Multi-Agent Robotic System In Dynamic Environments With AI-Driven Networked 4D Imaging Radars Based Perception System

The U.S. Army’s  Robotic and Autonomous Systems Strategy focuses on leveraging advanced technologies, including autonomous networked systems, AI, and machine learning, to build a more agile and capable force. This includes development of collaborative and multi-agent systems where multiple networked autonomous platforms work together in coordinated operations, particularly in complex or contested environments. MAS is key to increasing operational efficiency and effectiveness on future battlefields.1Autonomous robotic systems such Unmanned Ground Vehicles (UGV’s), Unmanned Aerial Vehicles (UAV’s), are critical assets to the Department of the Air Force (DAF) capable of performing a wide range of tasks like surveillance and reconnaissance and logistical support. Current sensor technology needed to enable multi-agent robotic systems faces limitations in terms of environmental adaptability, range, field of view, accuracy, as well as cost and scalability. Achieving highly accurate range, velocity, and azimuth/elevation angle of arrival measurements remain challenging. Additionally, the power and weight requirements of current DAF sensor technologies impose constraints.