To invent something that will look for people missing in the forest is not only an engineering task. Now any, even the most advanced technology, will only be a tool in the hands of people who should be able to search even with drones, even with wooden torches.
When engineers were looking for people with a thermal imager at the Odyssey competition, they all the same studied the surroundings and tried to understand how the missing person would behave in them. When they searched with the help of sound beacons, they relied even more not on their sound, but on the correct placement tactics developed over the years.
One of the experts of the competition was Grigory Sergeev, founder of the search squad “Lisa Alert”. During the final, I asked him about the decisions of the teams and the application of technologies not in competitive conditions, but in real searches.
If all of a sudden you don’t understand what the eyeliner was talking about - what kind of “Odyssey”, what kind of finale, what kind of thermal imagers and sound beacons - look at the previous articles on the topic:
- Story about the semi-finals of the Odyssey contest
- Report from the finals, where teams tried to find people on an area of 314 km 2 in 10 hours
- Full interview with technical expert of the contest Nikita Kalinovsky
- Report from the Beeline and Lisa Alert press conference on the launch of the search neural network
Grigory Sergeev
“Do you really have sound beacons in service?”
- This is a great thing. We have long been using sound beacons mounted on cars. These are sirens of special services that we ask you to put in a certain place, and pneumatic beeps in order to bring a person to the sound. Mobile beacons are very convenient.
We have such portable Nakhodka teams, similar in technology to lighthouses, but we don’t place them, but we go with them - so far in experimental mode. For example, "Extreme" have been used for a long time. It works for them instead of screaming in the forest. First they buzz - they wake a missing person, who can sleep at that moment, and then shout themselves.
A person who has been in the forest for more than one day periodically falls into a state of drowsiness, and he can skip the usual cry, so buzzing is useful. Even if it’s a moving group in search of a bedridden person. The rest, in different conditions, can come to the sound beacon themselves. But it is clear that this is not enough, and some combination of technologies is needed. For example, when we can photograph absolutely the entire search area in good quality and use an algorithm or a neural network to look at it.
The detachment "Lisa Alert" now uses ordinary dji mavic 2 drones. It flies, photographs the entire territory that interests us. The only problem is that it cannot fly far and covers only a small area. Then we upload these photos, and instead of the neuron, live people work who look at the photos. So we find the missing.
- You had a neural network?
- Recently launched a story with the Beeline neural network. I was already here at the competition, and this thing found a man in the Kaluga region. That is, here is the real application of modern technology. For search this is really useful.
It is very important to have a medium that flies for a long time and which allows you not to blur photos, especially at sunrise and sunset, when there is practically no light in the forest. If optics allows this, this is very good. In addition, everyone is experimenting with thermal imagers. In principle, the trend is correct, but the issue of price is always worrying.
- I asked you about thermal imagers in the summer at the launch of the Beeline neuron. You then categorically said that these were all fantasies, and in ten years you haven’t found a single thermal imager.
- We still have not found anyone with a thermal imager.
- But here everyone speaks well of the first team that uses it.
- The story is like that. Today I see something that I did not see then at a press conference with Beeline - a fall in prices and the emergence of Chinese models. This allows you to use normal high-quality thermal imagers on drones larger than Mavik. But for now, it still comes out wildly expensive. Dropping such a thing is twice as painful as the drone itself. And if the drone can already show something, then the imager needs specific conditions.
To fantasize that we will find someone under the crowns is not worth it. We will not find anyone, the crowns are not transparent to the greenhouse. In the south of Russia, starting from Voronezh, where there are continuous fields, it is impossible to work without it, otherwise we will not be able to use night time. But the technology should be one that is suitable for all units - and Voronezh, and Lipetsk, and Tambov.
- Why not customize the equipment? All areas are different, everywhere different searches.
- Of course, you need to touch the terrain, but today the thermal imager, which can show something decently, costs more than six hundred thousand. The second Mavik costs 120, plus batteries. As a result, for 200 we will definitely pack a complete set that can fly exactly as much as necessary - when the first battery has sat down, the second has already been charged. With this approach, we can buy six Maviks without a thermal imager for one thermal imager. Naturally, we will act with the Maviks.
“But a drone is even more expensive than a thermal imager.” It costs more than a million.
- The story is that while searches are not funded, while there are no government programs or support for large businesses, this is really just a technology contest. We are looking at a possible future here. In terms of technology - good. But when we practically equip Lisa Alert units in different regions with thermal imagers, I’m not ready to say.
If three years ago you would ask me a question, what is there with drones, and I would say - “in general, complete bullshit. A very cool thing, we even have it, it flies perfectly, the grass is cut even better. It’s impossible to find people. ” And today, every week we find the missing drone. Beeline neuron begins to be useful.
Or, for example, it’s great to make repeaters on balloons. They allow you to make a giant communication zone. Moreover, for the guys [Stratonauts] the balloon can be illuminated from the inside and work at night in the forest as a lighthouse.
But six thousand rubles for the rise ...
In the season we receive 20 - 30 applications for searches per day in Russia. In the highest season we can receive more than 100 applications per day. Therefore, we continue to seek financing, we are waiting for ready-made technologies and ready-made systems.
- If you were offered to pick up one of the four technologies for yourself, whose solution would you choose?
- Choose one? I will not succeed, probably from greed. Everything is so interesting. No wonder they are in the final, it all can work. Through the semi-finals, only that which is able to find made its way. There, on a much smaller plot, they succeeded. Here is just a giant plot, but the full beauty of the picture of the forest - it is transparent.
For example, these guys are [MMS Rescue]. There were two teams in the semi-final - in the final, there was one. They combined, and as a result of the combination of technologies, something more balanced was obtained. Beacon technology reduces the number of kilometers traveled by feet. The reverse side is the beacons that are scattered [MMS Rescue teams] now, suggest a further hike through the forest with the collection. And this is a distance that does not reduce the amount of human labor. That is, the technology itself is okay, but you need to refine something, think out in tactics - how to scatter it correctly, so that it is easier to collect.
The Yakut guys have the most finished product. They are already looking for and are already finding the missing with his help. These are people of practice who came to the competition not from a technical laboratory, but from the forest.
“I like them.”
- They bribe a lot of people with their practical part, but the competition is not about that. You need to see a reasonable bright look into the future. Under the crowns, we still do not know how to look at anything. My dream is to look through the forest to see a person lying through the trees. No matter what it will be. The problem is that for us, the forest is still an opaque environment.
- At the start of the contest, you didn’t feel like “These techies do not understand anything in the search, their toys will not help anyone”?
- Then we would not have started this story at all, we would not have molested the Sistema fund with the fact that it is very important and necessary. We need to move forward. Searching today is still a stone age with rare flashes of something new. Unless we go with ordinary torches, but with LED. We are still not at the stage when people from Boston Dynamics walk through the woods, and we smoke at the edge of the forest and wait for them to bring us the missing grandmother. But if you do not move in this direction, do not move all scientific thought, nothing will happen. We need to excite the community - we need thinking people.
The main thing is that outside the framework of this competition the teams do not say: “Play and that's enough.” Further movement is needed. We all plunged into this story for a year. And if you imagine another year, then another and another, then in ten years we will see fundamentally different things that are applicable throughout the world, in any conditions.