BEGIN:VCALENDAR

VERSION:2.0

PRODID:-//wp-events-plugin.com//6.6.4.4//EN

TZID:Asia/Jerusalem

X-WR-TIMEZONE:Asia/Jerusalem
BEGIN:VEVENT

UID:0-802@aerospace.technion.ac.il

DTSTART;TZID=Asia/Jerusalem:20240416T110000

DTEND;TZID=Asia/Jerusalem:20240416T120000

DTSTAMP:20240331T070615Z

URL:https://aerospace.technion.ac.il/events/structure-aware-probabilistic-
 inference-and-belief-space-planning-with-performance-guarantees/

SUMMARY:Structure Aware Probabilistic Inference and Belief Space Planning w
 ith Performance Guarantees
DESCRIPTION:Lecturer:Moshe Shienman\n Faculty:Technion Autonomous Systems P
 rogram\n Institute:Technion – Israel Institute of Technology\n Location:
 Zoom\n Zoom: https://technion.zoom.us/j/95719271567?pwd=WUNGZjVxS1A2bG5TZF
 RvRDBBN3NSUT09\n Abstract: \n\n\n\nIntelligent autonomous agents and robot
 s are increasingly utilized across various domains\, impacting our daily l
 ives in significant ways. From robotic surgery to automated warehousing\, 
 these agents are often expected to operate reliably and efficiently despit
 e facing limited environmental knowledge and uncertainty. Uncertainty aris
 es from various factors\, including noisy or restricted observations due t
 o physical constraints\, imprecise action execution\, and unpredictable ev
 ents in dynamic environments. In such scenarios\, a truly autonomous agent
  should be able to perform both inference\, which involves maintaining a b
 elief over the high-dimensional state based on available information\, and
  decision making under uncertainty\, also known as Belief Space Planning (
 BSP). In BSP\, the agent autonomously determines its optimal next actions 
 while considering the future evolution of beliefs. However\, solving bot t
 hese problems is computationally expensive and practically infeasible in r
 eal-world autonomous systems\, where the agent is required to operate in r
 eal time\, using inexpensive hardware and limited resources.\nIn our resea
 rch\, we utilize both topological structures\, induced from graph represen
 tations of posterior beliefs\, and specific structures of posterior distri
 butions\, to efficiently perform inference and BSP in high dimensional sta
 te spaces. Specifically\, for BSP\, we introduce a novel concept that leve
 rages topological signatures to approximate the solution. We establish ana
 lytical bounds for this approximation to ensure performance and empiricall
 y quantify its computational efficiency. In perceptually aliased environme
 nts\, where data association is not solved and posterior distributions are
  multi modal\, we leverage the structure of posterior distributions to bou
 nd the information loss when pruning hypotheses. This enables us to effici
 ently solve nonmyopic BSP problems without compromising on the quality of 
 the solution. Finally\, we introduce an innovative method for incremental 
 nonparametric probabilistic inference. Our approach leverages slices from 
 high-dimensional surfaces to efficiently approximate posterior distributio
 ns of any shape.\n\n\n\n Details: \n 
CATEGORIES:Seminars
LOCATION:Zoom

END:VEVENT

BEGIN:VTIMEZONE

TZID:Asia/Jerusalem

X-LIC-LOCATION:Asia/Jerusalem

BEGIN:DAYLIGHT

DTSTART:20240329T030000

TZOFFSETFROM:+0200

TZOFFSETTO:+0300

TZNAME:IDT

END:DAYLIGHT

END:VTIMEZONE
END:VCALENDAR