The Canadian Space Agency has selected Dartmouth Nova Scotia’s Global Spatial Technology Solutions (GSTS) to receive a Phase 2 Innovative Solutions Canada contract.
The award, worth $1.096M, will help GSTS develop new AI-driven space-based ship-tracking capabilities that could be a game-changer for Canadian agencies. This follows from their 2019 Phase 1 award for a concept study of the same technology.
GSTS specializes in using machine learning algorithms for “big data” analytics for marine intelligence applications. A press release on their system, called “OCIANA,” describes it as an “AI-based platform that ingests large volumes of data from a range of satellite- and ocean-based sensors, which includes information on ocean, weather, vessel, port, and marine species activity.” The data is processed using their “proprietary techniques,” and is presented to decision makers that can use them for a variety of maritime-related applications.
In an interview with SpaceQ, GSTS CEO Richard Kolacz said they focus on providing three main capabilities using OCIANA. They provide performance management that uses OCIANA to “manage and monitor the health of equipment onboard a vessel in order to reduce lifecycle costs.” They also can help commercial and civil operators with vessel management, telling vessels, port managers and shipping companies whether there are weather conditions, environmental considerations, port issues, or other factors that may change vessels’ optimal speeds, routes, and itineraries. If there is a storm, an overly-busy port, or even a pod of endangered North Atlantic Right Whales that may affect the vessels, operators will know and be able to make proactive decisions. Kolacz said it serves as a kind of “air traffic control for ships.”
(Kolacz focused on the Right Whales as a key example. Vessels are required to reduce speed around the whales to 10 knots or less in order to avoid deadly collisions. Knowing the pods’ location and heading will help operators avoid the pods and/or plan around any delays from ships needing to limit their speeds.)
Their third offering is risk management. One of the key data sources for OCIANA are satellites that track large vessels’ location, heading and speed using their Automatic Identification System transponders, such as the M3MSat built by COMDEV (now part of Honeywell Canada) and operated by DRDC and the CSA until recently. There are now hundreds of AIS-capable satellites, and they let OCIANA track individual ships using AIS transponders and understand their “pattern of life.” These patterns can indicate ships in distress for vessel management purposes, but can also spot when ships are doing things that are out of the ordinary. The system can use this to provide risk flags and risk ratings for the vessel and its behavior. This capability is central to the CSA’s contract.
Kolacz provided some examples. “Has it changed its name? Has it rendezvoused with another vessel that has turned off all of its transmission features? Has it gone out of the area, and then something that it normally doesn’t do? Those are all indicative of trying to do something that is illegal, or incorrect.” This can include anything from smuggling of drugs, weapons or people, “trying to break an embargo,” “taking fuel from a location that they shouldn’t be,” or environmentally-damaging activities like dumping, illegal fishing, or traveling at high speed around whale pods. There are a variety of Canadian agencies that benefit from this risk analysis.
If a vessel turns off its transponder or changes its AIS identification, however, agencies will require another solution. That’s what the CSA is looking for. GSTS is tasked with developing “a novel approach to classify vessels solely from time series of Satellite Automatic Identification System (SAIS) positional data…to reveal positional idiosyncrasies. The approach exploits positional information to derive identification information entirely independent of the vessel’s stated identifiers.”
So even if a vessel is providing improper AIS information, or perhaps isn’t providing it at all, the vessel can still be uniquely identified and tracked.
Their proprietary OCIANA-based approach, Kolacz said, “doesn’t rely on the information of the vessel transmitting who they are.” Instead, “we’re taking a look at the profile of how the vessel behaves in certain sea conditions,” as well as identifiers in the AIS signal like “acoustic features or transmission signal features.” These factors, among many others, let OCIANA uniquely identify vessels no matter how they may identify themselves. He gave the example of gait analysis as an analogy, saying “you may put on a coat, or a disguise, or a costume, but by the way you walk down the street we can tell that it’s you.”
Kolacz said it will allow them (for example) to identify a vessel “trying to do some illegal fishing and then zip out and say that it wasn’t them or that they weren’t there. We could confirm that they were, in fact, the ones doing it.” Even without the AIS transponders, the system can recognize the “pattern of life” of a vessel, and flag potential risk situations for review or intervention. Kolacz said that he’s been told by officials that “no country in NATO has this capability,” though he couldn’t get into detail on defence applications.
The contract is for 18 months, and they’re aiming to develop preliminary initial operational capability in the next six to eight months, with both a prototype stage and a final commercial version to follow. Kolacz said the commercial version will likely be ready some time after the 18 months are complete, and that they’re already receiving interest from potential customers.
They’re also planning a Series A funding round in 2023, after having already raised $4.1M in a round that concluded in September, once they can “demonstrate the full potential to the customer base that we now have on the defence, security, and commercial side.”