Postdoc in Robot Learning

• Pioneering work with many real robots (stationary single and dual-arm setups, as well as mobile manipulators) and shape the future of robot learning in the context of infrastructure robotics and beyond

• POSTED: September 9, 2025
• CLOSES: Open until filled
• LOCATION: Brisbane, Australia
• POSITION: Full-time, fixed-term (24 months)
• ORGANISATION: Queensland University of Technology
• SUPERVISOR: Prof. Niko Suenderhauf

About the Position

Prof. Niko Suenderhauf is seeking an enthusiastic Postdoctoral Research Fellow to make a decisive contribution to our research in robot learning at QUT’s Centre for Robotics (QCR). This is a unique opportunity to join one of Australia’s leading robotics groups and play a key role in a nationally significant research collaboration.

As part of the ARIAM Research Hub, you’ll work closely with me, our vibrant QCR team of academics, engineers, and PhD students, and expert colleagues from USyd (control) and ANU (planning). Together, we are pushing the boundaries of learning from demonstration, aiming to tackle key challenges in sample efficiency and generalisation.

This position is intended to be filled soon as possible (advertised in September 2025). Please contact us if you are interested.

Position Details

• Type: Full-time, fixed-term (24 months)
• Level: Academic Level B
• Salary: Starting from $114,000 AUD with annual increments, + 17% superannuation
• Leave: 20 days annual leave + paid personal/sick leave
• Location: QUT Gardens Point campus, Brisbane, Australia

The Research Vision

This project is about rethinking how robots learn from humans. We’re especially interested in:

• Learning from very few demonstrations
• Generalising to new tasks, objects, and environments
• Combining imitation learning with insights from control, planning, scene understanding, and LLM-driven reasoning

You’ll have the chance to do pioneering work with many real robots (stationary single and dual-arm setups, as well as mobile manipulators) and shape the future of robot learning in the context of infrastructure robotics and beyond.

You’ll also gain insights into research group management, including milestone tracking, budget planning, and broader research leadership – ideal preparation for an independent research career.

For a sample of our work in this area, check out our recent papers: You’ll continue our recent work in this area, such as:

• IMLE Policy: Fast and Sample Efficient Visuomotor Policy Learning via Implicit Maximum Likelihood Estimation – RSS 2025
• Learning from 10 Demos: Generalisable and Sample-Efficient Policy Learning with Oriented Affordance Frames – CoRL 2025
• Physically Embodied Gaussian Splatting: A Realtime Correctable World Model for Robotics – CoRL 2024 Oral Presentation
• Sayplan: Grounding Large Language Models Using 3d Scene Graphs for Scalable Robot Task Planning – CoRL 2023 Oral Presentation

Who Should Apply?

If you’re driven by curiosity and creativity, and excited about building the next generation of robot learning systems, this is for you. You should have:

• A strong background in robot learning – ideally with publications at CoRL, RSS, or similar venues
• Excellent communication skills, both written and verbal
• A clear vision for your research over the next two years
• A collaborative mindset and the ability to work effectively across diverse teams at QUT, ANU, and Uni Sydney
• Solid understanding of the mathematical and statistical foundations of deep learning
• A passion for real-world robotics and a desire to stay at the cutting edge of a fast-moving field

Key Responsibilities

• Lead and contribute to cutting-edge research in robot learning at QCR and the ARIAM Hub
• Publish at top-tier venues in robotics and machine learning
• Lead or co-organise workshops at major international conferences
• Collaborate with QCR’s researchers, engineers, and students to build reusable robot learning systems
• Translate research into practical demonstrations (e.g. asset management, infrastructure)
• Actively participate in the QCR and ARIAM communities, including events in Brisbane, Sydney, and Canberra
• Co-supervise PhD and undergraduate students
• Contribute to QCR’s Visual Learning and Understanding research program
• Support a collaborative and inclusive research culture in Australia through visits and joint publications

Selection Criteria

• Completion of a PhD in robotics or a related field
• A strong and clear research agenda to advance robot learning
• Demonstrated ability to conduct independent, high-quality research
• Track record of publication of robot learning work in top venues in robotics (especially CoRL and RSS)
• Deep expertise in robotic learning, general robotics, machine learning, and computer vision
• Proficiency in Python, PyTorch, and general software development on Linux
• Experience working with robot hardware: arms, mobile bases, cameras, other sensors
• Excellent written and verbal communication skills

How to Apply

To apply, please email the following materials to 📧 niko.suenderhauf@qut.edu.au

• A short video introducing yourself, your previous work, and your research vision for the next two years
• Your CV, including a full list of publications
• A motivation letter and research plan (max 3 pages)
• A statement on how you meet the selection criteria above (max 2 pages)
• A statement on how you plan to address the key responsibilities of the role (max 2 pages)
• Contact details for 1–2 referees, or attach reference letters