BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TU/e ME//Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Amsterdam
BEGIN:STANDARD
DTSTART:19701025T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:19700329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:event-47@tuemeche.nl
DTSTAMP:20261007T234657Z
DTSTART;TZID=Europe/Amsterdam:20261103T133000
DTEND;TZID=Europe/Amsterdam:20261103T150000
SUMMARY:World Modelling and Decision-Making for Robots in Precision Agric
 ulture
DESCRIPTION:Speaker: Ruben Beumer\nHost: René van de Molengraft\n\nThis 
 thesis investigates world modelling and decision-making for autonomous ro
 botic systems operating in precision agriculture under uncertainty and re
 source constraints. It contains four main contributions\, spanning both t
 heoretical developments and practical robotic applications.\n\nThe first 
 contribution addresses decision-making under costly sensing and actuation
  within a partially observable Markov decision process (POMDP) framework.
  The problem of when to sample and actuate is formulated as a stopping-ti
 me problem\, capturing the trade-off between state-dependent costs and in
 tervention frequency. Due to the intractability of optimal solutions\, ap
 proximate methods based on relaxed dynamic programming and event-triggere
 d control are developed\, providing performance guarantees. In addition\,
  a modified relaxed dynamic programming framework is introduced to explic
 itly bound the complexity\, while still providing performance guarantees\
 , designed through value function approximations and optimization via lin
 ear matrix inequalities.\n\nThe second contribution considers selective h
 arvesting of delicate crops\, such as table grapes\, under uncertainty. A
  method is presented that integrates multi-view mapping and tracking to i
 mprove the quality of crop information\, together with a recursive decisi
 on-making algorithm based on graphs containing reachability dependencies 
 between the products to optimize the harvesting order with respect to bot
 h product quality and execution time.\n\nThe third contribution focuses o
 n semantic world modeling for agricultural robots. A framework is develop
 ed that combines probabilistic object mapping with graph-based simultaneo
 us localization and mapping (SLAM)\, enabling the construction of semanti
 cally rich maps while maintaining accurate localization without relying s
 olely on GPS. Detected plant properties\, such as semantic attributes inc
 luding type and size\, not only enrich the resulting map but are also use
 d to improve data association.\n\nThe fourth contribution explores autono
 mous mechanical weeding using a small\, legged robotic platform. A system
  is developed that leverages the mobility of a quadruped robot equipped w
 ith a custom end-effector\, enabling precise weed removal while reducing 
 soil compaction. The proposed approach is supported by an integrated soft
 ware architecture and evaluated in both indoor and outdoor environments.\
 n\nOverall\, the thesis provides a perspective on uncertainty-aware world
  modeling and decision-making\, demonstrating how theoretical methods and
  robotic system design can be combined to address key challenges in preci
 sion agriculture.\n\nMore info: https://research.tue.nl/en/persons/ruben-
 m-beumer/
LOCATION:Atlas 0.710
URL:https://tuemeche.nl/peoplepages/event.php?id=47
CATEGORIES:PhD Defense
END:VEVENT
END:VCALENDAR