Tech talk: Individual and Group Decision-making in Open Environments from an Ecological Point of View

  • Date Tuesday, December 08, 2015  Time 11:00 AM
  • Speaker R. Andrew Goodwin, PhD
  • Location Galois Inc, 421 SW 6th Ave. Suite 300, Portland, OR, USA, (3rd floor of the Commonwealth building)
  • Galois is pleased to host the following tech talk.
    These talks are open to the interested public--please join us!
    (There is no need to pre-register for the talk.)

Animal movement is driven by responses to both social and non-social factors of the environment. Research across species shows that animals do not make decisions based solely on present information, e.g., at time t, but on an integrated assessment of information over recent time, e.g., t-1, t-2, etc. This raises the conundrum of how to relate distributions of animals (or humans) to conditions of the environment since prior individual experiences, possibly to unobserved conditions, contribute to the behaviors observed. I use a rich animal and environmental data set to harness and integrate the broad spectrum of behavioral research and construct a theoretical platform for describing the observed distributions of species populations. My work in the aquatic, terrestrial, and aerial environments falls into two categories, (i) better managing civil infrastructure that impacts animal movement and (ii) developing/evaluating computational abstractions of environmental sensing and individual decision-making to inform the engineering design of distributed systems in unbounded, open field environments. In the aquatic realm, I’ll share research suggesting that water acceleration “shapes” the individual movement trajectories of juvenile salmon near hydropower dams. Then, I’ll discuss how I am building off this research to understand Sonoran Pronghorn that can interfere with military training activities in the Arizona desert. Lastly, I’ll discuss bird flocking as a biological network that can inform the design of fast, light, mobile, compact, and collapsible systems (e.g., geospatial sensors) that are resilient to attack and can quickly hide.

R. Andrew Goodwin, PhD