<\/a><\/noindex><\/p>\nIn 2013, two little girls \u2014 Alina Ivanova and Ayana Vinokurova \u2014 went missing in the village of Sinsk in Yakutia. A massive effort was mounted to find them: a hundred volunteers, rescue teams, divers, and drones with thermal imagers were deployed. Footage from helicopters was made available online for everyone to view. But the efforts were not enough. What happened to the girls remains unknown to this day.<\/p>\n
Yakutia is immense. If it were a country, it would rank among the top ten largest by area. Yet, less than a million people live on this vast territory. In such an endless and desolate taiga, Nikolai Nakhodkin worked for 12 years in the Yakutia Rescue Service, 9 of which he served as a leader. When conditions are as tough as they can be and resources are scarce, new ways of searching for people have to be devised. And as Nikolai states, ideas come not from a good life. <\/p>\n
Nikolai Nakhodkin<\/i><\/p>\n
Since 2010, the Yakutia Rescue Service has been using drones. It is a separate organization from the Russian Emergencies Ministry, funded by the republic itself. There are no strict regulations regarding equipment, so the Emergencies Ministry began using drones much later. Additionally, there is a scientific group within the service where enthusiastic engineers work on developing applied technologies for rescuers. <\/p>\n
\"The existing methods of search used by the Ministry of Emergency Situations, rescue services, and various enforcement agencies have not changed since the 1930s. A tracker follows the trail, and a dog helps not to lose it,\" says Alexander Aitov, who led the research group. \"If a person is not found, an entire village, two, or three in Yakutia mobilizes. Everyone comes together and combs the forests. Every hour is crucial for finding a living person, and time passes quickly. There is never enough time. When the tragedy occurred in Sinsk, many people and equipment were involved, but without results. Similar situations arise in searches in the uninhabited taiga. To address this, the idea emerged to not perceive the missing person as a passive element but to leverage their own will to survive and active desire to live.\"<\/p>\n
\"Rescue engineers decided to create rescue sound-light beacons \u2014 fairly large but lightweight devices that emit loud sounds and shine for a long period, attracting attention both day and night. When a lost person approaches them, they will find water, crackers, and matches \u2014 along with instructions to stay put and wait for rescuers.\"<\/p>\n
\"These beacons are placed three kilometers apart from each other and encircle the approximate search area for the missing person. They emit low-frequency sounds, resembling a rumbling truck \u2014 because higher frequencies propagate poorly in the forest. Often, those rescued thought they were heading towards the sound of a road or a departing group of tourists.\" <\/p>\n
\"For example, they developed a floating suit for rescuers. The pants and jacket look like ordinary workwear, but in water, they keep a person afloat. To speak completely practically, the suit is dual-layered. Polyurethane foam pellets are sewn inside. They also have a design for divers working in low temperatures. When compressed air expands in the cold, the valves get covered in frost, and the person suffocates. Several institutions could not figure out what to do about it \u2014 they developed special materials, created electric heating, and implemented various modern approaches.\" <\/p>\n
Our team solved the problem for 500 rubles. The cold air coming from the cylinder (which can withstand down to -57) was passed through a coil running through a Chinese thermos. The air gets heated, people go underwater, and they can work there. <\/p>\n
However, the beacons were too simple, lacking many useful features. During search operations, rescuers often had to run great distances to check every beacon. If there are ten beacons, the rescuer must cover 30 km in the taiga every 3-4 hours. <\/p>\n
In 2018, the 'System' charity foundation launched the 'Odyssey' project \u2014 a competition for teams to find new ways to rescue missing people in the wild using new technologies. Nikolai Nakhodkin and Alexander Aitov decided to participate with their friends \u2014 they named their team 'Nakhodka' and brought their simple device to improve it in competition with others.<\/p>\n
\n
One of the less obvious solutions \u2014 or fantastical ones \u2014 is a dirigible equipped with a bio-radiolocation system. However, the team did not have a prototype and limited themselves to presenting their idea,\u201d says competition expert Maxim Chizhov.<\/p>\n
