dynamic team.
• Interesting work in an international research center with a diverse multi-national team.
• Paid Thesis.
• Mentoring program for new employees.
• Corporate Events.

APPLY NOW and JOIN OUR TEAM

Your Contact:

Barbara Cappello / Recruiting / + 43- 316- 873- 9028