CC BY
31Conclusions:
1. The movement of two-phase flows in the system of collection, separation and transport of gas was analyzed;
2. The results of gas-dynamic studies of gas-liquid mixture flows in various sections of pipelines are described on the basis of the structural form diagram in the joint movement of gas and liquid;
3. A technological scheme of a low-temperature gas separation unit is proposed using a separation heat exchanger and a pipe separator;
4. The advantages of the proposed scheme for the installation of low-temperature gas separation in the field preparation of gas for transport are shown.
References
1. Aliev E.U., Abdullaev E.A., Sultanov N.N. // Gas separation in pipelines. Ed. "Nafta-Press": Baku -2006, - 205 p.
2. .Mustafayev A.R., Abdullayev A.A., Panahov R.A., Sultanov N.N., Gadashova E.V. // Preparation of gases for transport, "Nafta-Press" : Baku,- 2015,- p.85-120.
COMPUTER TECHNOLOGY AND UNMANNED VEHICLES
Ivanko A.,
Professor, Moscow Polytechnic University
Ivanko M.,
Associate Professor of Moscow Polytechnic University, Ph.D.
Kolesnikova O., student of the Moscow Polytechnic University
Kulikova E.,
Associate Professor of Moscow Polytechnic University
Vinokur A.
Moscow Polytechnic University Professor
ABSTRACT
Innovative technologies, computer equipment make it possible today to put into practice artificial intelligence systems. There are control systems for airplanes and locomotives on railways, and at last real automatic control systems of modern cars began to appear. The era of high technology and automation of many activities, computers began to do a huge part of our work, both in everyday life and in the professional sphere. All this certainly simplifies our lives, so the developers do not stop and more often they surprise us with new projects.
To date, one of the most talked about new technologies is the development of unmanned vehicles. Many companies have started production of such cars, testing them on public roads. The article discusses the origins of the emergence of unmanned vehicles, analyzes the current state of the problem, assesses development prospects using information technologies.
Keywords: unmanned vehicles, IT-technologies, innovative projects, computer technologies, control automation.
An unmanned vehicle is a vehicle equipped with an automatic control system that can travel without human intervention.
The development of unmanned vehicles is accompanied by a number of ethical problems, including: moral, financial and criminal liability for accidents, decisions made by a car before a potentially fatal collision, problems of data protection and problems of losing jobs.
Consider the history of the creation of unmanned vehicles.
Many people may have a false opinion that the history of the development of unmanned vehicles originates in the XXI century. However, few people know that the first attempts to create a fully autonomous car were made in 1980. For example, if you go to the archive of The New York Times articles, then at the request of "unmanned vehicles", a large amount of materials 15 years ago will appear.
There are different data when the first fully autonomous vehicles appeared. The fact remains that initially
all such developments were created for military purposes. At the beginning of the 20th century, the first research began in the field of unmanned aerial vehicles. As early as 1916, Archibald Lowe created the first drone, a radio-controlled aircraft. During the First World War, air torpedoes and self-propelled German mines were already actively used.
However, until the middle of the 20th century, developments in the field of unmanned technology were experimental rather than practical, and, in one way or another, no single model could manage without direct human participation. Unmanned vehicles, like flying drones, were initially conventional remote control prototypes, and only gradually became autonomous.
The first experiments to create an unmanned vehicle date back to the early 1960s. In 1961, a student at Stanford University, James Adams, as part of his scientific work, created a prototype of a self-guided cart, better known as the "Stanford cart" (Fig. 1).
Picture 1.
The very first model was controlled by transmitting a signal through a cable. The second prototype of Adams has already made radio-controlled. In the 1970s, mathematician John McCarthy perfected the cart, equipping it with a vision system with which the device could partially autonomously move, orienting itself to a white line. The cart also had several cameras, a range finder and four channels for collecting information. Moreover, McCarthy attempted to create a three-dimensional mapping of the enviromnent.
In subsequent years, the main efforts of engineers were mainly focused on the development of a fully autonomous, rather than remotely controlled, controlled transport. At the initial stages, scientists from the USA, Japan and Germany achieved the greatest success. Thus, according to independent experts, the first fully autonomous vehicle was created by a group of German researchers led by robotics pioneer Ernst Dickmans in 1980 (Figure 2).
Figure 2.
