Modality data reduction and their effect for Teleoperation systems

MSc assignment

Context

Teleoperation systems

Teleoperation systems often exist out of multiple different modalities, such as video and arm control,
and audio. The cyber-physical components of such a system require that they are working in a "Firm
Real-time" setting: They require an update from the controller every so often. Missing this deadline
means the system comes to a safety stop and needs human interaction to restart again. This is a
situation which we would like to avoid.
Keeping communication as fast as possible is quite a challenge, especially when the operator and
remote are separated by some distance, or experience a slow communications channel.

Modality data reduction and encoding

For each of the modalities of a teleoperation system, certain choices can be made to reduce the
data. A downside often seen is that this does however delay the sending of the data, a trade-off
which might be problematic for the FRT requirements described above. Furthermore, choices in
transmission protocol can also be made. However, this does reduce the certainty of transmission
which is a trade-off which can affect the stability and transparency of the system.

Goal

The goal of this assignment is to explore the usefulness of encoding and transmission techniques
for teleoperation setups with audio, video, and haptic information streams, and how to use these in
current tele-operation systems to improve performance.
Towards this goal you will be doing the following:

  • Design and implement a system which allows for experimentation regarding the impact these
    techniques.
  • Design and execute an experiment which evaluates the usefulness of these techniques for
    teleoperation systems.

Your profile

For this assignment, best-equipped students are those who:

  • Are student in MSc. Robotics, MSc. EmSys, or related field.
  • Have some programming skills, most notably C++ and ROS.