Thesis Work for generating photo realistic virtual training data for physical AI models
Teknik, data och digitalt · Data, AI och analys · Maskininlärning · Mjukvaruutveckling · Robotik och automation
I korthet
Ansvarsområden
- Research solutions for simulating robots and capturing video footage en masse.
- Create 3D digital twins of robot setups using techniques like Gaussian Splatting and mesh generation.
- Evaluate robot performance using physical AI methods (e.g., ACT) trained on virtually generated video reference.
- Explore the necessary level of simulation fidelity for training AI models, balancing speed, complexity, and performance.
Krav
- Engineering student with a background in robotics, computer science, or a similar program.
- Experience with machine learning and AI training, preferably in AI vision systems or imitation learning.
- Knowledge of linear algebra.
Önskade kvalifikationer
- Experience with 3D rendering and CAD and/or 3D modeling techniques is a plus.
- Hands-on experience with ABB industrial robots is a plus.
Förmåner
- Partly on-site work.
- Thesis period: January - June 2027.
- 30 ECTS/högskolepoäng (hp) credits.
- Opportunity to work with world-class experts in a fast-moving, innovation-driven environment.
- Potential for future career opportunities with ABB.
Färdigheter
Om ABB
We are a technology leader in electrification and automation, enabling a more sustainable and resource-efficient future.
ABB is a technology leader in electrification and automation, enabling a more sustainable and resource-efficient future. The company’s solutions connect engineering know-how and software to optimize how things are manufactured, moved, powered and operated. Building on more than 130 years of excellence, ABB’s ~105,000 employees are committed to driving innovations that accelerate industrial transformation. www.abb.com\nPrivacy notice & house rules: https://social.abb/social-media-information\n\nBefore applying for a job at ABB , please read our Fraud Warning:…
Företag
ABBPublicerade jobb
för 1 dag sedan
Anställningstyp
Praktik
Arbetsform
Hybrid
Erfarenhetsnivå
Student
Platser
Sweden
Kvalifikation
Sökande
Ansök tidigt