For this project, Dickmans wrote several scientific papers that described in detail every detail of the mobile. Surprisingly, many technologies used more than 30 years ago, in fact, anticipated much of what is now used in modern UAVs. For the proper operation of their car, a group of German scientists applied the so-called Kalman filter, parallel computing mechanisms, and imitation of saccadic eye movements. In fact, this system was a machine learning model capable of adequately assessing the entire environment.
On the basis of the development of Dickmans from 1987 to 1995, the project "Prometheus" was in operation, aimed at improving unmanned vehicles. More than $ 1 billion was invested in Prometheus, which made it the most expensive project in the history of creating robotic vehicles in history. In 1994, the car "VAmP" Mercedes equipped with Dickmans technology for several hours independently traveled through the streets of Paris at a speed of up to 130 km / h, turned, overtook other cars and rebuilt from one lane to another (Fig.3).
Figure 3.
In the mid-1990s, a major impetus to the development of unmanned vehicles gave a breakthrough in the field of artificial intelligence, neural networks and machine learning.
In 2004, DARPA announced the first competition for fully autonomous vehicles. The cars were supposed to drive 230 kilometers without human intervention. Nobody got to the finish line then. Since then much lias
changed. In 2005, the race was the first winner (the Stanford team) and today all major automakers, startups and a significant number of IT companies are involved in the development of unmanned vehicles.
At the moment, there are already enough firms that have presented their models of unmanned vehicles on the market. For example, (Fig. 4) cars: • Renault-Nissan
A large-scale approach to the creation of unmanned vehicles applies the corporation Renault-Nissan. She plans to create 10 cars with autonomous driving functions until 2020. By 2018, they will be able to drive without a driver and change lanes, and by 2020, cross the intersections on their own to drive in traffic.
Figure 5.
Of course, unmanned vehicles are also being developed by Tesla (Figure 5). With the help of a software update, she introduced autopilot in her cars [2]. It still does not work everywhere and ignores traffic lights, but is able to go without a driver along a regular road. Ilon Mask promises that they will make the car completely unmanned within two years, by 2018.
Figure 6.
Alphabet unmanned vehicles (Figure 6) were the very first and to date have driven more than 2.2 million kilometers. Over the past 5 years of testing, there was information about only one accident caused by a car in February 2016 [3]].
It is not yet clear how and when the company is going to sell these cars to people.
Figure 7.
Presented by the corporation Mercedes - F 015 Luxury (Fig.7) is already used to ride enthusiastic journalists. Mass release, however, is expected only by 2030.
Figure 8.
In order to stand out from the background of other projects, Audi has made its car, the Audi RS7 (Fig. 8), high-speed and is testing it on race tracks. Already, he ideally passes the track and accelerates to 240 kilometers. Some even fear that this is the end for auto racing (4). The Audi A8 will be the first production car with an autopilot (but not fully autonomous), and it will be presented in 2017.
Also developed even unmanned trucks (Fig. 9) Mercedes-Benz Future Truck.
Are tested on the roads and trucks Mercedes. Judging by the logo, Mercedes expects to start selling them in 2025.
Figure 10.
Funds for the project of an unmanned Kamaz (Fig. 10) are allocated within the framework of the state program "Avtonet". Already there is a prototype.
Kamaz now expects to refine the unmanned model by 2022 and bring it to public roads by 2025-2027. In general, the development of unmanned vehicles in Russia will allocate 10 billion rubles in the next four years.
Figure 11.
The development team separated from Google in mid-2016 founded the Otto startup (Figure 11) aimed at creating unmanned trucks. Company executives are confident that the automation of trucks will bring the greatest economic benefits. The company does not plan to build its own trucks, but only to upgrade the existing ones for self-relocation.
Also developed buses and taxis.
Now let's look at the mechanism of the operation of an unmanned vehicle. How does he distinguish between objects? How to build a route? Stops to give way? Controls the speed of movement? And many other questions.
Figure 12.
The average person has everything you need to drive a car (Figure 12). We have eyes and ears to analyze the world around us; the brain that responds quickly to external processes; and, in most cases, a fairly reliable memory that allows us to drive quite confidently on a rather large number of roads. But there is a huge difference between how we perceive the world and how a computer does it.
"You can simply load the rules of the road into an on-board computer," says Katelin Jabari of the unmanned vehicle development team of Google, "but this will not allow for 99% of the things we encounter on the road. How abruptly should the car stop? How fast should she go into a turn? " Recreating these skills electronically is a possible, but not the easiest, task.
