[DC] Detection of an Isolated User’s (Inter)Actions during an Augmented Reality Guided Procedure: an Application in Space Medicine

With recent development of Augmented Reality (AR) and Artificial Intelligence (AI) technology, many new options have become available to overcome previously encountered problems (accurately detecting human activity, adaptive AR guidance, etc.). So when a collaborative scene between a "distant" user in an isolated location in need of assistance from a "local" user in a control centre is put into question, many different solutions can be considered. A solution could be to provide AR guidance for the isolated user with adjustable instructions from the local user who is immersed in a virtual representation of the isolated user’s environment. This allows the local user to quickly understand the problem and current state of the procedure, enabling them to provide the following steps proactively or to rearrange the steps in the current procedure. During my thesis work, our aim is to study how to detect the interactions of an isolated user, in order to create a more detailed virtual environment, improving the quality and priority of the instructed actions provided by the local user. This research subject has become more feasible with recent developments in the fields of AR technology, Human Activity Recognition (HAR) for lightweight or non-intrusive equipment as well as Application Domain Models to represent the environment using semantic information.