Thesis: Optimization of deep learning based perception in the context of industrial robotics*

Fixed-term for 3-6 months

SICK develops innovative solutions in the field of industrial automation to meet the growing demands of customers. The central research and development department supports this by exploring new technological and algorithmic approaches, particularly for the automation of warehouse processes. Robotics and camera-based sensors are used to improve efficiency, while AI-driven systems analyze inventory, optimize routes, and prioritize order processing. These systems often rely on classical neural networks, which face challenges when dealing with unknown objects or significant environmental changes.

Foundation models offer greater flexibility and scalability as they are pre-trained on large, diverse datasets and can be applied to various tasks without needing retraining for each one.

As part of a bachelor’s or master’s thesis, the existing bin picking application based on foundation models will be improved. Possible topics include the fine-tuning of foundation models for picking tasks, deployment on edge devices, integration of multimodal data for enhanced object recognition, automatic clustering of data as a basis for fine-tuning, and the comparison of agnostic object detection models on logistics data.

YOUR TASKS:

• Research literature on foundation models, bin picking applications, and automation in logistics
• Familiarize yourself with existing technologies and algorithms
• Analyze and discuss your results and document them thoroughly
  

YOUR PROFILE:

• Currently studying electrical engineering, computer science, mechatronics, mechanical engineering, or a similar field
• Strong object-oriented programming skills in Python
• Experience with training and using deep learning models with common frameworks
• Proficiency with Linux-based systems and Git
• (Very) good knowledge of English
• You are distinguished by your independent and structured way of working
• Strong teamwork and communication skills complete your profile

YOUR APPLICATION:

• We are looking forward to your online application
• Sarah Disch
• Job-ID 36391 
• All applications will be treated confidentially

*At SICK, we see people, not gender. 

We put great emphasis on diversity, reject discrimination and do not think in categories such as gender, ethnicity, religion, disability, age or sexual identity.

Stichworte: Intern, Internship, Abschlussarbeit