An autonomous car needs three basic skills for successful movement: first, it must understand where it is located; secondly, it must determine the safe route to the destination; thirdly, he must figure out exactly how he should go along this route. This is exactly what we do in the driving process: localization, awareness and movement. Fortunately, the third point is easy to implement with the help of technology, but there is a problem with the first two skills.
"We are often asked why we don't use GPS," said Will Maddern, a member of the Oxford research group that is building the Robot Car. "This is a very logical question: GPS is an amazing engineering achievement."
The main problem of GPS is that this technology has a number of significant limitations. Firstly, its use requires a "direct view of the sky", which excludes the possibility of its use in tunnels, closed parking lots and, even, in the forest. But, more importantly, GPS is not accurate enough to accomplish the task. The error of this technology fluctuates within meters, and when it
comes to driving along a busy street, every centimeter is important.
"We often use GPS as a point of reference in the localization algorithm," says Maddern, "but we do not rely on it when we determine in which lane the car is located." Therefore, unmanned vehicles rely on completely different technologies.
"It's necessary for two things to happen," explains Professor Sidhar Lakshmanan, an engineer who specializes in image processing and computer vision of unmanned vehicles at the University of Michigan. "The accuracy of the maps should increase significantly, and the process of registering objects should become more reliable."
Cars being developed at Oxford University, as well as the brainchild of Google, use 3D roadmaps, which project participants call prior maps. They look like the image above. Using on-board sensors, the car compares the current state of affairs, which it registers in real time with what is in its memory. "Basemaps allow the car to better" understand "where it is before starting processing real data," explains Jabbari. "Thus, he knows what should happen, analyzes what is actually happening, and can make a concrete decision based on a comparison of these data."
The need for such cards introduces some restrictions on the use of unmanned vehicles. "Before we can automate cars, we need to create fairly detailed road maps," admits Jabbari. Google and other developers create these cards bit by bit using the same cars that are used to test autonomous control systems, to fix the surrounding world with a very high level of detail.
Unmanned vehicles use rather exotic equipment. Google cars rely on lasers and radars. Laser Velodyne 64 (LIDAR) is mounted on top of the vehicle. You can see them in the photo below (Fig.13).
Figure 13.
LIDAR is capable of scanning 1.3 million points for each revolution around its axis. The system is able to recognize a 14-inch object at a distance of 160 feet. Systems that demonstrate two times lower accuracy are certainly less reliable, but in theory they should be sufficient to navigate in difficult weather conditions. Cars that are being developed at Oxford use stereo cameras and LIDAR. Others, such as the development of AutoNOMOS Labs, use a similar combination of sensors. Anyway, the main task of all this equipment is to collect as much information as possible about the car's surroundings.
There is quite a logical question: "If we have such exact means for analyzing the environment of a car in real time, why do we spend time on developing base maps?"
"Not all developers of unmanned vehicles need basic maps to navigate a vehicle," says Maddern. It is enough to follow road markings, signs and traffic lights, as well as taking into account the location and actions of other cars and road users. "
The Google team does not agree with this approach. Jabbari argues that "If you rely only on realtime data, you will have to perform a much larger number of operations to assess the situation. Basemaps make this process easier by increasing the level of security and control over the situation. " This does not mean that using only current data is impossible, it is just a more difficult task. Perhaps this will be the best approach, but it is unlikely to be the first to enter the market.
Even with sufficiently detailed base maps, the unmanned vehicle is constantly forced to gather information about the world, especially in difficult weather conditions. Maddern says that the cars that develop at Oxford are constantly creating new maps, although this approach requires significant amounts of memory.
How the car perceives the world around it depends on the equipment the research team uses. In the case of laser scanners, the equipment generates a picture of the surrounding world with millimeter accuracy, which is called a point cloud. This technology allows you to assess the situation with high accuracy, which facilitates the comparison of the data obtained with the base maps and ensures the high relevance of the data obtained.
The main problems of LIDAR are its high cost, high level of energy consumption and a large number of moving parts. Therefore, the cars that Oxford specialists are working on use ordinary cameras to compare the real situation with the information contained in the maps. On the other hand, the use of cameras reduces the accuracy of the analysis and complicates the comparison algorithm.
Instead of comparing "pixel to pixel," the computer identifies "points of interest," such as angles, borders, and other features. On both maps (base and obtained in real time), these points are combined into groups that are compared with each other. Maddern argues that the system works fine even with different image sizes and poor lighting.
