Project Description
Teleoperation systems are becoming increasingly important in fields such as remote e-health, remote maintenance, disaster response, and industrial robotics. These systems allow users to perform tasks remotely through robotic setups, often aided by technologies like motion tracking suits, VR headsets, and robotic manipulators. However, one of the largest challenges in teleoperation is ensuring smooth and seamless control, particularly when system latency affects how the user perceives their connection to the remote environment. High latency may disrupt the user’s sense of agency, embodiment, or confidence in completing tasks.
Our ongoing research explores the interplay of (network) latency, motion prediction, and user experience in teleoperation systems. We have built a teleoperation system that integrates a motion tracking suit, stereo camera, a VR headset, and a robotic hand and arm system. We aim to study how users perceive and adapt to latency during basic teleoperation tasks such as pick-and-place operations as well as tasks related to activities of daily living (ADL). By introducing cutting-edge technologies like motion prediction and EEG-based intention detection, we aim to push the boundaries of what is possible in teleoperation systems.
The following research directions are suggested to help narrow down your focus, but we are very open to adjusting the scope based on your interests and background.
Possible Research Directions
1. The Impact of Latency Compensation on User Experience
Latency compensation through predictive algorithms is a commonly used technique to improve teleoperation performance. This project investigates how such compensation affects user experience, specifically regarding task performance, embodiment, presence, and sense of control. A comparative study will be conducted between systems with and without predictive components, possibly researching various prediction horizons. The results can provide valuable data on the subjective and objective trade-offs of using prediction algorithms.
2. Integrating EEG for Enhanced Motion Prediction in Teleoperation
Predicting a user’s actions based solely on motion data might not always be sufficient, especially when dealing with high latency or unexpected movements. This project will explore the addition of EEG signals from a brain-computer interface (BCI) setup to improve motion prediction in teleoperation tasks. You will analyse whether EEG can provide reliable intention signals to reduce the impact of latency, evaluating both technical feasibility (e.g., prediction accuracy) and the user experience.
3. Task Complexity and Its Effect on Latency Perception and User Experience
The difficulty and complexity of tasks can affect how users perceive and adapt to latency. This project investigates how task complexity impacts the user’s sense of agency, presence, and latency awareness. By designing experiments with tasks of variable difficulty, from simple grasping tasks to more advanced tasks used in daily life, you will identify the latency tolerance thresholds for different types of operations. An additional focus could be placed on how these thresholds might guide the design of teleoperation systems for specific use cases.
Environment
The research for this thesis is conducted in the SINLAB at IFI. We have regular group meetings where we share our progress and discuss emerging issues. When working with us, you will get access to our lab, where students and researchers collaborate. Our robot uses the UR10e arm with a 5-finger Shadow Hand. Our controls are the Rokoko SmartSuit for body and hand motion. Additionally, the Vive Tracker can be used for tracking absolute positions.
Learning Outcomes and Expectations
By the end of the project, students will:
- Gain hands-on experience with advanced teleoperation systems, including VR/AR technologies, robotic control, and motion tracking.
- Gain knowledge in human-computer interaction (HCI), including the study of user experience concepts such as embodiment, agency, and presence.
- Learn to design, conduct, and analyse experiments involving human participants.
- Gain knowledge on ongoing research activities by actively participating in regular SINLAB meetings.
- Optional: Have the opportunity to co-author and publish a research paper based on their results.
Requirements for the Student
We are looking for motivated master students with an interest in robotics, human-computer interaction, neuroscience, and/or virtual reality. Background knowledge in one or more of the following areas will be helpful but is not strictly required:
- Programming skills (e.g., Python, MATLAB, or C++).
- Experience or coursework in robotics, control systems, and data analysis (e.g. IN5060)
- Interest in or familiarity with user experience measurement techniques and experimental design.
- Optional: Basic knowledge of EEG or familiarity with BCI concepts.