Seminar: Dr. Ted Pavlic, Arizona State University
All dates for this event occur in the past.
Details:
Sept. 13, 2018 at 3 p.m.
Dreese Lab, Rm. 259
"Thinking Outside the 'Bot: What Ecology, Physiology, and Conservation Biology Can Offer to Engineering Design"
ABSTRACT: Natural systems have long provided motivation as qualitative reference models for the design of autonomous systems. However, due to very little interaction between organismal biologists with engineers and computer scientists, biology has had very little effect on research and education in autonomous decision-making systems. Often approaches use high-performance computing or sophisticated mathematics from control theory to build algorithms that approximate superficial features of natural systems. Novel algorithms often have to be tested empirically to understand their improvement over the state of the art. Those ostensibly nature-inspired algorithms that do show impressive performance seem equally likely to have been developed with biology only entering in at late stages to provide analogies to improve dissemination. Simply put, while principles from chemistry and physics have permeated engineering, biology is a science that has been kept at arm's length. Recent advances in computation and manufacturing are leading to autonomous systems whose complexity begins to approach that of biological systems. In this talk, I discuss how my interdisciplinary lab seeks to bridge the ever-shrinking gap between biology and autonomous systems design. I will discuss how principles from behavioral ecology, such as cue ritualization, can inspire fundamentally new approaches int he design of multi-agent systems. I will show how apparently disparate research areas, such as nutrient ecology and parallel numerical computation, are actually closely related and can complement each other, leading to new decentralized resource-allocation algorithms as well as new approaches to thinking about threats to conservation of biodiversity. Time permitting, I will also discuss new work that merges sports science, conservation ecology, and operations research to better understand the interaction of speed and agility in heterogeneous multi-agent groups. Ultimately, I hope to show that maintaining a barrier between biology and engineering will only slow progress in the quest to make flexible, adaptive, life-like engineering systems.
BIO:Dr. Pavlic (@TedPavlic) has an interdisciplinary background, starting with a PhD in electrical and computer engineering from The Ohio State University, progressing through research appointments in computer science and then behavioral ecology, and leading to his current appointment as an Assistant Professor held jointly with the School of Sustainability and the School of Computing, Informatics, and Decision Systems Engineering at Arizona State University (ASU). He also serves as adjunct faculty in ASU's School of Life Sciences. His research uses a combination of theoretical, computational, and empirical methods to study both natural and engineered autonomous decision-making systems. His laboratory does empirical work with natural systems, such as social-insect colonies, and does engineering work building decision-making algorithms for artificial systems, such as decentralized energy management systems for the built environment. Dr. Pavlic is the Associate Director of Research for The Biomimicry Center at ASU. He is also active in several professional organizations and publication venues across engineering and life sciences. He is an ASU-SFI Fellow and affiliated with the ASU BEYOND Center for Fundamental Concepts in Science, the Center for Social Dynamics, and Complexity, and the Center for Biodiversity Outcomes. Outside of ASU, he is an external faculty with the Human Computation Institute.