Applications of Spatial Cognition for Intelligent Assistance Systems

Motivation and Aim

Artificial Intelligence is one of the main contributors to the interdisciplinary field of spatial cognition. In the last decade or so, large progress has been made in the AI related research fields in spatial cognition (e.g. in cognitive computational modeling, in autonomous robots, in spatial reasoning and planning, in geographic information and navigation systems, and in language processing). This workshop is dedicated to possible (fields of) applications that may result from the achievements in basic spatial cognition research in AI.

The aim of the workshop is to bring together researchers working on approaches that are or may be employed for intelligent spatial assistance and that take into account and/or integrate results of cognitive science. We are particularly interested in applications, i.e. computational systems already being developed, or on current approaches that may end up in computational spatial assistance systems.

Topics and Questions

Topics and questions of interest include (but are not limited to):

Wayfinding / navigation assistance: intelligent route planners or navigation systems capturing a user's previous knowledge, aspects of personalization, integration of environmental information (e.g., landmarks), path selection based on cognitive considerations, etc.
Service robotics / (semi-)autonomous machines as assistance systems: support for the elderly, aspects of human-robot interaction, safety and security issues, etc.
Applications in design and architecture: cognitive approaches to computer-aided architectural design, (semi-)automatic spatial design, systems supporting creativity and innovation, etc.
Spatial planning and scheduling: decision support tools for spatial planning, spatial reasoning (e.g., in under-/over-constrained systems), representation and communication of solution spaces, etc.

Typical questions to be discussed may be:

What are the options for intelligent computational support of humans in spatial tasks?
How can cognitive principles and mechanisms in AI systems be conceived of as a fruitful approach for representation of, reasoning about, and communication of spatial information?
How can these mechanisms be implemented efficiently, both with respect to runtime and performance?
When do we need to model cognitive processes (strong cognitive adequacy), and when is a cognitively plausible output sufficient (weak cognitive adequacy)?
How do findings from spatial cognition may enhance approaches from (classical) AI?

Organization

Unfortunately, we did not receive a reasonable number of suitable contributions to compile a sensible workshop program. Therefore, we decided to cancel the workshop. We would like to apologize for any inconvience that our decision may cause.

Workshop Organizers

Thomas Barkowsky, Universität Bremen
Kai-Florian Richter, Universität Bremen
Michael Schellenbach, MPIB Berlin