PhD Position in Safe Planning and Decision-Making for Long-Horizon Robot Autonomy

This PhD position at Delft University of Technology focuses on advancing general-purpose AI for robotics by combining fast reactive behavior with deliberate reasoning and planning under uncertainty. Operating within the EU-funded OPERA project and the Reliable Robot Control Lab, the successful candidate will design and analyze safe planning algorithms that balance performance, safety, and real-time computation over extended time horizons. The research involves formulating robust mathematical methods and validating them through both simulation and deployment on physical robotic systems. Candidates are expected to collaborate closely with European consortium partners, contribute to open-source toolboxes, and publish academic articles while completing a doctoral degree.

• Formulate and develop advanced planning and decision-making algorithms for robots operating under environmental uncertainty.
• Investigate the trade-offs between safety guarantees, system performance, and real-time computational constraints.
• Analyze theoretical guarantees, failure modes, and robustness of the developed control and planning frameworks.
• Implement optimization-based planners and controllers using languages such as C++ or Python.
• Validate theoretical models and algorithms through rigorous testing in simulation environments and on real robotic hardware.
• Collaborate with European partners within the OPERA consortium and contribute to joint project deliverables and open-source tools.
• Disseminate research findings by authoring papers for top-tier international conferences and academic journals.