BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20240310T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20231105T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20231117T210229Z UID:403D9FA4-3A33-4456-B929-9BD634AE0043 DTSTART;TZID=America/New_York:20231117T150000 DTEND;TZID=America/New_York:20231117T160000 DESCRIPTION:This work focuses on path planning for autonomous agents\, leve raging multiple sensing domains to provide navigation solutions in contest ed environments. The emphasis is on mutli-objective optimization\, finding optimal path costs that minimize uncertainty in the goal region.  The a lgorithm developed is based on the Rapidly-exploring Random Tree (RRT) pro babilistic planning algorithm\, but extends into the belief space to plan over uncertainty. The Rapidly-exploring Random Belief Alt-Nav Graph (RRBAN G) leverages the probabilistic guarantees of the RRT-based algorithms\, en suring the properties for probabilistic completeness and asymptotic optima lity. The algorithm is designed to be agent and measurement model agnostic \, but specifically how complementary navigation techniques obtain their m easurements when developing plans within a complex environment. The algori thm provides an offline\, initial plan for an agent given a priori world i nformation. There are several\, significant planned avenues for advancemen t\, targeting the algorithm itself\, extending to implement real-time dyna mic re-planning\, as well as benchmarking against other belief space plann ing (BSP) algorithms. Captain Machin will also discuss several ANT center research efforts focused on pushing Autonomy.\n\nCo-sponsored by: Wright-P att Multi-Intelligence Development Consortium (WPMDC)\, The DOD & DOE Comm unities\n\nSpeaker(s): Tim\n\nAgenda: \nThis work focuses on path planning for autonomous agents\, leveraging multiple sensing domains to provide na vigation solutions in contested environments. The emphasis is on mutli-obj ective optimization\, finding optimal path costs that minimize uncertainty in the goal region.  The algorithm developed is based on the Rapidly-ex ploring Random Tree (RRT) probabilistic planning algorithm\, but extends i nto the belief space to plan over uncertainty. The Rapidly-exploring Rando m Belief Alt-Nav Graph (RRBANG) leverages the probabilistic guarantees of the RRT-based algorithms\, ensuring the properties for probabilistic compl eteness and asymptotic optimality. The algorithm is designed to be agent a nd measurement model agnostic\, but specifically how complementary navigat ion techniques obtain their measurements when developing plans within a co mplex environment. The algorithm provides an offline\, initial plan for an agent given a priori world information. There are several\, significant p lanned avenues for advancement\, targeting the algorithm itself\, extendin g to implement real-time dynamic re-planning\, as well as benchmarking aga inst other belief space planning (BSP) algorithms. Captain Machin will als o discuss several ANT center research efforts focused on pushing Autonomy. \n\nVirtual: https://events.vtools.ieee.org/m/384182 LOCATION:Virtual: https://events.vtools.ieee.org/m/384182 ORGANIZER:a.j.terzuoli@ieee.org SEQUENCE:9 SUMMARY:Path Planning for Autonomous Agents URL;VALUE=URI:https://events.vtools.ieee.org/m/384182 X-ALT-DESC:Description: <br /><p>This work focuses on path planning for aut onomous agents\, leveraging multiple sensing domains to provide navigation solutions in contested environments. The emphasis is on mutli-objective o ptimization\, finding optimal path costs that minimize uncertainty in the goal region. &ensp\;The algorithm developed is based on the Rapidly-explor ing Random Tree (RRT) probabilistic planning algorithm\, but extends into the belief space to plan over uncertainty. The Rapidly-exploring Random Be lief Alt-Nav Graph (RRBANG) leverages the probabilistic guarantees of the RRT-based algorithms\, ensuring the properties for probabilistic completen ess and asymptotic optimality. The algorithm is designed to be agent and m easurement model agnostic\, but specifically how complementary navigation techniques obtain their measurements when developing plans within a comple x environment. The algorithm provides an offline\, initial plan for an age nt given a priori world information. There are several\, significant plann ed avenues for advancement\, targeting the algorithm itself\, extending to implement real-time dynamic re-planning\, as well as benchmarking against other belief space planning (BSP) algorithms.&nbsp\; Captain Machin will also discuss several ANT center research efforts focused on pushing Autono my.</p><br /><br />Agenda: <br /><p>This work focuses on path planning for autonomous agents\, leveraging multiple sensing domains to provide naviga tion solutions in contested environments. The emphasis is on mutli-objecti ve optimization\, finding optimal path costs that minimize uncertainty in the goal region. &ensp\;The algorithm developed is based on the Rapidly-ex ploring Random Tree (RRT) probabilistic planning algorithm\, but extends i nto the belief space to plan over uncertainty. The Rapidly-exploring Rando m Belief Alt-Nav Graph (RRBANG) leverages the probabilistic guarantees of the RRT-based algorithms\, ensuring the properties for probabilistic compl eteness and asymptotic optimality. The algorithm is designed to be agent a nd measurement model agnostic\, but specifically how complementary navigat ion techniques obtain their measurements when developing plans within a co mplex environment. The algorithm provides an offline\, initial plan for an agent given a priori world information. There are several\, significant p lanned avenues for advancement\, targeting the algorithm itself\, extendin g to implement real-time dynamic re-planning\, as well as benchmarking aga inst other belief space planning (BSP) algorithms.&nbsp\; Captain Machin w ill also discuss several ANT center research efforts focused on pushing Au tonomy.</p> END:VEVENT END:VCALENDAR