By and large, this accuracy is enough. The gif below shows how Robot Car perceives the surrounding space with a monocular camera and compares the data with the base map. Reducing the cost of equipment is of paramount importance, it reduces the cost of equipment from thousands to hundreds of dollars. The main problems faced by researchers: blurring the image, flashes of light and difficult weather conditions. But the problem of high cost is considered one of the most difficult in the development of unmanned vehicles, so its solution is of paramount importance.
Object recognition. "This is the holy grail of unmanned vehicles," says Lakshmanan, "Registering an unexpected obstacle, be it a pedestrian crossing the road, or a car changing its location abruptly, and adapting the car to new conditions."
There are several ways to solve this problem. The simplest is similar to the way in which Google's image recognition algorithm distinguishes a teapot from a kitten. Actually, it is his Google that is used in his project: "We teach cars to identify categories of things. The location of pedestrians, cyclists and other vehicles, and of various kinds, for example, a police car or a school bus. " Based on the type of object, the car must determine its range of motion in order to predict further developments. If the car "sees" that the cyclist has raised his left hand, this means that he plans to turn left or change lanes. The car adjusts its behavior based on this information. "
Roughly speaking, cars are quite capable of understanding what is happening around.
Figure 14.
There have been enough cases of DDP using unmanned vehicles. Some of them are the result of flaws in the software, which are carefully corrected. But others could not be prevented a priori.
The next problem is how self-driving cars need to be programmed to act in emergency situations where
either passengers or other road users are at risk. A classic example of the moral dilemma faced by car manufacturers and software developers is the problem of a trolley, in which the conductor of a trolley should choose to leave it on its original route and knock 5 people down or move the trolley to an alternate route and knock down one person (Fig. 15).
Figure 15.
In this problem it is necessary to address two main questions. First, what moral basis should a self-driving car be used to make such decisions? Secondly, how should this logic be transmitted in computer code? Researchers propose the use of two ethical theories for programming the behavior of self-driving cars: deontology and utilitarianism. The three laws of Asimov's robotics are a typical example of deontological ethics. According to this theory, a self-driving car must strictly follow the prescribed rules in any situation. According to utilitarianism, every decision made by a car must strive to maximize the usefulness of such a decision. In this case, it is necessary to give a definition of utility, one of which may be the maximization of the number of saved human lives. According to researchers, self-
driving cars must operate on the basis of a combination of several theories in order to be able to make morally informed decisions in emergency situations.
How is the use and testing of unmanned vehicles developing? The following is a list of 2018 news on this topic.
"On January 10, 2018, the startup Voyage, which manages the unmanned vehicle fleet, announced the launch of robotaxi in The USA, the largest old-age village - The Villages (located north of Orlando, Florida)."
"At the end of January 2018, it became known that the telecommunications company Ericsson, in cooperation with Swedish public transport operators and a number of local partners, would test unmanned buses
on the streets of Stockholm. The goal of the project is to test how autonomous vehicles will work in real conditions, traveling side by side with ordinary cars, cyclists and pedestrians. "
"In an effort to accelerate the development and implementation of technologies for unmanned vehicles in order to catch up with the United States in this area, on April 12, 2018, the Chinese authorities promulgated national rules for testing self-driving cars.
According to the requirements, which come into force on May 1, 2018, vehicles must first be tested in closed, specially designated areas, and only then tests on roads are allowed, which must also be predetermined. "
"At the beginning of May 2018, Aptiv (formerly Delphi), a automotive component manufacturer, announced the launch of 30 self-driving cars on public roads. They began driving in Las Vegas as part of a vehicle sharing service (ridesharing) offered by Lyft. "
"Since August 27, 2018, the world's first unmanned taxi service lias been launched in Tokyo. The
service in the test mode was launched by the Japanese developer of autonomous transport technologies ZMP together with one of the country's leading taxi companies Hinomaru Kotsu, Nikkei reports. "
"On September 26, 2018, the government of the Australian state of New South Wales announced the launching of unmanned bus trips on the road around the Olympic Park in Sydney to transport passengers."
"On October 13, 2018 in Dubai, the service of unmanned taxis was launched. They began testing it in one of the districts of the city to transport passengers from one particular point to another - for example, from a shopping center to a cinema or to the house of a passenger. For the project used modified cars MercedesBenz E-Class. "
News about the use of unmanned cars in Russia: "On July 17, 2018, Yandex announced the signing of an agreement on the development of unmanned vehicles in the city with the Moscow authorities. The five-year contract with the possibility of extension was signed at the Moscow Urban Forum.
