The U.S. Army has created a computational model that enables robotic systems to learn from soldiers via conversational interaction.
The service branch said Thursday it pursues this research in partnership with Tufts University under the Department of Defense's Laboratory University Collaboration Initiative (LUCI).
This research supports the Army's plan to implement autonomous robots that operate as part of soldier teams in the future. Researchers at the Army tested the model with a cognitive robotic architecture to demonstrate how the approach can support actual military activities.
“The model enables a robot to ask effective clarification questions based on its knowledge of the environment and to learn from the responses,” said Felix Gervits, researcher at the Army Combat Capabilities Development Command's Army Research Laboratory.
The team will move the project forward by making the model understand more types of questions.