Interaction Architecture

The premier  objective of this work package is to development of a platform that allows different research groups to carry out their experiments and to develop their own applications within an ambient intelligent robotic apartment environment. Importantly, this includes the integration of the robot in the ambient environment infrastructure, i.e. to allow for smooth data exchange on the one hand as well as the possibility for joint control, supporting parallel interactions. Furthermore, the architecture should provide the capability for 24/7 operability.

The Cognitive Service Robotics Apartment as Research Environment for Interdisciplinary Research.

Research on ambient intelligence and human-robot interaction needs to be conducted in a strongly interdisciplinary process to facilitate seamless integration of assistance technologies into the users' daily lives. With the Cognitive Service Robotics Apartment (CSRA) (Wrede et al., 2017), we are developing a novel kind of laboratory following this interdisciplinary approach. It combines a smart home with ambient intelligence functionalities with a cognitive social robot with advanced manipulation capabilities to explore the all day use of cognitive interaction technology for human assistance. This lab in conjunction with our development approach opens up new lines of inquiry and allows to address new research questions in human-machine, -agent and -robot interaction.


Enabling Parallel Interactions

In contrast to commercially available smart homes which only have a limited amount of interaction sites, the CSRA supports a wide range of interfaces with varying complexity. Thereby, parallel manipulation of the same device might occur, e.g., if two persons interact with the apartment at the same time in different locations. In order to keep interactions consistent for every user, potentially conflicting actions inside the CSRA are resolved by a central component that manages communicational resources. The apartment’s functionality is thereby provided in an architecture that differentiates between so-called preset scenes (that configure the apartment), scenarios (that provide a structured interaction with a temporal elongation), and more simple tasks (that trigger a single routine). Central research questions in this area are:

  1. How can an appropriate architecture arrange semi-scripted courses of action with autonomously performed system behaviors in highly parallel interactions?
  2. How can such an architecture support conflict resolution from a user’s perspective?