JOB DESCRIPTION

Neural Radiance Fields (NeRFs) and Gaussian Splatting have revolutionized 3D scene reconstruction and novel view synthesis, enabling real-time, high-fidelity rendering of a scene from an arbitrary viewpoint, enabling the manipulation of the scene. In our previous work, Neuro NCAP, we used these techniques to create crash test simulators for autonomous vehicles that can simulate actions in closed-loop by synthesizing novel viewpoints of the ego vehicle. Building on these breakthroughs, this PhD project at TU Delft explores AI-driven methods to enhance closed-loop simulation for safety-critical scenarios. A key focus is developing learned simulators that generate radar and lidar data from camera sensors. Additionally, the research will investigate end-to-end prediction and planning approaches that integrate radar-based perception and are trained in closed-loop, moving beyond traditional modular robotics pipelines to create more robust and scalable autonomous systems.
This fully funded, four-year PhD position at the Intelligent Vehicles Section of TU Delft. The research is part of the EU Horizon MOSAIC project and is conducted in partnership with NXP, a leading chip manufacturer. Your results will be published in top tier conferences like CVPR, ICCV, ECCV, ICRA and NeurIPS. For your work you will have access to the compute resources of TU Delft, ranging from personal machines, to shared GPU servers, the Delft AI Cluster that is shared across departments, as well as DelftBlue, which is one of the top 250 supercomputers in the world. Your primary supervisor will be Dr. Holger Caesar.

JOB REQUIREMENTS

• We are seeking PhD applicants with an interest in performing cutting edge research in an active and exciting research area.
• Prior experience working with Neural Radiance Fields or Gaussian Splatting.
• Prospective applicants should have a strong academic record with a solid background in Machine Learning (Deep Learning, generative models, diffusion models).
• Knowledge in sensor data processing and radaris a plus.
• Good programming skills (Python) and knowledge of deep-learning frameworks (PyTorch) are expected.
• A certain affinity towards turning complex concepts into real-world practice (i.e. vehicle demonstrator) is desired.
• The successful candidate is expected to be able to act independently as well as to collaborate effectively with members of a larger team and supervise Master and PhD students.
• Good English skills are required.
• The work takes place in Delft, The Netherlands, but allows for working two days per week from home.

Please refer the Job description for details