Thesis Work: Physical AI Through Simulation-Driven Learning and Planning for Industrial Robots

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:

Your role and responsibilities

Within ABB Robotics R&D Motion Control department we are proposing several master’s thesis projects next spring. The Motion Control department is responsible for a wide range of areas within the robot controller development spanning modeling, identification, control design, optimization, and path planning.

Background

Developments in physical AI for robotics are accelerating rapidly, but many open questions remain. Two of the more important questions include how to better utilize simulations to gather data for machine learning and how to best combine modern machine learning methods with classical planning. Below these two broad themes are several sub-questions that could be treated in a master thesis project. The specific questions to be addressed will depend on the interest and capabilities of the student and are to be decided after discussion together with the supervisors.

Objective

The objective of this thesis is to explore how modern simulation and AI methods can be used to support data-driven learning and planning for industrial robot systems. The work will include a review of relevant state-of-the-art methods, selection of a suitable problem formulation, implementation of a prototype, and evaluation on a realistic robot benchmark or experimental setup. Depending on the final scope, the thesis may focus on efficient generation of accurate simulations, Real2Sim methods, sensor modeling for cameras and force sensors, the trade-off between simulation speed and robot-model accuracy, or the use of foundation models to improve task planning. The expected outcome is a technical evaluation of the selected approach, together with recommendations for how the method could be further developed or applied in future industrial robot applications.

Research Questions

Some examples of possible research questions include:

• How can simulation environments be generated or adapted efficiently enough to support data-driven learning for industrial robot applications?
• Which Real2Sim methods are most suitable for reducing the gap between real robot behavior and simulated robot behavior in motion-control-relevant tasks?
• How accurately do cameras, force sensors, and other relevant sensors need to be modeled in simulation to enable successful transfer to a real robot system?
• What is the best trade-off between simulation speed and robot-model accuracy when generating data for machine learning-based robot control or planning?
• How can foundation models be combined with classical task- and motion planning to improve planning efficiency, robustness, generalization, or ease of programming for industrial robot tasks in unstructured or dynamic environments?

Details:

• Period: 2027 January – June
• Number of credits:  30
• Number of students for this thesis work: One or two
• Location: ABB Robotics office at Finnslätten Västerås

Qualifications for the role

• Studying for a Master of Science in Robotics, Engineering Physics, Computer Science, Electrical Engineering, Mechanical Engineering, Applied Mathematics, or similar.
• Strong interest in control theory, planning, machine learning, system identification, and/or physics-based modeling.
• Knowledge of Python is required. Basic programming in C/C++ is considered an advantage.
• Experience with robotics, simulation environments, numerical optimization, or machine learning frameworks is considered an advantage.
• Ability to work independently, take initiative, and approach open-ended technical problems in a structured way.
• Good written and spoken English skills.

More about us

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

Recruiting Manager "Niklas Durinder" <Niklas.d.durinder@se.abb.com>, 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.