NASA SISM Intelligent Systems Project Human-Centered Computing Human-System Modeling

Approach

The HCC Human-System Modeling (HSM) technical area focuses on computational modeling of distributed groups (scientists, engineers, astronauts, and mission operations personnel) interacting with advanced autonomous and robotic systems. Such models enable design decisions that reduce life cycle costs while enhancing system robustness and mission safety. This research focuses on work systems analysis and evaluation (i.e., understanding how people and systems are interconnected in practice) and on computational models of people interacting with automated systems. HCC methodology is based on the scientific study of cognition in people and machines, especially the differences between perceptual-motor, cognitive, and social aspects of people and present-day computers. It is a software engineering methodology, and should be integrated with other advanced software engineering methodologies. Human models fill a critical gap in our design capabilities. Life cycle costs of complex systems are largely determined by decisions made in early design phases, when ignorance is maximal. Model-based design methods can help envision and quantify real operational scenarios -- including interactions of cognitive, physical, and social factors. Designers can then incorporate cost-effective models of human behavior, procedural tasks, human-automation interaction, time-budgets for individual agents, and consistency-checks on synchronous and asynchronous (e.g., document-mediated) communication. This reduces both costs and risks. Without credible human behavior models, it is too easy to push aside complex performance and integration problems as "training issues." Modeling of total systems is the fundamental challenge to designing human-centered systems. Models useful for design must include people, objects, activities, and spatio-temporal constraints. System-level models capture the nature of joint activities, dynamic interactions, and plan modifications. Sophisticated models can answer key questions about facilities design, automation design and testing, science protocols, work scheduling, communication and coordination, and team training. An important goal of much HCC research is to design and model entire work environments -- such as mission control systems -- in which teams of humans and networks of machines are effectively integrated. The HSM technical area includes research in design-tool software, information flow modeling, cognitive task/work analysis, decision making under stress, psychophysiological reactivity, multi-person performance modeling, and work process modeling. Much of this research focuses on knowledge-based design tools and simulation frameworks for software agents. These tools will help uncover mission design flaws, deadlocks in decision-making, and bottlenecks in information flow.

Research Tasks

Human-System Modeling  Research Tasks
Principal Investigator	Lead Org.	Task Title
John R. Anderson	CMU	Simulating Learning of Complex, Dynamic Tasks
Jeffrey M. Bradshaw	UWF	Teamwork in Practice: Design for Collaboration in Mixed Human-Robotic Teams
Asaf S. Degani	ARC	Formal Conjoint Design and Analysis of Displays and Procedures
Michael A. Freed	ARC	Mixed-Initiative and Autonomous Surveillance
Nancy G. Leveson	MIT	Approaches to Human-Centered Software Development
Michael R. Lowry	ARC	Formal Analysis of Human-Automation Interaction
Maarten Sierhuis	ARC	Work Systems Simulation (Brahms)

NASA | SISM | IS | Human-Centered Computing

IS Tasks | HCC Tasks