PhD Position in Physical AI: Adaptive Foundation Models for Robotics

This PhD position at Aarhus University’s Department of Electrical and Computer Engineering focuses on developing adaptive foundation models for robotics, bridging multimodal perception, reasoning, and action. The successful candidate will join the Adaptive & Agentic AI (A3) Lab, supervised by Associate Professor Behzad Bozorgtabar and Professor Qi Zhang. The project aims to create models that enable robots to understand instructions, predict physical-world interactions, and adapt reliably under dynamic conditions. Research will emphasize novel learning algorithms, rigorous evaluation, and contributions to top-tier conferences in machine learning, computer vision, and robotics. The candidate will conduct original research targeting high-impact publications while collaborating with robotic and edge-computing platforms.

• Develop and evaluate adaptive foundation models for robotics, focusing on multimodal perception, reasoning, and action.
• Investigate vision-language-action models to enable robots to generalize across tasks, objects, and environments.
• Design world models that predict physical-world interactions and support planning and learning from interaction.
• Explore methods for maintaining reliable robotic behavior under dynamic environments and resource constraints.
• Conduct rigorous evaluations using public datasets, simulations, and robotic platforms to assess task success, generalization, and computational efficiency.
• Publish original research in top-tier conferences and journals in machine learning, computer vision, and robotics.
• Collaborate with supervisors and lab members to refine research directions and contribute to the broader field of adaptive AI.