Another team decided to use a seismic sensor \u2014 a device that can recognize human footsteps among ground vibrations and show the direction they are coming from. With the help of the prototype, they even managed to find an extra who played the 'lost person' (as they affectionately called them), but the team did not progress far in the competition. <\/p>\n
By June 2019, after several training trials in the forests of the Leningrad, Moscow, and Kaluga regions, 19 top teams advanced to the semifinals. Their task was to find two actors in less than 2 hours over an area of 4 square kilometers. One was moving through the forest, while the other was lying still. Each team had two attempts to locate the person. <\/p>\n
\"Among the semifinalists, one team wanted to create a swarm of drones that would fly under the tree canopies, controlled by artificial intelligence, determining the direction of movement, maneuvering around trunks, and avoiding branches and twigs. Using AI, it would analyze the surrounding environment and identify the person. <\/p>\n
The team \"ALB-Search\" was close to success. They had a speaker on board that connected to the radio, a microphone capable of listening to the surrounding space, a camera, and a computer with AI and a trained neural network that processed images from the camera in real time, where a person might appear. <\/p>\n
\"The operator could analyze not thousands of images, which is physically impossible, but dozens or even a few, and then decide whether to change the drone's route, if an additional drone was needed for reconnaissance, or if the search group should be sent immediately.\"<\/p>\n
However, most teams faced similar problems \u2014 the technologies proved unsuitable for real forest conditions.<\/p>\n
\n
Thermal imagers, which about a third of the teams hoped for, also proved ineffective. In summer \u2014 when most people go missing \u2014 the foliage heats up so much that it becomes a continuous hot spot. It's easier to conduct searches in a small window at night, but there are still many heat spots \u2014 heated stumps, animals, and much more. A camera could help verify suspicious locations, but it's of little use at night. <\/p>\n
Moreover, it turned out to be difficult to obtain thermal imagers. 'Unfortunately, due to restrictions imposed on us by the EU and other countries, good thermal imagers are unavailable in Russia,' said Alexey Grishaev from the 'Vershina' team, which relied on this technology.<\/p>\n
'The thermal imagers available on the market have a digital output at a frequency of 5-6 frames per second and an additional analog video output with a high frame rate but low image quality. In the end, we found a very good Chinese thermal imager. You could say we were lucky \u2014 there was only one like it in Moscow. But it displayed the image on a small monitor where nothing could be seen.<\/p>\n
Most teams used the video output. Our team managed to improve the model and get a quality digital image at a frequency of 30 frames per second. The result was a thermal imager of a very serious level. Probably only military models could be better.' <\/p>\n
But even these problems were just the beginning. In the short time that the UAV flew over the search area, the cameras and thermal imagers captured tens of thousands of images. Transmitting them on the go was impossible \u2014 there was no internet or cellular network over the forest. Therefore, the drone returned to the point, and its recordings were unloaded, taking at least half an hour, and in the end, they had so much material that it was physically impossible to review it in just a few hours. In this situation, the 'Vershina' team utilized a special algorithm that highlighted images where thermal anomalies were detected. This reduced data processing time. <\/p>\n
'We saw that not all teams that came for the qualification trials understood what a forest is. That radio signals behave differently in a forest and dissipate quite quickly,' announced Maxim Chizhov at a press conference. 'We observed the surprise of the teams when communication was lost already at a distance of one and a half kilometers from the starting point. For some, the absence of internet over the forest was a surprise. But this is reality. This is the forest where people get lost.' <\/p>\n
The technology based on light and sound beacons performed well. Four teams reached the finals, three of which relied on this solution. Among them was 'Nakhodka' from Yakutia.<\/p>\n
\n
By the semifinals, there were only three people on the team who walked through the forest and set up beacons in the search area. And while many were solving engineering tasks, Na\u0445\u043e\u0434\u043a\u0430 operated like rescuers. \u201cWe need to apply a rescue mindset, rather than an agricultural one, when simply covering an area. We must act like a rescuer, putting ourselves in the place of the missing person, considering the approximate direction they might take and the paths they could follow.\u201d<\/p>\n
