A few days ago, the Odyssey competition ended, in which engineering teams were looking for the best technology to find people missing in the forest. In the summer I talked about and posted it yesterday .
The organizers set a colossally difficult task - to find two people on the territory of 314 km2 in 10 hours. The ideas were different, but (spoiler) no one managed. One of the technical experts of the competition was Nikita Kalinovsky. I discussed with him the participants, their decisions, and also asked what other ideas were remembered throughout all stages of the competition.
If you've already read the coverage of the finale, you'll find some of the lines here as well. This is just a complete interview with minimal editing.
If you have not read more than one article from this series, I will briefly retell the context.
In previous episodesThe AFK Sistema Foundation launched the Odyssey competition to find ways to introduce modern technologies in the search for people who are lost in the wild without means of communication. Out of 130 teams, four teams reached the final - only they were able to find people in a 4 km2 forest twice in a row.
The Nakhodka team, founded by veterans of the Yakut Rescue Service. These are search engines with vast experience in real forest conditions, but perhaps the least advanced team in terms of technology. Their solution is a large sound beacon, which, with the help of a special signal configuration, is clearly audible at a distance of up to one and a half kilometers. A man comes to the sound and sends a signal from the lighthouse to rescuers. The chip is not so much in the technology as in the tactics of its use. Search engineers use a minimum of beacons to protect the search perimeter and, gradually narrowing it down, to find a person.
The Vershina team is the exact opposite of Nakhodka. Engineers rely entirely on technology and do not use ground forces at all. Their solution is drones equipped with customized thermal imagers, cameras and loudspeakers. The search among the footage is also carried out by algorithms, not by people. Despite the skepticism of many experts about the uselessness of thermal imagers and the low level of algorithms, Vershina found people several times both in the semi-final and in the final (but not the right ones).
Stratonauts and MMS Rescue are two teams that use a whole range of solutions. Sound beacons, area communication balloons, photography drones and real-time search trackers. The Stratonauts were the best in the semi-finals because they found the missing people the fastest.
Sound beacons have become the most effective and common solution, but they can only find a person who is able to move. A lying person has almost no chance. It seems that it is best to look for it with a thermal imager, but the thermal imager does not see anything through the crowns, and also hardly distinguishes heat spots from people from all other objects in the forest. Photography, algorithms and neural networks are promising technologies, but so far they are not performing well. There were exotic technologies too, but each of them had more limitations than advantages.
What do you do outside of competition?
β Group of companies INTEC, Tomsk. The main area is industrial design, development of electronics and software, including embedded. We have our own small pilot and small-scale production, we help bring the product from idea to mass production. One of our most famous projects is the NIMB project, which we have been developing since 2015. In 2018, we received the Red Dot Design Award with this project. This is one of the most prestigious awards in the world of industrial design.
- What does this thing do?
- This is a security ring, an alarm button that the user presses when an alarm event occurs. Looks like a regular finger ring. On its lower part there is a button, inside there is a Bluetooth module for communication with a smartphone, a microelectric motor for tactile indication, a battery, a three-color LED. The base is a combined flexible-rigid board. The main part of the body is metal, the cover is plastic. This is a well known project. In 2017, they raised about $350 on Kickstarter.
β How are you here? Teams live up to expectations?
- In some teams, people have extensive experience in searching, they have been in the forest more than once, they have held such events more than once. They understand well how to find a person in real conditions, but they understand very little in technology. In other teams, the guys are very well versed in technology, but they have absolutely no idea how to move through the forest, in summer, winter, autumn conditions.
Is there no golden mean?
- I have never seen it yet. The general opinion of all experts is this: if all teams are combined, driven into a single collaboration, forced to combine solutions, take the best from each and implement it, then you get a very cool complex. Naturally, it needs to be finished, brought to a sane product state, brought to the final presentation. However, it will be a very cool solution that can really be used and that will really save people's lives.
But individually, each of the solutions is not fully effective. Somewhere there is not enough all-weather capability, somewhere there is not enough round the clock, someone is not looking for people who are unconscious. You always need to approach comprehensively and most importantly, you always need to understand that there is a certain theory of people search and the complex must correspond to this theory.
Now the decisions are raw. Here you can see two classes of projects: the first is very simple and very reliable systems that work. Those sound signal beacons that the guys from Yakutia brought, the Nakhodka team is a unique device. It is evident that it was made by people with a lot of experience. Technically, it is very simple, it is an ordinary pneumatic signal with a LoRaWAN module and a MESH network deployed on it.
What is so unique about her?
- You can hear it for one and a half kilometers in the forest. For many others, this effect is not observed, although the volume level is approximately the same for everyone. But the right frequency and pneumatic signal configuration give such results. I personally recorded the sound at a distance of about 1200 meters, with a very good understanding that this is really the sound of a signal and the direction to it. In real life, this thing works great.
