Thesis Work -Depth-Aware Physical AI Models for Robot Manipulation

At ABB, we help industries outrun - leaner and cleaner. Here, progress is an expectation - for you, your team, and the world. As a global market leader, we’ll give you what you need to make it happen. It won’t always be easy, growing takes grit. But at ABB, you’ll never run alone. Run what runs the world.

This role sits within ABB's Robotics business, a leading global robotics company. We're entering an exciting new chapter as we’ve announced the plan for SoftBank Group to acquire ABB Robotics. SoftBank is a globally recognized technology group and investor/operator focused on AI, robotics, and next-generation computing.  By joining us now, you’ll be part of a pioneering team shaping the future of robotics—working alongside world-class experts in a fast-moving, innovation-driven environment.

This Position reports to:

R&D Center Lead

 

Your role and responsibilities

Physical AI models such as Vision-Language-Action (VLA) models have recently emerged as a promising approach to robot automation. Unlike traditional robot programming, VLA models learn robot manipulation tasks directly from human demonstrations. They take camera images and language instructions as input, and output robot actions directly. Currently, most VLA models rely solely on standard RGB (color) images, which lack explicit spatial and geometric information about the environment.

 

This thesis investigates how to incorporate depth information (e.g., from RGB-D cameras) into VLA or other physical AI models to improve robot manipulation performance.

 

The student is expected to:

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

The student is encouraged to propose their own ideas on how depth information can best be represented and utilized within the model.

 

Details:

• Period: January to July, 2027
• Number of credits:  30 ECTS/högskolepoäng (hp)
• Number of students for this thesis work: 1
• Location: on-site, Västerås

Qualifications for the role

• Master's student in Computer Science, Electrical Engineering, Robotics, Mechatronics, or other related fields
• Strong interest in robot learning, machine learning, computer vision, or physical AI
• Programming experience in Python and familiarity with deep learning frameworks (e.g., PyTorch, huggingface, LeRobot platform).
• Experience with git, docker, etc is a plus.
• Experience with or interest in 3D data (depth images, point clouds) is a plus
• Hands-on experience with real robot systems is a plus but not required
• Ability to work independently and communicate findings clearly

More about us

Recruiting Manager LiWei Qi, +46 73 021 2309, Supervisor: Chi Zhang, [email protected] , Marco Iannotta [email protected] will answer your questions.

 

Positions are filled continuously. Please apply with your CV, academic transcripts, and a cover letter in English.

 

We look forward to receiving your application!

A Future Opportunity
Please note that this position is part of our talent pipeline and not an active job opening at this time. By applying, you express your interest in future career opportunities with ABB.

We value people from different backgrounds. Could this be your story? Apply today or visit www.abb.com to learn more about us and see the impact of our work across the globe.