BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
BEGIN:STANDARD
DTSTART:19510909T000000
TZOFFSETFROM:+1000
TZOFFSETTO:+0900
TZNAME:JST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260829T061514Z
UID:6D740F0A-EB5D-460D-902F-FDE3C97DC31A
DTSTART;TZID=Asia/Tokyo:20260925T110000
DTEND;TZID=Asia/Tokyo:20260925T120100
DESCRIPTION:Pacific (or Vancouver\, Canada) time = September 24\, 2026 7-8p
 m.\n\nAbout the Talk:\nIntelligent control of UAVs (Uncrewed Aerial Vehicl
 es) swarms typically requires the swarm to navigate effectively while avoi
 ding obstacles and achieving continuous coverage over multiple mission tar
 gets. Although traditional Multi-Agent Reinforcement Learning (MARL) appro
 aches offer dynamic adaptability\, they are hindered by the semantic gap i
 n black-boxed communication and the rigidity of homogeneous role structure
 s\, resulting in poor generalization and limited task scalability. Recent 
 advances in Large Language Model (LLM)-based control frameworks demonstrat
 e strong semantic reasoning capabilities by leveraging extensive prior kno
 wledge. Nevertheless\, due to the lack of online learning and over-relianc
 e on static priors\, these works often struggle with effective exploration
 \, leading to reduced individual potential and overall system performance.
 \n\nThis session will feature Prof. Honggang Zhang from Macau University o
 f Science and Technology\, who will address these limitations by proposing
  RALLY: a role-adaptive navigation framework that leverages Large Language
  Models (LLMs) for autonomous\, collaborative coordination among UAV swarm
 s. The talk addresses key limitations of conventional multi-agent reinforc
 ement learning by introducing LLM-driven semantic reasoning and adaptive r
 ole-switching. We believe this will be of great interest to anyone working
  in multi-agent systems\, robotics\, or AI-driven control frameworks.\n\nT
 alk Title:\nRALLY: Role-Adaptive LLM-Driven Yoked Navigation for Agentic U
 AV Swarms\n\nSpeaker:\n\nProf. Honggang Zhang\nSchool of Computer Science 
 and Engineering\nMacau University of Science and Technology\n\nModerator:\
 n\nEdward Au\, Ph.D. (Editor-in-Chief\, IEEE Open Journal of Vehicular Tec
 hnology)\n\nCo-sponsored by: IEEE Japan Office (Website:https://jp.ieee.or
 g/)\n\nVirtual: https://events.vtools.ieee.org/m/574633
LOCATION:Virtual: https://events.vtools.ieee.org/m/574633
ORGANIZER:mr_anthony_cheung@ieee.org
SEQUENCE:82
SUMMARY:RALLY: Role-Adaptive LLM-Driven Yoked Navigation for Agentic UAV Sw
 arms
URL;VALUE=URI:https://events.vtools.ieee.org/m/574633
X-ALT-DESC:Description: &lt;br /&gt;&lt;p&gt;&lt;span lang=&quot;EN-US&quot;&gt;&lt;strong&gt;Pacific (or Van
 couver\, Canada) time = September 24\, 2026 7-8pm.&lt;/strong&gt;&lt;/span&gt;&lt;/p&gt;\n&lt;p
 &gt;&lt;span lang=&quot;EN-US&quot;&gt;&lt;strong&gt;About the Talk:&lt;/strong&gt;&lt;br&gt;&lt;/span&gt;&lt;span lang=
 &quot;EN-US&quot;&gt;Intelligent control of UAVs (Uncrewed Aerial Vehicles) swarms typi
 cally requires the swarm to navigate effectively while avoiding obstacles 
 and achieving continuous coverage over multiple mission targets. Although 
 traditional Multi-Agent Reinforcement Learning (MARL) approaches offer dyn
 amic adaptability\, they are hindered by the semantic gap in black-boxed c
 ommunication and the rigidity of homogeneous role structures\, resulting i
 n poor generalization and limited task scalability. Recent advances in Lar
 ge Language Model (LLM)-&lt;/span&gt;&lt;span lang=&quot;EN-US&quot;&gt;based control frameworks
  demonstrate strong semantic reasoning capabilities by leveraging extensiv
 e prior knowledge. Nevertheless\, due to the lack of online learning and o
 ver-reliance on static priors\, these works often struggle with effective 
 exploration\, leading to reduced individual potential and overall system p
 erformance.&amp;nbsp\;&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;span lang=&quot;EN-US&quot;&gt;&amp;nb
 sp\;&lt;/span&gt;&lt;/p&gt;\n&lt;p&gt;&lt;span lang=&quot;EN-US&quot;&gt;This session will feature Prof. Hon
 ggang Zhang from Macau University of Science and Technology\, who will add
 ress these limitations by proposing RALLY: a role-adaptive navigation fram
 ework that leverages Large Language Models (LLMs) for autonomous\, collabo
 rative coordination among UAV swarms. The talk addresses key limitations o
 f conventional multi-agent reinforcement learning by introducing LLM-drive
 n semantic reasoning and adaptive role-switching. We believe this will be 
 of great interest to anyone working in multi-agent systems\, robotics\, or
  AI-driven control frameworks.&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;span lang
 =&quot;EN-US&quot;&gt;&amp;nbsp\;&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;strong&gt;&lt;span lang=&quot;EN-U
 S&quot;&gt;Talk Title:&lt;/span&gt;&lt;/strong&gt;&lt;span lang=&quot;EN-US&quot;&gt;&lt;br&gt;RALLY: Role-Adaptive 
 LLM-Driven Yoked Navigation for Agentic UAV Swarms&lt;br&gt;&lt;br&gt;&lt;strong&gt;Speaker:
 &lt;/strong&gt;&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;span lang=&quot;EN-US&quot;&gt;Prof. Hongga
 ng Zhang&lt;br&gt;School of Computer Science and Engineering&lt;br&gt;Macau University
  of Science and Technology&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;span lang=&quot;EN
 -US&quot;&gt;&amp;nbsp\;&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;strong&gt;&lt;span lang=&quot;EN-US&quot;&gt;M
 oderator:&lt;/span&gt;&lt;/strong&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;span lang=&quot;EN-US&quot;&gt;Edw
 ard Au\, Ph.D. (Editor-in-Chief\, IEEE Open Journal of Vehicular Technolog
 y)&lt;/span&gt;&lt;/p&gt;
END:VEVENT
END:VCALENDAR