Physical AI Meets Virtual Reality: Immersive Human-Robot Interaction and Spatial Planning [Guided Research, Master Thesis]

23.10.2026, Abschlussarbeiten, Bachelor- und Masterarbeiten

Background and Motivation
The convergence of Physical AI, computer vision, and virtual reality (VR) opens new possibilities for how humans interact with intelligent robotic systems. While recent advances in embodied AI enable robots to perceive, reason about, and act within the physical world, interacting with these systems remains challenging. Traditional robot programming interfaces often require specialized technical knowledge and provide limited support for intuitive spatial reasoning, task specification, and motion planning.

Virtual reality offers a compelling alternative: humans can enter an immersive, three-dimensional representation of their physical environment, interact directly with reconstructed objects, and communicate desired robot actions through natural spatial gestures. Combined with state-of-the-art 3D reconstruction, RGB-D perception, and robot motion planning, VR can become an intuitive interface for programming and collaborating with physical AI systems.

This thesis explores the intersection of Human–Computer Interaction (HCI), Human–Robot Interaction (HRI), computer vision, and immersive technologies. The central idea is to develop a VR-based interaction framework that connects a digital representation of the physical environment with a real robotic system, enabling users to specify tasks, manipulate virtual objects, and plan robot motions in an immersive environment.

The project will use Franka robotic arms, RGB-D cameras, and modern 3D reconstruction techniques to build and evaluate a working prototype. The ultimate goal is to investigate how immersive interaction can make intelligent robotic systems more accessible, intuitive, and effective.

The central research question is:

How can VR-based human–robot interaction enable active camera viewpoint planning to improve 3D reconstruction of physical environments?

The ideal candidate has strong programming skills and interdisciplinary expertise in computer vision, robotics, and VR development, with experience in RGB-D perception, 3D reconstruction, robot motion planning, or immersive interaction particularly valuable.



Interested?
drop me an email: hannah.schieber@tum.de

Kontakt: hannah.schieber@tum.de