Professor Etsuro Shimizu of the Tokyo University of Marine Science and Technology is collaborating with Groke Technologies to develop their Situational Awareness (SA) system as Groke works to deploy the system within the Japanese maritime industry. Professor Shimizu tells us that the system must be able to respond to a specific combination of issues that is particular to Japanese waters, one aspect of which is population.
“Anyone with the legal right to fish can basically just drop their gear wherever they need to, but to a certain degree, fixed nets are set up in specific places. But what about when someone is fishing for octopus, for example? They drop octopus pots attached to a bamboo rod that sticks out of the water with a triangular flag on top that tells people something is there. It’s difficult to avoid equipment like that. Japan has a high population, so there is a lot of equipment in the water and more fishing boats than you might find in other countries.”
One benefit of Groke’s SA system is that it can be used by larger vessels to detect and identify the vast number of fishing boats and their gear in the waters around Japan.
Some issues still need to be resolved, Professor Shimizu notes. Cameras must be able to distinguish between vessel running lights and sunlight reflected in the water. In addition, the light from buildings in heavily urbanized areas like Tokyo can reflect off the water’s surface and potentially confuse systems.
One new idea Professor Shimizu is looking forward to proposing to Groke is designed to improve safety for smaller vessels.
“When a fishing boat or other small vessel around ten meters long hits the wake of a larger vessel, it can sway and roll violently,” he explains. “I think we need the system to be able to detect and identify that sort of phenomenon so that smaller vessels can change course and avoid it. We’re working on this in our lab, but I want to talk to Groke about incorporating it into the next phase of the project, maybe specifically for small vessels since large vessels don’t really need to worry about it.”
The fact that Professor Shimizu feels himself free to offer new proposals to Groke is perhaps part of why he has such respect for them and their collaboration.
“Groke works very quickly, and they are very highly capable developers. To be honest, Japanese companies often move quite slowly. A company might work with AI, but they won’t know much about marine vessels. In contrast, the people at Groke thoroughly research what they need to know about ships. We also generally have similar ideas on the issues we think need to be solved. They seem to have a very deep understanding of how marine vessels work.”
Professor Shimizu and his research lab gain much from collaborating with a company in the marine industry.
“Our perspectives really do align regarding what needs urgent development, but as a university lab, we can’t do much. It’s not that the students don’t have the ability, it’s more that it’s hard to maintain research continuity because students are constantly leaving as new ones come in.”
Professor Shimizu feels that one benefit of industry over academia is that it is easier to accumulate knowledge in a continuous way when you are working with a company because, as a business, the company has a stake in achieving success.
He also tells us that, as an engineer, research can be uninteresting because much of it involves presenting results to other researchers at academic conferences. That is why he appreciates being able to actually contribute to a company building something that will go to market.
The fact that the ultimate goal is commercialization does make the work more interesting for Professor Shimizu, but he has another reason for working with Groke.
“Of all the things I’ve praised Groke for today, I think they are the best in the field of AI obstacle detection,” he says.
Professor Shimizu hopes that Groke will see comparable benefits from collaborating with his research laboratory.
“When you work in academia, you always have to be generating ideas that are a step ahead of everything else, and my goal is to provide Groke with information on research they haven’t even thought about yet.”
Professor Shimizu also expects that generational differences are another characteristic of academia that can be harnessed.
“The times are always changing, and we have the benefit of our students being very young. I think younger people have a different sensibility when it comes to obtaining information…I hope we can help Groke by serving as a place in which we incorporate the opinions of these young people – the people who will lead the next generation – and harness those new sensibilities in sharing ideas.”
Article published originally in Groke’s exclusive customer newsletter in April 2021
This is the second part of the interview. First part is found from here.