Thesis: Deep learning-based environment recognition using 3D sensor technology*

Winter Semester 2025/26 – Limited to 3-6 months, Start possible from 01.09.2025

YOUR TASKS:

• Develop algorithms for environmental perception to prevent collisions of mobile robots and machines.
• Research and apply state-of-the-art deep learning methods for 2D/3D environment perception (segmentation, object detection, and tracking).
• Train deep learning models and evaluate various algorithms.
• Design a multi-sensor system for 3D environmental perception, e.g., for obstacle detection in logistics applications or outdoor environments.
• Process and fuse high-resolution RGB and depth data from real or simulated stereo or RGB-D cameras.
• Integrate, optimize, and evaluate systems on state-of-the-art AI accelerator hardware such as NVIDIA Jetson and Hailo.
• Document your work results in detail.

YOUR PROFILE:

• You are currently pursuing a Master’s degree in Computer Science, Robotics, Mathematics, or a related field.
• You have strong programming skills, ideally in C++ or Python.
• You have a solid understanding of image processing and working with 2D and 3D data.
• You have expertise in deep learning and machine learning, with experience in frameworks such as TensorFlow or PyTorch.
• Ideally, you have knowledge of applications involving mobile robots or autonomous driving, e.g., using ROS and/or NVIDIA Jetson.
• Your systematic and structured thinking and approach set you apart.
• Creativity in problem-solving and enthusiasm for innovation complete your profile.
   

YOUR APPLICATION:

• We are looking forward to your online application
• Sarah Disch
• Job-ID 36765 
• 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: Abschlussarbeit