Internship | A performance assessment study of point-cloud based localization algorithms with a real-time dynamic – TNO – Helmond


About this position

At TNO, we develop software for Automated Driving vehicles. Localizing the vehicle is a critical and fundamental component of any autonomous driving architecture. Specifically, LiDAR-based localization techniques have shown huge potential for centimeter-level localization accuracy; however, deciding the best algorithm for this purpose remains an open question. Such an algorithm must be able to, given an input point cloud (yellow points in the figure), find the vehicle position within a pre-existing point cloud (map, white points in the figure). In a dynamic environment, the input point cloud will contain dynamic objects such as cars and pedestrians which are not part of the ground truth map. The goal is to compare the localization performance of the legacy approach and an approach considering a real-time dynamic points remover.

What will be your role?

The goals of this internship are twofold:
1. Define metrics/KPIs for point cloud localization performance assessment in the autonomous vehicle domain
2. Compare (in real-time) a state-of-the-art point cloud-based localization algorithm and the same algorithm extended with the online dynamic point remover.
Thus, the open question is: what is the performance increase, in point-cloud localization, when extended with a real-time dynamic point remover?
Proposed approach:

  • Literature research on state-of-the-art point cloud localization algorithms with open-source packages in ROS (1 weeks)
  • Literature research on state-of-the-art online dynamic point removers (1 weeks)
  • Define metrics/KPIs to assess the performance of point cloud localization (0.5 weeks)
  • Implement the point cloud remover (2.5 weeks)
  • Test and assess the performance of the localization packages in a simulation environment (ROS) in real-time and with real data (2 weeks)
  • Report your findings (2 ~ 4 weeks)
  • What we expect from you

    What do we require of you?
    We require that you are familiar with the general functioning of an autonomous car. It is important that you are knowledgeable in C++, Linux and ROS or you are willing to learn as this is the means for testing and comparison of the algorithms. The internship fits into a 3 month project (based on full time availability).

    What can you expect of your work situation?
    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. (more info on the department: ). 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. The way of working is hybrid, normally 3 days at the office and 2 days at home.

    Related videos
    localization video

    What you’ll get in return

    You want to work on the precursor of your career; a work placement 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. Naturally, we provide suitable work placement compensation.

    Lees hier meer

    Deel deze vacature: