Prev-Talk | Next-Talk | All-Talks | Talks-Sorted | MOOS-DAWG'26
Talk-32: AI Based Planning for Small Fleets
Prof. Greg Dudek
McGill University, Centre for Intelligent Machines
Coordinating teams of autonomous vehicles across marine, aerial, and terrestrial domains presents a growing command-and-control challenge. As the number and diversity of vehicles increase, operators must translate mission intent into platform-specific tasks while monitoring vehicle status, sensor information, communications, and changing conditions. Independent Robotics is developing IMPAC, a mission-control system that enables operators to direct heterogeneous multi-agent systems using natural-language commands.
IMPAC translates operator intent into executable mission plans while emphasizing interoperability, human oversight, and traceable decision-making. Designed to operate at the tactical edge under limited or intermittent communications, the system aims to reduce operator workload and support coordinated missions involving USVs, AUVs, UAVs, and other autonomous platforms. This talk will provide an overview of IMPAC, its approach to human-machine interaction, and its ongoing operational evaluation with Defence Research and Development Canada.
Categories:
- Collaborative Autonomy
- Planning