Practices of the Future: How Schoolchildren Improved Drones, Growing Bacteria in New Environments, and Creating Assistants for AI Developers

Improving quadcopters and helping novice pilots

During the first days of the shift, as part of the track “Unmanned Transport and Logistics Systems”, students learned to program in Python, use indoor navigation systems, work with libraries for computer vision and use various sensors on drones. Then the project activities began – the teams were offered to work on cases of using quadcopters to perform autonomous tasks indoors. For example, Anton Babkin and Igor Fetisov were among the guys whose task was to protect drones from collisions and damage.

– Quadrocopters are very expensive and fragile, which makes it scary for beginners to work with them. Therefore, they learn more slowly than they could, – explains Anton Babkin. – There were two parts in our project. The first part is a program that prevented the quadrocopter from dropping from a great height, intercepting control from the pilot. The second part of the project is physical collision protection. By the end of the shift, we had a drone that was an order of magnitude more reliable and stronger than the standard one.

The group, which included Matvey Menshinin, worked on simplifying the control of the drone to help novice pilots.

– The essence of the DODGE project was that in modern realities there are a lot of obstacles for novice drone pilots. As part of the school, we were able to make an MVP for our project: we assembled Clover and wrote a program using machine vision. With the help of this program and the installed camera, the drone recognized the flying ball by color and flew away from it.

Investigating bacteria suitable for soil fertilization

Participants of the track “Microbiology and Molecular Biology” studied the microbiological and molecular biological characteristics of microbiomes using the example of the Moscow region. The students got acquainted with the methods of soil science, microbiology, molecular biology, with the genetics and ecology of microorganisms, population dynamics, and then started projects.

The team, which included Stanislava Rychka, studied the vital activity of bacteria of the genus Azotobacter. The guys identified the dependence of the dynamics of bacterial growth on soil acidity and tried to find species of the genus Azotobacter that live in acidic soils, including in the Moscow region.

– Articles on the Internet claimed that bacteria do not live in environments with acidity below 4, – says Stanislava Rychka. – We hypothesized that such species exist, and we got an incredible result! In an extremely acidic environment with an acidity of 1, bacteria grew, for a long time I could not believe that it was Azotobacter. On the same evening, we saw the bacteria in a microscope – this confirmed that the bacteria that grew up belong to the genus Azotobacter. And then the same was confirmed at the genetic level by the results of electrophoresis. Our hypothesis was confirmed. It was everyone’s delight! The found types of bacteria can be used as fertilizers instead of mineral fertilizers, the use of which is extremely irrational – it leads to soil degradation. It is very valuable for me that our project is aimed at solving the current problem.

We use machine learning technologies

The popular track Machine Learning had three laboratories at once: Analysis of Academic Literature by Means of Computational Linguistics, ML in Marketing and Web Analytics, and Development of Question-Answer Systems.

Roman Tikhomirov’s team decided to make a product that directly helps machine learning specialists. According to Roman, programmers working on a neural network can spend several hours looking for an error. The guys decided to create a system that will allow the user to instantly receive instructions on how to fix these problems. As a result, the team created the Aurora Assistant project – a question-and-answer system that helps IT specialists eliminate errors in the development of artificial intelligence.

The group, which included Viktoria Shchur, worked on its own ML-project – a web service with a nutrition recommendation system.

– Our system offers the user several dishes in the daily menu, based on his financial capabilities, body characteristics, the number of meals, culinary abilities. The user can choose the dish he likes and order ingredients from well-known supermarket chains by courier or self-pickup. During our work at the design school, we also managed to make an MVP.

Yaroslav Oleinik, together with the team, worked on an assistant to help a person communicate. The assistant is planned to be ported (made compatible, in other words) to any messenger or social network.

– The messenger will give the user various hints, for example, at what time it would be convenient for the interlocutor to respond to the user’s message. We managed to create a workable MVP of the project, which was a web application that played the role of a test messenger, and a system for analyzing emotions in the application imported into it.

Sofia Semyonova’s team wanted to make an art project using machine learning. One of the track’s experts showed them a neural interface that is usually used in meditation. But the guys understood: on its basis, they can bring those frames from films closer to reality when a person controls all the power of thought.

– The essence of the neural interface is simple: it reads waves from the brain and transmits them to a phone or computer. But why? If we say that there is a neural network capable of generating an image based on the word entered by the user, no one will be surprised. My team and I decided to improve this neuron so that it is based not only on the received word, but also on the mood of a person, which another neural network would determine by receiving data from the neural interface. As a result, we got a simple site, in the backend of which two neural networks are hidden: one analyzes emotions, the other creates an image based on them.

“Practices of the Future” – these are one-day and multi-day hackathons, design schools, competitions and challenges of the NTI Circle movement, where schoolchildren and students are completely immersed in work on a real problem, design future practices related to solving urgent problems with the help of new technologies. The “Practitioner of the Future” methodologies link schoolchildren, students, representatives of business, education, and science in a single space of joint work. In 2019-2021, Practices of the Future held more than 120 events, in which more than 18,500 people took part.

Source: Автономная некоммерческая организация "Редакция журнала «Наука и жизнь»" by

*The article has been translated based on the content of Автономная некоммерческая организация "Редакция журнала «Наука и жизнь»" by If there is any problem regarding the content, copyright, please leave a report below the article. We will try to process as quickly as possible to protect the rights of the author. Thank you very much!

*We just want readers to access information more quickly and easily with other multilingual content, instead of information only available in a certain language.

*We always respect the copyright of the content of the author and always include the original link of the source article.If the author disagrees, just leave the report below the article, the article will be edited or deleted at the request of the author. Thanks very much! Best regards!