Research

The guiding vision of the interdisciplinary Thematic Network on Interactive Intelligent Systems is to:

  • Elucidate the principles and mechanisms that underlie cognitive processes in humans and animals.
  • Replicate these principles in technical systems so as to make them more versatile and easier to use for ordinary people.
  • Create a scientific basis for cognitive interaction technology as a key mission for shaping the technology of tomorrow in a way that suits our human needs and helps us to address the societal challenges of the future

The goal of this research is to endow machines with the intelligence required to recognize how they can assist their users. Instead of requiring a great deal of programming, intelligent systems will be able to anticipate the user's wants and needs in a given context. In the future, such systems should be able to learn to acquire new knowledge and skills while they are in operation.
Advancing ways to complement and share expertise is closely connected to coordinating research efforts. While we do consider the frequent visits between network partners to already contribute significantly to towards exchange of expertise, we plan to go far beyond such “baseline-gains.” Instead, we seek to implement programs that will take this exchange of expertise to a deeper level. The type of approaches we envision include the following:

  • Initial, short-term research stays to enable researchers to acquire the necessary information in order to plan studies and set up experiments. This is documented by tutorial videos, initial test datasets, and conducted pre-studies.
  • As a next step, experiments are planned in detail off-site using initial test datasets and simulation tools.
  • Finally, experimental data can be recorded on-site (through short-term research stays) or remotely, the latter of which is documented through an appropriate meta-description-format stored in web-based catalogue.
  • A shared software toolkit (“Virtual Lab Toolkit”) to support remote experimentation.

Systems Sharing:

There is a steady increase in the technological requirements of technical systems to be used in sophisticated experimental setups. Examples of this include robotic devices that are expensive to develop and maintain, the need for dedicated microelectronics components that may be commercially unavailable, or neuroimaging and motion tracking technologies whose operation requires a large critical mass of researchers in order to make it viable. The partners of the Thematic Network will cooperate towards developing interfaces, protocols and processes to make such devices available among network partners. This will include offering access to selected equipment during research visits, but also the development of methods that enable remote access and remote experimentation, which in turn should help to achieve a new level of flexibility in systems sharing.

Lab Rotation Program:

We will implement a lab rotation program designed to facilitate the spread of cutting-edge experimental methods between the labs of partner institutions. This will be achieved through three complementary activities: (1) short-term research stays, particularly for doctoral and post-doctoral researchers; (2) data management activities that provide a joint standard and web-based platform for the documentation of the experimental setup, study design, and data; and (3) establishing an infrastructure for remote experimentation, which will enable a virtual network of distributed lab environments.

We expect these measures to have a strong impact on the integration of system development and thorough evaluation, thus contributing to wider testing of computational models on a large number of diverse, complementary robot platforms, ranging from virtual agents to android platforms. Strategic partners each offer distinct research expertise, complementing methods, and technological approaches, all of which will be integrated to enable synergistic effects in developing social artifacts and to allow for cross-national evaluations.