LJAD : "Learning to navigate in dynamic environments: animal behavior and artificial intelligence" 10.5.2017 11:00 pm at the seminar room of LJAD

Next Wednesday, 10.5.2017, Antonio Celani from International Center for
Theoretical Physics (ICTP, Trieste, Italy) will deliver a seminar at
11:00 pm at the seminar room of Dieudonné Lab.(Valrose)

Title:

Learning to navigate in dynamic environments: animal behavior and
artificial intelligence

Abstract:
I will discuss two examples of difficult navigation tasks in highly
dynamical environments. The first one is thermal soaring in birds. Birds
and gliders piggyback ascending currents to fly with a reduced
expenditure of energy, for example, during migration, and to extend
their flying range. Flow in the thermals is highly turbulent, which
poses the challenge of the orientation in strongly fluctuating
environments. The second one is gravitaxis by smart microswimmers. These
are self-propelled particles that can sense basic mechanical cues from
their surroundings and respond to them in order to reach the highest
altitude within some time horizon. We combine numerical simulations of
fluid flow with reinforcement learning methods to identify strategies of
navigation that can cope with potentially dangerous flow configurations
and exploit favorable ones.