BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:MechE Colloquium: From Motion to Mission Planning via Augmented Gr
 aphs of Convex Sets
DTSTART:20261006T120000
DTEND:20261006T130000
DTSTAMP:20260917T011104Z
UID:d6ea740fe7270afdc1363db2f70bd2759c5cad6323a27ba1ecd4050c
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Tyler Summers\, Mechanical engineering\, The University
  of Texas at Dallas\nAbstract: Robot motion planning has traditionally fo
 cused on navigating obstacle-laden environments: computing smooth\, collis
 ion-free trajectories from start to goal. Many real-world missions\, howev
 er\, require satisfying logical precedence constraints: collecting resourc
 es before accessing restricted zones\, completing subtasks in a prescribed
  order\, or acquiring tools before they can be used. This talk presents 
 the augmented graph of convex sets (augmented GCS) framework\, which unifi
 es continuous trajectory optimization and combinatorial task sequencing wi
 thin a single optimization problem.\nThe key insight is that a layered aug
 mented GCS\, built on an exact convex partition of the free space\, simult
 aneously selects an optimal task completion sequence and computes an optim
 al continuous trajectory. A shortest path in the augmented GCS solves both
  problems at once\, yielding an exact solution up to a finite Bézier curv
 e parameterization. The layered structure of the augmented GCS turns out t
 o implement precisely the Bellman-Held-Karp (BHK) dynamic programming algo
 rithm for the Traveling Salesman Problem\, establishing a formal correspon
 dence between our framework and the combinatorial TSP literature. This mak
 es augmented GCS a continuous-geometry generalization of BHK\, achieving t
 he same singly exponential worst-case complexity\, an exponential improvem
 ent over general-purpose temporal logic tools. We further develop a librar
 y of mission specification variations (including ordered collection\, disj
 unctive keys\, conjunctive doors\, timed constraints\, and conditional log
 ic) each with proven correctness\, substantially expanding the range of ex
 pressible mission types. Numerical experiments confirm exponential speedup
 s in practice and near-global-optimality on a large benchmark suite.\nTime
  permitting\, I will briefly discuss ongoing extensions\, including spacet
 ime augmented GCS for dynamic environments\, safety-aware planning via con
 formal prediction sets\, and a galactic survey benchmark inspired by space
  mission planning that stress-tests the framework at scale.\n\n\n\nBiograp
 hy: Tyler Summers is an associate professor at the University of Texas at
  Dallas. Prior to joining UT Dallas\, he was an ETH Postdoctoral Fellow a
 t the Automatic Control Laboratory at ETH Zurich from 2011 to 2015. He rec
 eived a PhD degree in Aerospace Engineering at the University of Texas at 
 Austin in 2010. He was a Fulbright Postgraduate Scholar at the Australian 
 National University in Canberra\, Australia in 2007-2008. He received the 
 National Science Foundation CAREER Award in 2021 and a Young Investigator 
 Program award from the US Army Research Office in 2017. His research inte
 rests are in feedback control\, optimization\, and learning in complex dyn
 amical networks\, with applications in robotics and power networks.
LOCATION:MED 0 1418 https://plan.epfl.ch/?room==MED%200%201418 https://epf
 l.zoom.us/j/61360740951
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR