PhD: Safe, Long-Horizon Robot Planning Under Uncertainty

TU Delft; Delft; On-site; EUR 34,000 - 43,000; Full time

Benefits offered by this job

Customisable compensation package

Discounts on health insurance

Relocation assistance

Dual Career Programme

Job summary

TU Delft is seeking a PhD candidate to develop planning and decision-making methods that enable safe, reliable robot behavior over long time horizons, even with partial knowledge of the environment. The research sits at the intersection of model-based planning and learning, with emphasis on formal guarantees and failure modes. You will work under Prof.

Robert Babuška and Dr. Laura Ferranti in the Cognitive Robotics Department, collaborating across Europe on simulation, perception, and tooling.

Qualifications

• MSc in Systems and Control, Robotics, Applied Mathematics, or closely related field.
• Strong interest in safe planning and decision-making for robots.
• Rigorous, theoretically grounded research, validated in simulation and on real robots.

Responsibilities

• Develop planning and decision-making methods for long-horizon robot safety.
• Analyze guarantees and failure modes of proposed methods.
• Collaborate with TU Delft and OPERA partners; publish results.

Skills

Control theory

Optimization

Motion planning

Probability & statistics

Dynamical systems

English communication

Education

MSc in Systems and Control / Robotics / Applied Mathematics

Tools

acados

CasADi

ROS / ROS 2

Python

C++

Job description

TU Delft is seeking a PhD candidate to develop planning and decision-making methods that enable safe, reliable robot behavior over long time horizons, even with partial knowledge of the environment. The research sits at the intersection of model-based planning and learning, with emphasis on formal guarantees and failure modes. You will work under Prof.

Robert Babuška and Dr. Laura Ferranti in the Cognitive Robotics Department, collaborating across Europe on simulation, perception, and tooling.