Real-World Cognitive Multi-Taskings and Problem Solving:

A Large-Scale Cognitive Architecture Simulation Through High Performance Computing - Project CASIE

Sponsor: Air Force Research Laboratory

Modern problem solving tasks range from the mundane everyday elements of life such as deciding what to eat for dinner and what to buy while driving home from work to highly specialized activities such as planning and executing a large-scale air operations campaign in the desert.  From a computational standpoint, both of these problems are extremely difficult to model, let alone have ready algorithmic solutions that are effective or even efficient.  While both of these problems have a myriad of challenges arising from vagueness, uncertainty, domain scope, etc., they share a fundamental core. There are numerous tasks and subtasks at varying levels of complexity that are either working in cooperation, or, just as likely, in competition with other individual tasks, groups of tasks, or even organizations of tasks.  

This two-year project addresses the preliminary investigation on the fundamental problem of modeling and solving these "communities" of tasks from a cognitive point of view different from existing approaches such as multi-agent systems, parallel and distributed task decomposition, neuro-inspired cognition systems, and cellular automata.  For example, using purely autonomous agents incurs a massive burden of coordination and communication due to their distributed philosophy.  All decisions are made locally and coordination and communication must be negotiated between the agents. 

This is a joint collaboration project between SUNY Binghamton and Dartmouth College.

Project Personnel: 

PIs:

PhD student:

AFRL Program Managers:

Collaborating Organizations:

Publications:

 

This material is based upon the work supported by the Air Force Research Lab under project No. FA8750-05-2-0284.

Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Air Force Research Lab.

Go back to the Multimedia Computing Research Lab homepage