PhD Position in Physical AI and Robotic Foundation Models
Explore vision-language-action models and real-world robotic applications in a fully funded PhD position at Aarhus University.
Best for:Best for PhD candidates with a strong background in machine learning, robotics, or computer vision, eager to bridge theory with hands-on robotic system development and industrial collaboration.
Overview
This fully funded PhD position at Aarhus University focuses on advancing **Physical AI**—specifically vision-language-action models, world models, and safe test-time adaptation of robotic foundation models. The work involves hands-on development and testing on real robotic systems, addressing challenges like distribution shifts, uncertainty, and real-world constraints. The position is based in the **A3 Lab**, led by Assoc. Prof. Behzad Bozorgtabar, and offers collaboration opportunities with industrial partners in robotics and autonomous systems. Applicants will join an international, English-speaking research environment with access to Denmark’s AI ecosystem, including the Pioneer Centre for AI (P1) and the DIGIT Centre. The role emphasizes real-world impact over benchmarks, ensuring research directly addresses industrial challenges.
What you'll do
- Develop and refine vision-language-action models for robotic systems, focusing on real-world applicability.
- Design and implement world models that account for uncertainty and distribution shifts in robotic environments.
- Conduct hands-on experiments with real robotic systems to test and validate AI models under real-world constraints.
- Collaborate with industrial partners to ensure research directly addresses practical challenges in robotics and autonomous systems.
- Publish findings in top-tier machine learning and computer vision venues (e.g., NeurIPS, ICML, CVPR).
- Engage in knowledge exchange within the ELLIS network and Denmark’s AI ecosystem (P1 and DIGIT Centre).
- Present research outcomes at conferences, workshops, and internal lab meetings.
Requirements
Degree required: Master's degree (or equivalent) in Computer Science, Robotics, Machine Learning, Artificial Intelligence, or a related quantitative field.
Degree offered: PhD
Language: English; proficiency at a level suitable for academic research and collaboration (no explicit test score required but implied).
Instruction: English
- Proven expertise in machine learning, particularly in vision-language models or reinforcement learning.
- Experience with robotic systems or hands-on robotics development.
- Strong programming skills in Python and familiarity with deep learning frameworks (e.g., PyTorch, TensorFlow).
- Background in computer vision, natural language processing, or autonomous systems.
- Ability to work in an interdisciplinary team with industrial collaborators.
- No explicit nationality or residency restrictions mentioned.
- Open to applicants with a strong academic background in relevant fields.
- No age or prior experience caps specified.
Funding & terms
How to apply
Deadline: 2026-11-01 (CET)
Applications must be submitted exclusively through the official Aarhus University PhD application portal. No email submissions will be accepted. The process involves uploading required documents directly via the provided link. Ensure all materials are complete and adhere to the portal’s guidelines.
- CV/resume
- Motivation letter or personal statement
- Certified copies of academic transcripts and diplomas
- Proof of English proficiency (if applicable, though not explicitly required)
- Reference letters (number not specified but typically 2-3)
Contact
Behzad Bozorgtabar· Associate Professor (Supervisor)
Create a free profile to unlock your personal match score, the full requirements, funding details and how to apply — in English or Persian.