PhD Position Safe GPAI Planning, Decision-Making and Active Perception for Reliable Robot Autonomy

Join TU Delft and the EU-funded OPERA consortium to develop safe GPAI-based planning, decision-making and active perception methods for reliable robot autonomy.

In this PhD position, you will develop methods for safe GPAI-powered planning and decision-making for autonomous robots. Your research will focus on the interface between model-based control, learning-based decision-making, and active perception.

Robots operating in dynamic, real-world environments must make reliable decisions despite uncertainty, incomplete perception, changing conditions, and the presence of people. In the EU-funded OPERA project, TU Delft contributes to the development of General-Purpose AI for robotics by combining fast reactive behavior with more deliberate reasoning, planning, and uncertainty-aware decision-making.

You will investigate how robots can decide when to act reactively, when to plan over longer horizons, and when to actively gather additional information before executing a task. This directly supports OPERA’s work on GPAI-powered planning and decision-making in complex and dynamic environments, where robots must combine hybrid data-driven and physics-based architectures, safe navigation envelopes, contingency planning, and active perception.

A central part of the project will be active perception for reliable autonomy. You will study how robots can select informative viewpoints, fuse multimodal sensor data, estimate uncertainty, and adapt their safety margins or plans when perception is incomplete or ambiguous. This connects to OPERA’s work on uncertainty-aware and safe active environment perception, where uncertainty estimates guide safety heatmaps, adaptive sensing, and safety-aware navigation and manipulation.

The core of the research will be safe planning and decision-making under uncertainty, with active perception as a mechanism for reducing uncertainty before or during task execution. Depending on your background, this may involve model predictive control, trajectory optimization, uncertainty-aware planning, sensor fusion, viewpoint selection, contingency planning, or selected elements of safe learning.

You will collaborate with OPERA partners developing perception, world models, simulation tools, and GPAI components, and your work will contribute to the OPERA open-source toolbox.

You will work in the Cognitive Robotics Department at TU Delft under the supervision of Prof. Robert Babuška and Dr. Laura Ferranti. You will be embedded in the Reliable Robot Control Lab and contribute to TU Delft’s OPERA work on reliable, adaptive, and trustworthy robot autonomy.

Job requirements

The ideal candidate for this PhD position has a strong technical background and is enthusiastic about contributing to safe, intelligent, and trustworthy robot autonomy. We welcome applicants from all backgrounds who are motivated to work at the intersection of planning, learning, and perception.

You have:

• A MSc degree in Systems and Control, Computer Science, Applied Mathematics, Robotics, Mechanical Engineering, Artificial Intelligence, or a closely related field.
• A strong interest in working across multiple research domains, including motion planning, control, perception, and machine learning.
• Excellent programming skills, particularly in Python and/or C++, and experience with modern software development tools.
• A passion for ground breaking theoretical research combined with an eagerness to test ideas on real robotic systems.
• Strong analytical and mathematical abilities, enabling you to work confidently with algorithms, optimization, probability, or learning frameworks.
• Excellent communication skills and proficiency in English (written and verbal), as required for academic publication and international collaboration.

You are particularly encouraged to apply if you have experience in one or more of the following areas:

• MPC, optimal control, trajectory optimisation, motion planning.
• Safe/robust control and decision-making under uncertainty.
• Active perception, sensor fusion, uncertainty estimation, viewpoint planning.
• Reinforcement learning or safe learning as useful, but not the main identity.
• ROS/ROS2, simulation and real robotic validation.

As part of OPERA, you will be requested to travel to visit the different partners in the consortium and attend the regular project meetings.

TU Delft

TU Delft offers an inspiring research environment within the Cognitive Robotics Department and the Reliable Robot Control Lab, where you will contribute to reliable, adaptive, and trustworthy robot autonomy as part of the OPERA consortium.

Conditions of employment

Doctoral candidates will be offered a 4-year period of employment in principle, but in the form of 2 employment contracts. An initial 1.5 year contract with an official go/no go progress assessment within 15 months, followed by an additional contract for the remaining 2.5 years assuming everything goes well and performance requirements are met.

As a PhD candidate you will be enrolled in the TU Delft Graduate School. The TU Delft Graduate School provides an inspiring research environment with an excellent team of supervisors, academic staff and a mentor. The Doctoral Education Programme is aimed at developing your transferable, discipline-related and research skills.

TU Delft offers a customisable compensation package, discounts on health insurance, and a monthly work costs contribution. Flexible work schedules can be arranged.

Will you need to relocate to the Netherlands for this job? TU Delft is committed to make your move as smooth as possible. The HR unit, Coming to Delft Service, offers information on their website to help you prepare your relocation. In addition, Coming to Delft Service organises events to help you settle in the Netherlands, and expand your social network in Delft. A Dual Career Programme is available, to support your accompanying partner with their job search in the Netherlands.

As part of knowledge security, TU Delft conducts a risk assessment during the recruitment of personnel. We do this, among other things, to prevent the unwanted transfer of sensitive knowledge and technology. The assessment is based on information provided by the candidates themselves, such as their motivation letter and CV, and takes place at the final stages of the selection process. When the outcome of the assessment is negative, the candidate will be informed. The processing of personal data in the context of the risk assessment is carried out on the legal basis of the GDPR: performing a public task in the public interest.