There is a growing need for coverage detection of large maritime areas, mainly in the exclusive economic zone (EEZ). Since physically accessing such a large area is practically impossible, satellite-based sensors offer an efficient and cost-effective solution. However, detection of moving vessels with satellite sensors is a challenging problem. Satellite imagery is expensive, covers a very small area, and can be acquired only at predefined acquisition opportunities.
All civilian ships are required to transmit an Automatic Identification System (AIS) signal, which indicates the vessel’s position and identification parameters. However, since this information is unreliable, total maritime awareness is achieved through the integration and analysis of visual data from electro-optic (EO) and SAR satellites, enabling visual identification of vessels instead of relying only on cooperative vessel AIS transmission. The time it takes from vessel transmission of an AIS signal until it can be analyzed in ground segment (known as a data cycle) is typically over one hour. On the other hand, current commercial satellite mission planning may take several hours. Therefore, in order to task a satellite to detect a moving vessel with no additional aids after three hours, you must cover a very large area (about 17,000 sq km). This will end up costing US$150,000–$200,000 per ship, which is not affordable. At the same time, manually choosing relevant satellite imagery is equivalent to looking for a needle in a haystack: one satellite image is commonly of size 50–100 sq km (i.e., less than 1% of the total area’s size).
Natalie Fridman dives into this challenging problem and shares ISI’s AI-based solution along with successful examples of detecting maritime vessels with ISI’s satellites. Natalie focuses on efficient ways to decrease the search area for the vessel by using vessel behavior prediction to increase the probability of acquiring relevant information from the satellite imagery. Reduction of position uncertainty leads to a much more economical solution. Using a multiagent system, ISI developed a prediction algorithm for the vessel behavior and selection algorithm, which recommends the best satellite and observation window for a given mission. Deep learning enables the system to autonomously detect vessels in existing satellite imagery, and by correlating this detection with various other sensors, it can detect uncooperative vessels as well.
Natalie Fridman is the vice president of research and innovation at iSi, where she manages the development of the company’s newest and advanced technologies. Natalie’s expertise lies in algorithms for behavior analysis and prediction, decision making systems, and autonomous agents. Previously, she spent eight years at artificial intelligence and robotics research group the MAVERICK Lab; was a lecturer in the Computer Science Department at Bar Ilan University; and was an artificial intelligence team leader at Elbit Systems, where she managed all aspects of the research in artificial intelligence fields. Natalie has more than 20 publications, including highly refereed AI journal articles, conference papers, and book chapters, and has two patents for algorithms in the AI field. She holds a PhD in computer science from Bar Ilan University, where she was a president’s scholar for excellence in her studies.
©2018, O’Reilly UK Ltd • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org