BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
DTSTART:20230312T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20221106T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20221209T041700Z
UID:1D719C9A-6439-4810-881E-9F7FBB54EC0A
DTSTART;TZID=America/New_York:20221208T183000
DTEND;TZID=America/New_York:20221208T193000
DESCRIPTION:If a generalized likelihood ratio (GLR) test cannot make a good
  decision\, then there is no good decision to be made. If the test is as t
 o whether or not a very low observable (VLO) target is present in heavy cl
 utter\, the GLR should be the maximum-likelihood probabilistic data associ
 ation (MLPDA) tracker. The MLPDA is very effective\, but has several opera
 tional shortcomings that its close cousin\, the maximum-likelihood probabi
 listic multi-hypothesis tracker (MLPMHT) avoids. We will discuss and compa
 re both algorithms\, plus show some fortuitous new MLPMHT developments. Pe
 rhaps most interesting\, we are now able to set the MLPMHT threshold accur
 ately and confidently\, as would be a requirement for real-time operation.
  And since one cannot do better than ML\, we are now able to make fundamen
 tal statements about which targets can be tracked and which cannot: These 
 statements are essentially a bound\, as opposed to algorithm-specific perf
 ormance experience.\n\nSpeaker(s): Dr. Peter Willett\, \n\nRoom: Hazard Ro
 om\, Bldg: Coastal Institute (bldg. 26)\, 215 South Ferry Rd.\, Narraganse
 tt\, Rhode Island\, United States\, 02882
LOCATION:Room: Hazard Room\, Bldg: Coastal Institute (bldg. 26)\, 215 South
  Ferry Rd.\, Narragansett\, Rhode Island\, United States\, 02882
ORGANIZER:jason.e.gaudette@ieee.org
SEQUENCE:16
SUMMARY:Maximum-Likelihood Methods in Target Tracking and Fundamental Resul
 ts on Trackability
URL;VALUE=URI:https://events.vtools.ieee.org/m/331779
X-ALT-DESC:Description: <br /><p>If a generalized likelihood ratio (GLR) te
 st cannot make a good decision\, then there is no good decision to be made
 . If the test is as to whether or not a very low observable (VLO) target i
 s present in heavy clutter\, the GLR should be the maximum-likelihood prob
 abilistic data association (MLPDA) tracker. The MLPDA is very effective\, 
 but has several operational shortcomings that its close cousin\, the maxim
 um-likelihood probabilistic multi-hypothesis tracker (MLPMHT) avoids. We w
 ill discuss and compare both algorithms\, plus show some fortuitous new ML
 PMHT developments. Perhaps most interesting\, we are now able to set the M
 LPMHT threshold accurately and confidently\, as would be a requirement for
  real-time operation. And since one cannot do better than ML\, we are now 
 able to make fundamental statements about which targets can be tracked and
  which cannot: These statements are essentially a <em>bound\,</em> as oppo
 sed to algorithm-specific performance experience.</p>
END:VEVENT
END:VCALENDAR