Systems Perspective on Human Spatial Behavior based on Invariants in Agent-Environment Interactions

Presented by Bérénice Mettler

Tuesday, November 29, 2016
12:00 p.m.
ICSI Lecture Hall

Abstract: My general research interests are understanding how human achieve versatile and adaptive spatial guidance and control skills. These skills manifest in manual control such as surgery, as well as, in vehicle control such as helicopter control. Advances in communication, computing and sensing open up a broad range of opportunities to augment human skills and, at the same time, they raise questions about the best integration of human operators in human-machine systems (HMS). Human factors is a field that studies human-machine interactions combining behavioral and biological sciences with engineering. Due to complexity of HMS the field has heavily relied on empirical methods, and formal results have been limited to specific domains, such as continuous control, discrete decision making, or planning. A general characteristic of human and animal spatial behavior is that they are dynamically coupled with their environment. My research, therefore, has focused on formal modeling of the system-wide organization of perceptual, control, guidance, and planning functions. I will describe the general approach use to tacked various aspects of human spatial control. This approach is based on the hypothesis that humans’ sensory-motor patterns are governed by invariants in their interactions with the task environment. These invariants are relevant to understanding the sensory and control functions, and, at the same time, provide insights into structural characteristics of spatial behavior, which support the understanding of planning functions. I will cover examples of this general approach to various aspects including: modeling and assessment of human motor skills; modeling the perceptual functions and attention when learning to operate in cluttered environments; and augmenting robot tele-operation in search tasks.

Bio: Bérénice Mettler received her diploma in mechanical engineering from ETH in 1996, and her doctorate from Carnegie Mellon University in 2001. She was a postdoctoral researcher at MIT’s Laboratory for Information and Decision Systems (LIDS) from 2001 to 2004 and then worked there as a research scientist until 2006. She then joined the faculty of the Department of Aerospace Engineering and Mechanics at the University of Minnesota, where she started the Interactive Guidance and Control Lab (IGCL). She is best known for her work on dynamics and control of miniature rotorcraft (Identification and Characteristics of Miniature Rotorcraft, Kluwer, 2003). Her current research focuses on understanding control, perceptual and cognitive mechanisms employed by humans to enable versatile and adaptive spatial guidance and control. Her work has been supported by the National Science Foundation, the Office of Naval Research, NASA, the Army Flightdynamic Directorate, and the Jet Propulsion Lab. She is a recipient of the National Science Foundation Faculty Early Career Development Award. She obtained tenure in 2013, has been on an entrepreneurial leave from 2014-16. She is the CEO and founder of iCueMotion with the goal of building technologies to augment human skills combining artificial intelligence and real-time feedback. She recently left her full-time faculty position to focus on running user trials.