BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:65c488382d4d85cbf24df7deeac96948
DTSTAMP:20260809T164924Z
SUMMARY:Control and Optimisation seminar – Alex Vladimirsky (Cornell)
DESCRIPTION:Title\nSurveillance-evasion games under uncertainty\nAbstract\n
 Adversarial path planning problems are important in robotics applications 
 and in modeling the behavior of humans in dangerous environments. Surveill
 ance-Evasion (SE) games form an important subset of such problems and requ
 ire a blend of numerical techniques from multiobjective dynamic programmin
 g\, game theory\, numerics for Hamilton-Jacobi PDEs\, and convex optimizat
 ion. We model the basic SE problem as a semi-infinite zero-sum game betwee
 n two players: an Observer (O) and an Evader (E) traveling through a domai
 n with occluding obstacles. O chooses a pdf over a finite set of predefine
 d surveillance plans\, while E chooses a pdf over an infinite set of traje
 ctories that bring it to a target location. The focus of this game is on 
 “E’s expected cumulative exposure to O”\, and we have recently devel
 oped an algorithm for finding the Nash Equilibrium open-loop policies for 
 both players. I will use numerical experiments to illustrate algorithmic e
 xtensions to handle multiple Evaders\, moving Observes\, and anisotropic o
 bservation sensors. Joint work with M.Gilles\, E.Cartee\, and REU particip
 ants.\nBio\n\nAlex Vladimirsky is an applied mathematician at Cornell Univ
 ersity\, whose research interests include numerical analysis\, nonlinear P
 DEs\, dynamical systems\, optimal control & differential games\, mathemati
 cal biology\, and algorithms on graphs. His past and current projects incl
 ude numerical methods for (& homogenization of) Hamilton-Jacobi PDEs\, mul
 tiobjective & randomly-terminated optimal control\, uncertainty quantifica
 tion in randomly-switching systems\, multipopulation mean-field games\, su
 rveillance-evasion games under uncertainty\, inverse problems in geophysic
 al explorations\, dimensional reduction in turbulent combustion\, approxim
 ation of invariant manifolds\, macro-scale models of pedestrian interactio
 ns\, optimization of drug therapies for cancer patients\, models of microb
 ial competition\, sailboat navigation under wind-pattern uncertainty\, beh
 avioral ecology\, immunology\, and traffic engineering. Vladimirsky joined
  Cornell University in 2001 after receiving his Ph.D in Applied Mathematic
 s at UC Berkeley under supervision of James Sethian . Vladimirsky’s rese
 arch has been partially funded by the US National Science Foundation\, the
  Air Force Office of Scientific Research\, the Simons Foundation\, and the
  Royal Society (through a Wolfson Visiting Fellowship to support his sabba
 tical year at Imperial College London in 2025-26).\n
URL:https://www.imperial.ac.uk/events/201280/control-and-optimisation-semin
 ar-alex-vladimirsky-cornell/
DTSTART;TZID=Europe/London:20251105T140000
DTEND;TZID=Europe/London:20251105T150000
LOCATION:410\, Huxley Building\, South Kensington Campus\, Imperial College
  London\, London\, SW7 2AZ\, United Kingdom
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:STANDARD
DTSTART:20251105T140000
TZNAME:GMT
TZOFFSETTO:+0000
TZOFFSETFROM:+0000
END:STANDARD
END:VTIMEZONE
END:VCALENDAR