Problem formulation

Teleoperation systems are increasingly vital in various fields, such as robotics, healthcare, and remote operations. However, the effectiveness of different interfaces, especially for dexterous robot hands, remains largely unexplored. This research, which aims to systematically compare the performance, usability, and application potential of a diverse range of interfaces—from gloves and motion capture systems to VR setups and RGB cameras—is at the forefront of innovation in this field.

Task definition

This thesis will involve designing and conducting experiments to evaluate and compare the performance metrics of various teleoperation interfaces for dexterous multi-finger robotic hands, including glove-based systems, motion capture setups, VR environments, ORBBEC pose tracking, and RGB cameras. The goal is to assess these interfaces based on accuracy, latency, user comfort, and ease of integration, providing actionable insights for selecting the optimal interface for applications with dexterous robotic hands.

You shall offer

• Solid knowledge base and experience in computer vision, deep learning, and robotics.

• Coding skills in Python and C++.

• Experience with ROS beneficial.

We will offer

• The most state-of-the-art technologies in deep learning and computer vision.

• Working in a lab with a Germany-wide unique Shadow Teleoperation System.

• Tight support from supervisors, including a workshop on scientific writing.