Fully funded PhD in Physical AI for Adaptive Robot Manipulation (tactile sensing)
- Deadline
- 4 Jan 2027 · 43 days left
- Location
- Aarhus, Denmark
- Type
- PhD Position
- Compensation
- Danish state academic salary, basic salary grades 4–8, plus monthly non-pensionable allowance
- Duration
- 36 months · 3 years
- Start date
- 1 Mar 2027
- Working language
- English
- Posted
- 21 Nov 2026
- Verified
- 21 Nov 2026
About this opportunity
Aarhus University is offering a fully funded three-year PhD in Physical AI at the Adaptive & Agentic AI (A3) Lab in its Department of Electrical and Computer Engineering. The position belongs to STORM (Soft Tactile Memory for Safe Physical-AI Correction of Robotic Foundation Models), a project funded by the Independent Research Fund Denmark. The project asks how robot foundation models can use touch as well as vision to notice and fix failing actions, such as a slipping grasp, while the robot is still moving.
The research has three parts: learning contact representations from tactile sensors, including a soft-fluidic skin developed at the University of Southern Denmark (SDU); predicting how contact should change over short time horizons; and using the gap between expected and felt contact to adapt pre-trained policies safely. The methods will be tested on dexterous handling of fragile and unfamiliar objects and on safe human-robot interaction.
What is offered
- A fully funded three-year PhD position at the Graduate School of Technical Sciences, Aarhus University.
- Employment as academic staff at basic salary grades 4–8 under the Danish state agreement with the Confederation of Professional Unions, plus a monthly non-pensionable allowance.
- Supervision by Associate Professor Behzad Bozorgtabar, working with Professor Ahmad Rafsanjani and the SDU Soft Robotics group.
- A possible research stay at KTH Royal Institute of Technology in Stockholm.
- Preferred start date: 1 March 2027 or later.
Who can apply
- Applicants with a master's degree (120 ECTS or equivalent) in computer science, machine learning, AI, robotics, electrical or computer engineering, or a related quantitative field. The degree must be completed or nearly completed before the start date.
- Advanced Python and PyTorch skills, with experience training or adapting modern architectures such as Transformers, sequence models, and diffusion or flow-matching models.
- A strong background in machine learning, deep learning or computer vision, and interest in tactile sensing, world models and online adaptation.
- Willingness to run experiments on real robots. Hands-on robotics experience (grippers, tactile or force sensors, ROS 2) helps but is not strictly required for strong ML candidates.
- It is also an advantage to have experience in one or more of: robot policy learning (imitation learning, VLA or world-action models, benchmarks such as LIBERO, CALVIN or ManiSkill), world models, or model adaptation (test-time adaptation, continual learning, PEFT/LoRA).
- A strong master's thesis, open-source code, or publications at leading ML, vision or robotics venues count strongly in your favour.
- Excellent written and spoken English.
The university describes the position as highly selective and will shortlist applicants.
How to apply
- Apply through the online portal linked from the position page. The deadline is 4 January 2027 at 23:59 CET. Only documents received before the deadline will be assessed.
- Upload a one-page statement of interest.
- Upload a CV with a publication list and a technical project page, with links to code or videos where you have them.
- Upload transcripts and diplomas for your bachelor's and master's degrees.
- Upload a PDF copy of the project description from the announcement. The portal requires this for technical reasons.
The committee may ask for more information or invite you to an interview. See the application guide for details. Questions about the project can go to Behzad Bozorgtabar (behzad@ece.au.dk), and questions about admission to admission.gradschool.tech@au.dk.
Requirements
-
Open to all nationalities
-
PhD candidate
-
Minimum degree: Master's
Fully funded PhD in Physical AI for Adaptive Robot Manipulation (tactile sensing)
4 Jan 2027 · 43 days left