TNO
About this position
Automated vehicles (AVs) need to safely deal with all kinds of scenarios. To assess whether an AV can do so, enormous amounts of simulations are conducted of scenarios with many varying parameters. The large number of scenario parameters make the (numerical) extrapolation of the simulation failure rate to the real-world failure rate problematic. In this thesis project, your goal is to develop methods to find the optimal sub-set of parameters to enable accurate estimation of the real-world failure rate. Are you interested in this challenge? Please read further!
What will be your role?
Automated vehicles (AVs) have become widely popular in recent years since they have the potential to bring several benefits such as increased road safety, traffic throughput and fuel efficiency. To achieve higher levels of automation, an AV must be capable of dealing with all kinds of challenging scenarios. Virtual simulations are conducted to assess whether an AV can safely cope with the different scenarios. However, even if virtual simulations are cheaper to conduct compared to physical tests, the number of virtual simulations need to be limited.
A scenario comes with many parameters that can be varied. However, to analyze all possible variations of these scenario parameters and to estimate the overall failure rate of an AV, an infeasible number of simulations is required. Fortunately, not all scenario parameters have a similar influence on the failure rate of an AV: some parameters have significant influence on the outcome of the simulations, while other parameters may only have minor influence or no influence at all. The challenge of this thesis is to arrive at a small set of parameters while only reducing the accuracy of predicting a failure slightly.
Initially, this might be simply a subset of the original parameters, but better results might be obtained when (linearly or non-linearly) mapping the original parameters to a transformed parameter sub-space.
To conduct your research, you will be provided with simulation results of a real application from an AV developer whom TNO is collaborating with. Your task is to develop methods to reduce the number of parameters and to apply your methods on a real application.
During this assignment you will be investigating the following research question:
How can the set of scenario parameters be reduced while minimizing the estimation error of the failure rate of an automated vehicle?
To answer the research question, you will:
You will work at the Integrated Vehicle Safety department of TNO on the Automotive Campus in Helmond. In this department people are working on developing software for automated driving vehicles. The developed software is tested in pilots and on the public road. The people are young, enthusiastic and driven. You will work in an open area, within your own team. One of our employees will be your mentor. He will help you to get acquainted with the department and give you guidelines for your research in order to help you to get the best out of it.
What we expect from you
We are seeking candidates with experience in Python, MATLAB, or R, with a strong preference for Python. Proficiency in English communication is also essential. We expect candidates to have an average grade above 7. The thesis fits to a 6 to 9-month project (based on full time availability). We require you to work in Helmond at the TNO office at least once or twice a week to enable you to work with our tools and to have short communication lines. We would like you to start as soon as possible.
What you’ll get in return
You want an internship opportunity on the precursor of your career; an internship gives you an opportunity to take a good look at your prospective future employer. TNO goes a step further. It’s not just looking that interests us; you and your knowledge are essential to our innovation. That’s why we attach a great deal of value to your personal and professional development. 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:
