Problem formulation

In teleoperation systems for dexterous manipulation, ensuring safety and reliability is paramount, particularly in high-stakes environments. Traditional teleoperation methods often lack robust mechanisms for detecting anomalies and predicting potential failures, leading to risks of unintended actions and system breakdowns. Developing a system that integrates real-time anomaly detection with proactive failure prediction could significantly enhance the safety and reliability of teleoperation for complex manipulation tasks.

Task definition

This thesis will focus on developing a safe and reliable teleoperation framework for dexterous manipulation, incorporating advanced anomaly detection and proactive failure prediction mechanisms. The research will involve designing algorithms that monitor system performance in real-time, detect anomalies, and predict potential failures before they occur, thereby enabling preemptive corrective actions. The system's effectiveness will be tested through a series of teleoperation tasks, evaluating its ability to maintain safety, minimize errors, and enhance overall reliability in complex and dynamic manipulation scenarios.

You shall offer

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

• Coding skills in Python and C++.

• Experience with ROS and communication technology.

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.