The US Army is using its operational data to develop artificial intelligence models that can automatically recognize and track targets across military missions.
Sightline Intelligence’s Aided Target Recognition (AiTR) platform will give army teams tools to train and evaluate target-recognition models using military datasets. The platform includes model-development tools and a training interface for government users.
The resulting models could support a range of army applications, including missile seekers, launched effects, intelligence, surveillance, and reconnaissance, long-range precision fires, and air and missile defense.
Turning Army Data Into Mission AI
The effort is being carried out under a Cooperative Research and Development Agreement between Sightline and the US Army Combat Capabilities Development Command Aviation & Missile Center (DEVCOM AvMC).
Sightline’s technology uses AI-enabled image processing at the edge to detect, classify, track, and geolocate targets.
But for the AI models to operate effectively in specific missions, they need access to relevant operational data.
“One of the biggest challenges in defense AI is that the most important operational data isn’t available commercially,” said Sightline chief product officer Stephen Bornstein.
“This collaboration allows DEVCOM AvMC to leverage its operational expertise and army-relevant datasets to create capabilities tailored to real-world army requirements, while providing a path for those capabilities to be developed, deployed, and sustained at scale.”