Headquartered in London, UK, Orca AI is the creator of a first-of-its-kind digital watchkeeper to maximise voyage safety and operational efficiency for ships and fleets.
Powered by maritime purpose-built machine learning and computer vision technologies, the Orca AI platform empowers crew to make data-driven decisions in congested waters and in low visibility conditions. Additionally, it allows fleet managers and operators to have a better understanding of their fleets’ performance and identify unsafe or risky and inefficient behaviours easily.
In 2022, Orca AI empowered the world's first commercial autonomous voyage in partnership with Designing the Future of Full Autonomous Ships (DFFAS) and The Nippon Foundation.
Orca AI’s solution is already utilised by leading shipping companies worldwide, including Maran Tankers, MSC, SeaSpan and NYK with more than 1,000 vessels booked with the platform.
TechNews180 interviewed Shoval Bolotin, Orca AI's Chief Operations Officer, discussing their recent fundraising efforts and the groundbreaking maiden voyage of the world's first autonomous commercial ship, and more.
Neil: Congratulations on raising $23 million in new funding. Can you share how you plan to utilise these funds to specifically further develop and expand Orca AI's technology and operations?
Shoval: By raising this capital, Orca AI will be able to invest in technology, expand internationally, and grow. All of which will help contribute to reducing maritime carbon emissions, improving efficiency, and safety of navigation.
Neil: Orca AI powered the world's first autonomous commercial ship voyage in 2022. What were the biggest challenges you faced during this milestone, and how did you overcome them?
Shoval: The major challenge was Covid, when everything was remote. After the world was released from restrictions, Japan was still blocked. Another challenge was shipping the equipment and developing the process of working together with the companies. There was a lot of back and forth with both of our R&D departments, until we created this integration between the Orca system and the other manufacturing systems on this vessel. All sea trials were done without us being present in the lab or on the vessel. Once the trial of the test in the lab had ended, we had a call with the R&D arm of NYK Shipping, and we incorporated the feedback into the product to modify certain protocols between the companies.
Neil: Your platform has significantly reduced close encounters and crossing events for major shipping companies. Could you elaborate on the specific AI technologies and processes that contribute to these safety improvements?
Shoval: One of the major advantages of Orca is that, when it’s deployed over 400 vessels, each one of the vessels becomes another sensor and point of data. Any vessel which is running our system is actually recording data and covering more nautical miles. So, first of all we take the information that we gather and classify it for multiple cases. For example if visibility is good or bad, or cloudy or rainy. All of these factors have an impact on navigation. Whether it's a vessel or a fishing boat, a very small buoy or whale fins.
The ability of the Orca AI platform is to gather this information. We have the classification and what we call the annotators. So some of the algorithm actually learns by itself. One of the main differences in this industry is that the Orca platform is constantly improving, compared to more static radars.
In fact, one of the additional things that we’re doing is running feedback and remote sessions with the people on board to understand directly if there are things that for some reason were not detected or are not classified, so we can help close the loop. When we got started, one of the original comments was that the system is also detecting birds which it should not. So with our feedback process, this was easily fixed. We can now even detect small items like drones. So for us, this is a new opportunity. But this is an example of something we did not think of at the beginning, and then customers requested and we consequently responded.
The main point is that every vessel becomes like another sensor and it enables increased engagement with the crews to get feedback, so they can understand where we stand and what we need to improve.
Neil: In light of the increasing threats such as drone attacks and piracy, how does Orca AI's technology enhance the security measures onboard vessels?
Shoval: The drone aspect is still a work in progress, it's not something we’ve yet deployed. There's also a lot of requests regarding piracy. I will say that today, because of the view of the cameras, we are not looking at the aft. In our discussions with officers and fleet managers, we can see our system often records events and close encounters and we’re always looking at ways we can improve this and enhance the safety of the vessel itself.
Neil: Orca AI has achieved impressive results in reducing fuel consumption and CO2 emissions. Can you discuss how your platform's navigational decisions contribute to these environmental benefits?
Shoval: Yes. In the maritime industry, there is a huge hype on fuel efficiency and sustainability. But actually when we're training or teaching seafarers on a corporate vessel, this is not something that was previously touched upon. If you are now the officer, there's no system that gives them real-time feedback on things that they're doing. The crew are not measured on fuel safety. We’ve understood that our ability to give an early detection on different targets actually allows the crew on board to perform more drastic manoeuvring. Imagine we’re driving a car. You want to make a very common manoeuvre and another driver is reckless. The Orca platform gives you the ability to see there's a vessel that we will potentially need to clear. It gives you the ability to do this for 20 minutes. Then, the manoeuvre is easier and you consequently divert less from your route.
This is the first system that gives you personalised information and visibility.
Neil: Looking ahead to the rollout of fully autonomous ship technology in 2025, what are the key steps Orca AI is taking to ensure successful implementation and industry adoption?
Shoval: There are a few measures that we’re doing on the vessel. I think that the most important one is working with The International Maritime Organization. This is the organisation that regulates how to operate at sea. We're working with them on what we call the Watchkeeper. Once you have a computer vision platform, you can reduce the number of crew members that you need on certain shifts. The first step would be to reduce the workload they have which is one of their major problems, and in the future it will allow for reducing the crew, thus presenting more concrete ROI.
We’re also seeing IMO regulation that says you must have at least two watchkeepers. One of the things that we're pushing for is that you’ll have one human being watchkeeper and one platform and computer vision to enable this.
The second thing that was mentioned was partnering with insurance companies. We are now partnering with one such company and they have already seen improvement on the vessel that we are installed upon.
With the shipping companies themselves, they are occasionally reluctant for change and the team is constantly explaining where the value is. The things that we found within these conversations were concerns that this system would replace them. However, this couldn’t be further from the truth and are now treated as a helping hand for the crew, and definitely do not pose a threat to it. Orca is one part of the overall solution but in the future, we aim to be the full solution.
