BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards Physical AI: Time-Series Prediction for Intent-Aware Robot
  Learning - Mukund Mitra\, Indian Institute of Science
DTSTART:20250805T110000Z
DTEND:20250805T120000Z
UID:TALK234802@talks.cam.ac.uk
CONTACT:Melanie Ellwood
DESCRIPTION:Understanding human intent is fundamental to robots that can c
 ollaborate naturally and effectively. Intent prediction involves forecasti
 ng time-series data - such as human motion trajectories\, gaze patterns\, 
 and interaction data - to enable machines to anticipate human actions\, re
 spond appropriately\, and learn from interaction. This capability paves th
 e way for safer\, faster\, and intuitive human-robot collaboration.\n\nThi
 s work presents a framework that combines Imitation Learning techniques wi
 th Foundation Models to advance intent-aware robot learning. The approach 
 is demonstrated across diverse tasks\, including target prediction in exte
 nded reality\, human-robot handovers\, and multi-robot coordination. By le
 veraging multimodal cues—such as hand motion\, gaze\, and interaction hi
 story - the system enhances prediction accuracy. Additionally\, large lang
 uage and vision models enable the interpretation of high-level human instr
 uctions for task planning and robot navigation. Together\, these contribut
 ions move toward the goal of Physical AI\, where robots can learn from hum
 ans and understand and act on their intent in real-world environments.\n\n
 *Bio*: Mukund Mitra holds a B.Tech in Mechanical Engineering from NIT Raip
 ur and is currently affiliated with the Robert Bosch Centre for Cyber-Phys
 ical Systems at the Indian Institute of Science (IISc) Bangalore. He was a
 warded the Prime Minister’s Research Fellowship (PMRF)\, conferred to th
 e top 1% of researchers in India. He has published in top venues including
  ICRA\, ACM IUI\, and ACM Transactions on Human-Robot Interaction (THRI).\
 n\nHis research focuses on developing predictive models for Physical Artif
 icial Intelligence\, with emphasis on imitation learning and generative mo
 dels. His work contributes to time-series data prediction\, with applicati
 ons spanning user interface design\, eXtended Reality (XR)\, and motion pl
 anning for autonomous systems.
LOCATION: Cambridge University Engineering Department\, JDB Seminar Room
END:VEVENT
END:VCALENDAR