ML-based control architecture for uncrewed surface vessels in development for DARPA
NewsJuly 11, 2023
CAMBRIDGE, Massachusetts. Boeing subsidiary Aurora Flight Sciences is working on the development and testing of machine learning-based introspection technologies to help uncrewed surface vessels (USV) better operate in a maritime environment, the company announced in a statement.
These technologies aim to equip physical systems, such as UVSs, with the ability to adapt their control laws in response to uncertainties or unexpected events, the company says.
This work is a part of the seedling program titled "Fast Adaptation and Learning for Control Online" (FALCON) in support of the Defense Advanced Research Projects Agency's (DARPA) Learning Introspective Control (LINC) program.
Aurora's FALCON solution is an online learning and adaptive control architecture designed to help USVs more quickly adapt to changing dynamics at sea, making it more effective and survivable, the statement reads.
