Depth-Aware Physical AI Models for Robot Manipulation Thesis

About the role

Join ABB's Robotics business to investigate how depth information can improve Vision-Language-Action models for robot manipulation. This thesis runs from January to July 2027 on-site in Västerås.

Key responsibilities

• Survey existing approaches for integrating depth/3D information into VLA or other physical AI robot manipulation policies.
• Design and implement methods to incorporate depth (e.g., additional input modality, 3D representations, point clouds).
• Evaluate performance on real robot manipulation tasks in ABB's lab environment.
• Document findings and contribute to ABB's Physical AI pipeline.

Required profile

• Master's student in Computer Science, Electrical Engineering, Robotics, Mechatronics or related field.
• Strong interest in robot learning, machine learning, computer vision or physical AI.
• Ability to work independently and communicate findings clearly.

Required skills

• Programming experience in Python.
• Familiarity with deep learning frameworks such as PyTorch and HuggingFace.
• Experience with git and Docker (optional).
• Experience with or interest in 3D data (depth images, point clouds) (optional).