BEGIN:VCALENDAR
CALSCALE:GREGORIAN
VERSION:2.0
METHOD:PUBLISH
PRODID:-//Drupal iCal API//EN
X-WR-TIMEZONE:America/New_York
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
DTSTART:20070311T020000
TZNAME:EDT
TZOFFSETTO:-0400
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
DTSTART:20071104T020000
TZNAME:EST
TZOFFSETTO:-0500
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
SEQUENCE:1
X-APPLE-TRAVEL-ADVISORY-BEHAVIOR:AUTOMATIC
234151
20260403T151506Z
DTSTART;TZID=America/New_York:20260408T120000
DTEND;TZID=America/New_York:2
 0260408T130000
URL;TYPE=URI:https://www.wpi.edu/news/calendar/events/robot
 ics-engineering-colloquium-speaking-series-professor-george-konidaris
Robotics Engineering Colloquium Speaking Series: Professor George Konidaris
Unifying the Stack: A Principled Structuralist Approach to Intelligent Robo
 t Control\n\n\n\n      \n      \n\n\n\nAbstract: There are two dominant ap
 proaches to designing intelligent robots. One, typified by language behavi
 or models, leverages unstructured deep neural networks and learning from d
 emonstration to generate behavior. These approaches have had several impre
 ssive successes but face scaling, trust, and explainability challenges. Th
 e second approach seeks to integrate, rather than discard, technologies fr
 om existing subfields (like motion planning and SLAM) into a coherent cont
 rol architecture that retains their favorable properties while accessing t
 he strengths of deep network. The primary challenge here is that there is 
 no unifying theoretical framework for all of robotics: each subfield was d
 esigned and studied largely in isolation. I will propose a unifying framew
 ork that models the control stack as layers of increasingly abstract decis
 ion processes. Each layer combines perceptual and action abstractions, to 
 generate a more tractable decision process by exploiting structure in the 
 world or the robot. Existing technologies fit naturally into this stack as
  observation or action abstractions. The result is a natural hierarchy wit
 h a few missing technologies. I will discuss my group's recent results on 
 both filling in these missing technologies, and more generally in learning
  decision process abstractions from pixel-level data.\nBio: George Konidar
 is is an Associate Professor of Computer Science at Brown, where he direct
 s the Intelligent Robot Lab. George is also the co-founder of two technolo
 gy startups: Realtime Robotics, which commercializes his work on hardware-
 accelerated motion planning, and Lelapa AI, which is based in his home cou
 ntry of South Africa and develops African language models. George is the r
 ecent recipient of an NSF CAREER award, young faculty awards from DARPA an
 d the AFOSR, and the IJCAI-JAIR Best Paper Prize.\n
END:VEVENT
END:VCALENDAR