Human Research and Engineering Directorate Research Areas

Cognitive Modeling in Support of a Robotic Control System

Advisor: Cassenti, Daniel N. (daniel.cassenti@us.army.mil)   410.278.5859

Location: Human Research and Engineering Directorate, Aberdeen Proving Ground, MD 21005-5425

Key words: cognitive modeling, robotics, cognition, ACT-R, cognitive psychology

The use of robotic assets in the field instead of Soldiers is one approach to protecting Soldiers from harm. The approach requires the development of intelligent robots that also have a high degree of autonomy. To accomplish this, ARL has developed a robotic control system that has increased the capabilities of robots to act without requiring constant human operator supervision. The system is based on the cognitive modeling system ACT-R (Adaptive Control of Thought–Rational), which itself is based on a plethora of empirical research investigating cognitive psychology. Associates will ideally have a strong background in cognitive psychology and particularly cognitive modeling within cognitive modeling systems (e.g., ACT-R Soar, EPIC). A strong background in robotics is not required, though learning the robotic control system will be required once the Associate begins the program. Research will be conducted to identify important cognitive skills that humans use to solve spatial navigation problems and how these skills may be incorporated into the robotic control system. Common tools used in cognitive psychology, such as experimentation, task analysis, and human factors analysis may be incorporated into the proposed research project. Use of ARL’s facilities such as the Robotic Test Bed is strongly encouraged. The study of autonomous robotic systems is a field that represents tremendous growth and opportunity to be published in the psychology, cognitive, cognitive modeling, and computer science literatures.

References:
Kelley TD: International Journal of Advanced Robotic Systems 3: 219, 2006