The new capabilities combine HawkEye 360's RF geolocation services with a customised machine learning model developed through Amazon Web Services' (AWS) Machine Learning (ML) Solutions Lab. The capabilities, which will be integrated into the HawkEye 360 portfolio of products, leverage underlying vessel characteristics and behaviour to predict whether a given vessel is likely to engage in similar activity as sanctioned vessels.
HawkEye 360 used Amazon SageMaker Autopilot, a fully managed service that helps make it easy to build, train and deploy ML models quickly, to develop the purpose-built, proprietary algorithms undergirding the new capabilities. These algorithms can help generate deeper insights into RF data in half the time than was previously possible.
The new algorithms evaluate vessels' historical data and known interactions, along with contextual vessel characteristics to generate insights into the complex connections involved in illicit maritime vessel activity, such as illegal fishing, human trafficking, ship-to-ship transfer of illegal goods, smuggling and more. This provides analysts with a holistic view of maritime activity and the ability to detect, predict and zoom in on high-risk activity.
This RF signals analysis and machine learning ability can help make the oceans a safe place by supporting a variety of applications, including commercial maritime activity, national security operations, maritime domain awareness, environmental protection and more.
"RF signals can provide valuable insight into commercial vessel activity across the globe, even when bad actors seek to hide their location," said Tim Pavlick, vice president of product at HawkEye 360. "With these machine learning-backed capabilities, we will empower customers to cut through an ocean full of noise to obtain more timely and critical insights from maritime RF data to improve mission outcomes and prevent illegal and illicit activities.”