Eric Tüschenbönner

AR devices require accurate localization to enable 3D scene understanding – leverage existing 3D building models (CAD/LiDAR) as environmental maps instead of time-consuming scans. • Implemented synthetic data generation pipeline • Adapted Scene Coordinate Regression (deep neural network) for training on generated 3D coordinates • Explored transfer learning techniques to bridge the domain gap between real-world and synthetic data

• Developed AI-based visual quality inspection system concept for industrial automation. • Managed mobile banking app development; coordinated international teams and led client communication. • Gained expertise in agile project management, cross-functional collaboration, and technical communication.

Modular robotics platform for flexible automation, focusing on AI-driven visual inspections programmable via human-robot interaction. • Designed and integrated custom hardware components using CAD & 3D printing, enabling rapid iteration & scalable production • Enhanced software architecture & documentation; refactored codebase & initiated CI/CD workflow • Demonstrated interaction capabilities, improving operational efficiency & usability • Recognized for reliability, independence, & high-quality work

Focus on 3D Computer Vision. Thesis @ Magic Leap / CVG: Visual Localization against Universal 3D Models. Semester Project @ ASL: Extrinsic Camera Calibration using Map Information. Courses: - Robot Dynamics - Planning and Decision Making for Autonomous Robots - Autonomous Mobile Robots - Recursive Estimation - System Identification - Dynamic Programming and Optimal Control - Vision Algorithms for Mobile Robotics - Image Analysis and Computer Vision - Machine Learning - Mixed Reality

Bachelor Thesis: Design of a high-speed racing drone integrating different subsystems, that resulted in a novel wing-based design with thrust-vectored propellers and elevons. Within the group project, I was responsible for the control system design and guiding meetings/presentations as chairman of the team of 10.