But the beacons of Na\u0445\u043e\u0434\u043a\u0430 by this point were no longer as simple as they were a few years ago in Yakutia. With the help of grants from \u2018Sistem\u044b\u2019, the engineers developed a radio communication technology. Now, upon finding a beacon, a person presses a button, and rescuers immediately receive a signal, knowing exactly which beacon the missing person will be waiting at. Drones are not needed for searching but to lift a radio signal relay into the air and increase the range of the activation signal transmission from the beacons.<\/p>\n
Another two teams developed entire search complexes based on sound beacons. For example, the M\u041cS Rescue team created a network of portable beacons, where each beacon serves as a relay, allowing the activation signal to be transmitted even in the absence of direct radio communication with the search headquarters. <\/p>\n
\u201cWe have a group of enthusiasts tackling such a task for the first time,\u201d they say. \u201cWe have worked in other fields \u2014 technology, IT, and we have specialists from the aerospace sector. We gathered, brainstormed, and decided to create this solution. The main criteria were low cost and ease of use, so that people could take it and apply it without training.\u201d<\/p>\n
Another team, the \u2018Stratonauts\u2019, managed to find an actor faster than anyone else with a similar solution. They developed a special application that tracked the position of the drone, the locations of the beacons, and the positions of all the rescuers. The drone delivering the beacons also served as a relay for the entire system, ensuring that the signal from the beacons was not lost in the forest.<\/p>\n
\u201cIt was not easy. One day, we got completely soaked. Two of our team ventured into the forest through the underbrush, and they realized it was far from a picnic outing. But they returned tired yet satisfied \u2014 after all, we found the person in both attempts in just 45 minutes,\u201d says Stanislav Yurchenko from \u2018Stratonauts.\u2019 <\/p>\n
\u201cWe used drones to deploy beacons in the center of the area to ensure maximum coverage. In one flight, a drone can carry one beacon. It takes a long time \u2014 but is faster than a person. We used small, compact DJI Mavic drones \u2014 each carrying a single beacon. This is the maximum weight it can carry, but it keeps costs low. Of course, we would like to find a fully autonomous solution. With AI, so the drone can scan the forest and determine drop points. Right now, we have an operator, and after a kilometer, without additional devices, communication is lost. So in the next phase, we will brainstorm solutions.\u201d<\/p>\n
But no team found the immobilized person, and most importantly, they did not figure out how to do it. Theoretically, the chances of finding him remained only for the \u2018Vertex\u2019 team, which, despite all difficulties, managed to locate the person and reach the final using a thermal imaging camera.<\/p>\n
\n
The advantage of such devices is that they do not fall when the engine shuts off; they continue gliding and land on a parachute. The downside is that they are not as maneuverable as quadcopters. <\/p>\n
The main drone, \"Vershina,\" is equipped with an upgraded thermal imager and a high-resolution camera, while the second drone only has a camera. On board the main UAV is a microcomputer that uses specially developed software to autonomously detect thermal anomalies and sends coordinates with detailed images from both cameras. \"This way, we don\u2019t have to manually review all the material, which amounts to about 12,000 shots per hour of flight time.\"<\/p>\n
However, the aircraft technology was developed recently, and there were still many problems \u2014 with the launch system, with the parachute, and with the autopilot. \"We were afraid to take it for testing \u2014 it could simply crash. We wanted to avoid technical complications. So we opted for a classic solution \u2014 DJI Matrice 600 Pro.\"<\/p>\n
Despite all the difficulties that led many to abandon cameras and thermal imagers, Vershina managed to find a statistician. This required extensive work, firstly with the thermal imager and secondly with the search methods themselves.<\/p>\n
Over the course of three months, the team tested the technology that enabled the thermal imager to view the ground between canopies. \"There was a bit of luck, because the route of the statistics involved traversed forests where no thermal imager would see anything. And if a person got tired and sat down somewhere under a fir tree \u2014 it would be impossible to find them.