βAt the same time, it looks the most non-technological.
β It really is. They are made from a piece of PVC pipe and are the simplest, most reliable and very effective solution. But with its own limitations. We cannot take and use these devices to find a person who is unconscious.
β The second class of projects?
- The second class is complex technical solutions that implement various specific search models - search using thermal imagers, combining thermal and three-color images, drones, and so on.
But, everything is very raw. Neural networks are used in some places. They are deployed on personal computers, on nvidia jetson boards, on the aircraft themselves. But all this is still incomplete. And as practice has shown, the use of linear algorithms in these conditions worked much more efficiently than neural networks. That is, the definition of a person by a spot on the image from a thermal imager, using linear algorithms, by the area and shape of the object, gave a much greater effect. The neural network did not find practically anything.
Because there was nothing to teach her?
- They claimed that they taught, but the results were extremely controversial. Not even that controversial - there were almost none. Neural networks did not show themselves here. There is a suspicion that they were either taught incorrectly or taught the wrong thing. If you apply neural networks correctly in these conditions, then most likely they will give a good result, but you need to understand the entire search methodology.
β They say that neural networks are promising. If they are done well, they will work. About the thermal imager, on the contrary, they say that it is useless in any case.
βNevertheless, the fact was recorded. The thermal imager is really looking for people. As in the case of neural networks, you need to understand that we are talking about tools. If we take a microscope, then to examine small objects. If we hammer a nail, then it is better not to use a microscope. With a thermal imager and neural networks, everything is the same. A properly tuned, correctly applied tool in the right conditions gives a good result. If we apply the tool in the wrong place and in the wrong way, it is natural that we will not get the result.
- Well, how to use a thermal imager if they say here that even a rotting stump gives more heat than a missing grandmother?
- Not more. Checked, looked - no more. The person has a clear pattern. You need to understand that a person is a very specific object. Moreover, at different times of the year these are different objects. If we are talking about summer, then this is a person in a light T-shirt or a T-shirt or a shirt that glows with a powerful spot on the thermal imager. If we are talking about autumn, about winter, then we see a head covered with a hood with a remnant of a thermal trace that is knocked out from under the hood or from under the cap, luminous hands - everything else is hidden by clothes.
Therefore, a person can be seen in a thermal imager, clearly visible, I myself saw it with my own eyes. Another thing is that wild boars, elks, bears are just as clearly visible, and we need to filter very clearly, which we still observe. You definitely wonβt get by with a thermal imager alone, you canβt just take it, show it to a thermal imager and say that it will solve all our problems. No, there must be a complex. The complex should include a three-color camera that gives a full-color image or monochrome with LED illumination. It must be with something else extra, because the thermal imager itself just gives spots.
- Of those teams that are now in the final, who is the coolest?
No favorites, to be honest. I can throw a solid brick at everyone. Let's just say I really liked the decision of the first Vershina team. They just had a thermal imager plus a three-color camera. I liked the ideology. The guys search by technical means without involving ground forces, they didnβt have mobile crews at all, they searched only with drones, while they found people. I will not say whether they found the one they needed or not, but they found people and found animals. If we compare the coordinates of the object on the thermal imager and the object on the three-color camera, then we will be able to identify the object and determine whether there is a person there.
I have questions about the implementation, the synchronization of the thermal imager and the camera was done carelessly, it was practically not there at all. Ideally, the system should have a stereo pair, one monochrome camera, one tricolor camera, and a thermal imager, and they all work in the same time system. Here it was not. The camera worked in a separate system, the thermal imager in a separate one, and they had artifacts because of this. If the speed of the drone were a little higher, this would already give very strong distortions.
β Did they fly a copter or was there a plane?
βNo one here had a copter. Rather, the copters were launched by one of the teams, but this was a purely technical function to ensure communication in the search area. A Lorov repeater was hung on them, and it provided communication within a radius of 5 kilometers.
As a result, all search aircraft here are of the aircraft type. This entails its own problems, because taking off and landing is not easy. For example, yesterday the weather conditions prevented the Nakhodka team from launching their drone. But I would say this - the drone that they had in service would not have helped them in the form in which it is configured now.
- In the semi-finals, they wanted to use the drone only for relaying.
- The drone at Nakhodka was made for photo-video filming and notification. There is an alarm, a thermal imaging camera and a color camera. At least that's what I've heard from them. They didn't even unpack it yesterday. As they brought it, it was packed. But even if they got it, most likely they would not use it. They have a completely different tactic - they searched with their feet.
Today, the guys want to sow the forest with beacons and find a person with their help. This is my least favorite solution. I have big doubts that they will then collect those 350 lighthouses that they brought here. Or rather, how, we will force them to collect, but not the fact that they will collect everything. Most of all I liked the decision of the first team, because it involves the complete abandonment of ground forces.
