TU Delft: PhD: Safe Planning for Long-Horizon Robot Autonomy
This is a PhD position in the Cognitive Robotics Department at TU Delft. It sits in the Reliable Robot Control Lab, under Prof. Robert Babuška and Dr. Laura Ferranti, and is part of the EU-funded OPERA project on general-purpose AI for robotics. The work develops planning and decision-making methods that let robots act safely over long time horizons when their knowledge of the environment and of people is incomplete.
What you'll do
- Combine model-based planning and control with learning-based models, balancing safety, performance and real-time computation
- Formulate the methods and analyse their guarantees and failure modes
- Validate the methods in simulation and on real robots
- Collaborate with OPERA partners across Europe on simulation, perception and the OPERA open-source toolbox
- Travel to partner meetings and attend regular consortium meetings
Who can apply
- A MSc degree in Systems and Control, Robotics, Applied Mathematics or a closely related field
- A strong interest in safe planning and decision-making for robots, at the intersection of planning, control and learning
- Strong analytical and mathematical skills, in particular in optimization, probability and statistics, and dynamical systems
- Excellent written and spoken English
- Preferred: hands-on experience with optimization-based planners or controllers (for example MPC with acados or CasADi) in C++ or Python
- Preferred: experience with safe or robust control under uncertainty, learned dynamics models, uncertainty quantification, or real-robot deployment with ROS or ROS 2
How to apply
Apply through the application button on the TU Delft page by 15 November 2026. Upload a letter of motivation, a updated CV, your Bachelor and Master transcripts, and a copy or draft of your MSc thesis (or a link to it). Applications can be addressed to Laura Ferranti and Robert Babuška.
Official page: careers.tudelft.nl/talentcommunity/apply/1375678657/?locale=en_US