According to Alexey, using several drones in conjunction is very expensive (one carrier with a technical solution for searching on board costs over 2 million rubles), but in the end, it will be necessary. He believes this gives a chance to detect a stationary statistician. \"We initially wanted to find a person lying down. We thought that we could find a moving one anyway. And teams with beacons were only searching for moving individuals.\"<\/p>\n
\n
He pondered. It seemed to me that all the other teams spoke with smiles and sparkles in their eyes about solving engineering problems. The guys from MMS Rescue joked that a dropped beacon could fall right onto a person lying down. The 'Stratonauts' admitted that it's an extremely complex super task, for which there are currently no ideas. And the rescuer from 'Find' spoke with what seemed to me a mixture of sadness and hope:<\/p>\n
\u2014 A girl went missing in the taiga, at three and a half years old. She spent twelve days there, and searches conducted by a large number of people lasted for ten days. When she was found, she was lying in the grass, almost invisible from the top. They only found her by combing through the area.<\/p>\n
If there had been beacons set up... at three and a half years old, a child is already quite aware. And she might have approached it and pressed the button. I believe that some lives would have been saved.<\/p>\n
\u2014 And was she rescued?<\/p>\n
\u2014 Yes, she was.<\/p>\n
\nIn the fall, the remaining four teams will go to the Vologda region, and the task before them will become several times more difficult\u2014finding a person in a radius of 10 kilometers. That is, over an area of more than 300 square kilometers. In conditions where a drone has half an hour of flight time, vision gets obstructed by the treetops, and communication disappears after just one kilometer. As Maxim Chizhov says, so far no prototype is ready for such conditions, although he believes everyone has a chance. Grigory Sergeev, the head of the search and rescue squad 'Liza Alert', adds: <\/p>\n
\u2018Today we are already ready to use a couple of technologies we saw, and it will be effective. And I urge all participants and non-participants\u2014guys, test the technologies! Come to us for searches! Then it will not be a secret to anyone that the forest is opaque for radio signals and that thermal imaging cannot see through the treetops. My main dream is to find more people with fewer resources.\u2019habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"\u041a\u0430\u0436\u0434\u044b\u0439 \u0433\u043e\u0434 \u0441\u043f\u0430\u0441\u0430\u0442\u0435\u043b\u0438 \u0438\u0449\u0443\u0442 \u0434\u0435\u0441\u044f\u0442\u043a\u0438 \u0442\u044b\u0441\u044f\u0447 \u043f\u0440\u043e\u043f\u0430\u0432\u0448\u0438\u0445 \u0432 \u0434\u0438\u043a\u0438\u0445 \u0443\u0441\u043b\u043e\u0432\u0438\u044f\u0445 \u043b\u044e\u0434\u0435\u0439. \u0418\u0437 \u0433\u043e\u0440\u043e\u0434\u043e\u0432 \u043d\u0430\u0448\u0430 \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0447\u0435\u0441\u043a\u0430\u044f \u043c\u043e\u0449\u044c \u043a\u0430\u0436\u0435\u0442\u0441\u044f \u043d\u0430\u0441\u0442\u043e\u043b\u044c\u043a\u043e \u043e\u0433\u0440\u043e\u043c\u043d\u043e\u0439, \u0447\u0442\u043e \u0435\u0439 \u043f\u043e \u043f\u043b\u0435\u0447\u0443 \u043b\u044e\u0431\u0430\u044f \u0437\u0430\u0434\u0430\u0447\u0430. \u0412\u0440\u043e\u0434\u0435, \u0432\u043e\u0437\u044c\u043c\u0438 \u0434\u0435\u0441\u044f\u0442\u043e\u043a \u0434\u0440\u043e\u043d\u043e\u0432, \u043f\u043e\u0432\u0435\u0441\u044c \u043d\u0430 \u043a\u0430\u0436\u0434\u044b\u0439 \u043f\u043e \u043a\u0430\u043c\u0435\u0440\u0435 \u0438 \u0442\u0435\u043f\u043b\u043e\u0432\u0438\u0437\u043e\u0440\u0443, \u043f\u0440\u0438\u043a\u0440\u0443\u0442\u0438 \u043d\u0435\u0439\u0440\u043e\u0441\u0435\u0442\u043a\u0443 \u0438 \u0432\u0441\u0435 \u2014 \u043d\u0430\u0439\u0434\u0435\u0442 \u043a\u043e\u0433\u043e \u0443\u0433\u043e\u0434\u043d\u043e \u0437\u0430 15 \u043c\u0438\u043d\u0443\u0442. \u041d\u043e \u044d\u0442\u043e \u0441\u043e\u0432\u0441\u0435\u043c \u043d\u0435 \u0442\u0430\u043a. \u0414\u043e \u0441\u0438\u0445 \u043f\u043e\u0440 […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-35805","post","type-post","status-publish","format-standard","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t
![\"Engineers](\"\/wp-content\/uploads\/d8f939d4057610c75ab470ea570af535.jpeg\")