- Only because of this? After all, if you really take such a huge area in quantity, it might work.
- It will most likely work, but I didnβt like the drop configuration, nor the configuration of the beacons themselves.
- There was a brick for the Stratonauts.
β The Stratonauts have a great solution. If they had done it the way they wanted to do it, they would have succeeded. But they also had problems with flying machines.
They have a system for providing search groups. The main emphasis is on mobile ground forces. They are issued with beacons, communicate with groups and communicate with ground beacons to deploy search groups in the right places and in the right directions. They have balloons with repeaters that provide communication over the area. They have ground stationary beacons, but there are very few of them, and they themselves admit that they made them at the last moment, and this is not their main tactical unit - they made them for the sake of testing. There are quite a few of them and they did not make a special contribution to tactics.
The main tactic was that each searcher in the group has its own personal tracker, which is combined into a single information network along with the headquarters. They can clearly see who is where. Combing in real time, the direction is corrected.
- Everything looks like it really wants to be combined into one.
β Yes, absolutely. Grigory Sergeev and I walked, he looks and says - βDamn, what a cool thing, I would like itβ, we come to others - βDamn, what a cool thing, I would like it hereβ, we come to the third - βDamn, what a cool thing , I would have found a person there and there.
Separately, they are sectorally good solutions for certain conditions. If you combine them, then you get a very good complex that has a single communication field, there is a system deployment over a long range with the help of balloons, there is a system for tracking and controlling ground forces in real time, there are beacons that strike at a fairly long range and can proper use and dissection of the search area into sectors to provide signaling to a person so that he goes to them, and then everything turns into a matter of technology. There is flying weather - some forces are used, there is no flying weather - others, night - still others.
But it's all catastrophically expensive.
Some are expensive, some are not.
β For example, one drone, which is now taking off, probably costs as much as a Boeing.
Yes, they are quite expensive. But you need to understand that if used correctly, this is a one-time purchase. You need to buy once, and then just carry it around the country and apply it. Such a one-time investment in capable hands will last a long time if properly maintained and operated.
- When you looked at the applications for the competition, was there something that you liked, but did not make it to the final?
- There were a lot of funny things.
- What is the funniest thing you remember?
β Very memorable bioradar suspended on a balloon. I laughed for a long time.
It's even scary to ask what it is.
- The trick is that this is a really good way to determine. Bioradar is aimed at determining exactly biological living objects against the background of everything else that is reflected. Chest vibration, pulse is usually used. For this, very high-frequency radars under 100 GHz are used, they shine at a fairly good distance and shine through the forest at a depth of 150 by 200 meters.
Why is it funny then?
- Because this thing only works when installed permanently, and they wanted to hang it on a balloon. And they say - "this is a stationary object." Here we are now looking at a balloon, it is constantly sausage, and they want to hang a thing on it, which must be tightly screwed to the ground, otherwise the picture will be such that nothing will be clear on it at all.
Another very funny were cardboard unmanned aerial vehicles.
β Cardboard?
β Yes, cardboard drones. It was very funny. Aircraft glued from cardboard and painted with varnish. He flew just as God puts on his soul. The guys wanted him to fly in one direction, but he flew anywhere, just not in the right direction, and eventually crashed, saving himself from torment.
Very funny was "a flying bagel that can be reconfigured into a flying sausage" - a real quote from the application. The outer braid of the fire hose is taken, the rubber is removed, inflated and becomes a long pipe twisted on both sides. They tie him up, and it turns out a flying bagel, on which they hang a camera. And that a bagel can be easily transformed into a flying sausage - everyone was laughing over the sausage. Why, why the sausage is not clear, but it was very funny.
- I heard about cubes that are placed on the ground, and they read vibrations and steps.
Yes, indeed, there were such things. You need to understand that the thing is actually quite working. I know of several commercial products that do just that. This is a security tuned seismograph for perimeter security systems. But this thing is used exclusively for critical infrastructure and military installations. I know that gas pumping stations have three-level access control systems, the first of which is just seismographs.
- Sounds like a good idea. Why not then?
β The fact is that it is one thing to protect a closed perimeter of a critical infrastructure facility with a small area, and another thing is to sow the entire forest with these seismographs. Their range is very short, and most importantly, you can hardly distinguish between a running boar, a running man and a running bear. Theoretically, of course, if you turn on the materiel correctly, but this greatly complicates the technique, there are much simpler methods, it seems to me.
Everyone was recommended to go to the quarterfinals, everyone was recommended to try their hand. Those whom we see here are those who actually managed to find people. All other people were not found, so the competition, it seems to me, is quite objective. You can, for example, trust the opinion of experts, you can not trust, but the fact remains - found or not found.
Source: habr.com