Internship | Traffic rules to formal logic: advancing the automation of rule interpretation for automated driving

Internship | Traffic rules to formal logic: advancing the automation of rule interpretation for automated driving

About this position

Automated vehicles (AVs) rely on clear driving rules to operate safely and predictably. These rules guide both system design and the assessment of how well an AV behaves in real traffic. They also help ensure the vehicle is socially compliant—although in some situations, bending a rule may actually be the more socially appropriate choice. Traffic rules are written for people, not machines. In everyday driving, drivers rely on experience, context, social norms and shared understanding to apply them. Today, formalizing traffic rules from natural language is still a manual, time‑consuming task. Traffic rules are often long and vague. They also differ across regions and jurisdictions. Engineers must interpret the text, decide on what is relevant in the text and translate it into logical structures. This takes time and does not scale well.

What will be your role?

The assignment is to advance the TNO approach that converts traffic and driving rules written in natural language into precise, structured formal rules.

Proposed approach:

• Review existing methods and prior work completed within the different projects.
• Identify linguistic structures common in traffic and driving rules.
• Break down rules into context, actors, conditions, and expected behaviors.
• Develop a consistent structure for expressing a formal version of each rule.
• Translate the formal version into a clear logical statement using mathematical expressions.
• Handle exceptions and edge cases that naturally appear in rules.
• Evaluate the clarity, consistency, and completeness of the translated rules.
• Test the method on a diverse set of rules from different sources.
• Recommend improvements to make the process more systematic and scalable.

Expected Outcomes:

• Methodology for turning natural‑language rules into formal representations and corresponding mathematical statements.
• Set of translated rules that demonstrate the method across different rule types.
• Practical insights on how to extend or automate parts of the workflow.

What we expect from you

Candidates should have a background in Artificial Intelligence, Data Science, Computer Science, Systems and Control Engineering or a related field. They should be proficient in programming languages (e.g., Python, basic data processing). Experience with formal methods or mathematical reasoning, NLP libraries or language models is preferred. The internship/ master thesis is expected to last 6 to 9 months. We require you to work in Helmond at the TNO office at least twice a week to enable you to work with our tools and to have short communication lines.

What you'll get in return

You will, of course, be properly supervised during your work placement and be given the scope for you to get the best out of yourself. Furthermore, we provide:

• A highly professional, innovative internship environment, within a team of top experts.
• A suitable internship allowance (615 euro for wo-, hbo- and mbo-students, for a full-time internship).
• Possibility of eight hours of free leave per internship month (for a full-time internship).
• A free membership of Jong TNO, where you can meet other TNO professionals and join several activities, such as sports activities, (work-related) courses or the yearly ski-trip.
• Use of a laptop.
• An allowance for travel expenses in case you don’t receive an OV-card.

TNO as an employer

At TNO, we innovate for a healthier, safer and more sustainable life. And for a strong economy.