Collaborative Research: Research Network on Computational Human-Embodied AI Interaction Models for Engineered Systems

Advances in robotics, artificial intelligence, machine learning, and automation are enabling machines to perform tasks that once required continuous human control. These technologies are increasingly used in factories, autonomous vehicles, hospitals, and many other settings. However, machines still struggle to work smoothly side by side with people because human actions can vary with experience, surroundings, goals, emotions, and social interactions. Intelligent machines must therefore become better at recognizing what people are doing, adapting as circumstances change, and responding safely when unexpected situations arise. Achieving this capability requires new knowledge at the intersection of computing, engineering, human factors, and the cognitive and behavioral sciences. This award supports the Computational Human–Embodied Intelligence Interaction Models for Engineered Systems (CHIMES) Research Network, which will bring together multi-disciplinary researchers to solve this scientific gap. Through workshops, student training, research exchanges, and shared resources, CHIMES will help advance safer, more effective intelligent technologies that work with people in everyday and high-risk environments. The CHIMES Research Network will address the scientific challenge of developing computational human models that enable seamless interaction between humans and intelligent dynamical systems. The network will integrate expertise in control theory, robotics, machine learning, human factors, cognitive science, social psychology, behavioral economics, and human-computer interaction. Its activities will focus on three complementary areas: interdisciplinary synthesis of human-modeling approaches, models of bidirectional human-machine interaction and co-adaptation, and contextual reasoning that captures the contingent, evolving, and socially situated nature of human behavior. CHIMES will convene annual workshops and graduate student practicums, support interdisciplinary research exchanges, and establish a centralized repository of datasets, computational models, benchmarks, simulation platforms, and experimental practices. These activities are designed to advance creative cross-disciplinary approaches to computational human modeling, including predictive and adaptive models of human behavior and cognition, improve the safety, robustness, and responsiveness of autonomous systems, establish a sustainable interdisciplinary research community, development and dissemination of shared knowledge, and creation of a functional database for sharing of data, models, simulation platforms, and experimental practices that advance computational human-embodied AI modeling. CHIMES outcomes are expected to contribute to position the United States as a global leader in safe human-centered intelligent engineered systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.