Figure 16.
"On September 26, 2018, the Monitoring Station opened in the Skolkovo Innovation Center (Figure 16) - a high-tech base for testing unmanned vehicles (BPTS). Testing will be conducted in conditions close to public roads. The station uses a promising 5G network. The first tests were of the second generation NAMI-KAMAZ buses 1221 of the Shatl project. "
"On November 26, 2018, Russian Prime Minister Dmitry Medvedev signed a decree on testing unmanned vehicles on public roads. The experiment will begin on December 1, 2018 in Moscow and Tatarstan and will last until March 1 of 2022. "
These quotes are presented from the articles "Unmanned vehicles on the world market" and "Unmanned vehicles in Russia" from the portal tadviser.ru.
Conclusion and perspectives. Most drivers are now owners of their own cars. Many see cars as a continuation of their own "I." But some experts predict the end of this attitude to cars, pointing to a decrease in the number of car owners among young city dwellers and a small number of young people who still spent time on obtaining a driver's license.
Psychologist Joseph Coughlin is sure: "We, most likely, will never be owners of unmanned vehicles." But this does not mean that the level and quality of services will be the same for different people. Wealthy people, thinks Coughlin, will issue a premium package, according to which every morning a luxury car will drive up to their house. Those who save every penny, on the contrary, will be able to buy a subscription for one month and will share a small car that stops around the corner with other passengers.
For investors, Uber and Lyft are a huge plus for unmanned vehicles - a significant reduction in costs, namely for the driver.
GM representatives believe that this innovation will also be beneficial to customers: the current cost of the trip - $ 1.5 per kilometer - will more than halve.
While engineers are solving technological problems, company representatives are carefully looking at the government. They worry that any law regarding any technology can quickly become outdated and, thus, hinder further innovation.
Despite technological advances, society is still wary of drones. According to the Deloitte survey, 81%
of South Koreans surveyed (81%), 79% of Japanese and 74% of US residents doubt the reliability and safety of unmobiles.
In Russia, 51% agreed with the statement: "I would like to have my unmanned vehicle." All over the
world, they are 42%. The least willing to acquire this type of transport - in Germany (24%). According to the survey results, only 15% of Russians said they would not use an unmanned vehicle (Fig. 17).
B. 2. Какое из приведённых ниже утверждений отражает ваше мнение?
• Я хочу поскорее начать пользоваться беспилотным автомобилем
• Я не уверен(а), но идея появления беспилотных автомобилей кажется мне интересной
• Я против появления беспилотных автомобилей, никогда не буду ими пользоваться
Figure 17.
Unmanned vehicles can be a revolution of vehicles and completely change the lives of people. How this technology will evolve will show time.
References
1. https://habr.com/post/395601/.
2. https://habr.com/post/395663/.
3. https://habr.com/post/393237/.
4. https://goo.gl/Atg4ad
5. https://goo.gl/69JnSr
6. https://goo.gl/4bmAL3
7. https://goo.gl/dHievU
Pile foundations iN THE CONSTRUCTED CONDITIONS OF THE CITY
Karpyuk V.M.,
Professor of the Department of concrete and masonry structures; Odessa State Academy of Building and Architecture, Odessa, Ukraine
Karpyuk I.A.,
Associate professor, candidate of engineering sciences, faculty of hydrotechnical and transport building;
Odessa State Academy of Building and Architecture, Odessa, Ukraine
Pancheva N. V. student;
Odessa State Academy of Building and Architecture, Odessa, Ukraine
ABSTRACT
This article highlights the problem of foundations in constrained urban environments, and also describes the results of laboratory tests in a pile model of piles, which are immersed by the method of driving and indentation.
Keywords: pilot, injection, carrying capacity, sediment, ground, model research, national experiment, interaction, calculation, coefficient.
In recent years, more and more often there is a need for the construction of residential and civil buildings adjacent to existing buildings and structures. In such conditions, the use of piles, immersed by driving or vibration, is dangerous, since, in this case, there are dynamic effects from which damage occurs in the structures of existing buildings. Thanks to the joint fruitful work of scientists and builders, a number of technologies and mechanisms have been developed in
the cities of St. Petersburg, Moscow, Kiev, etc., allowing plunging prismatic piles [1, 5, 7, 8]. This issue was addressed by such scientists as: Abelev M.Yu., Dalma-tov B.I., Gdalin S.V., Simagin V.G., Sotnikov S.N., Shvets V.B., Ginzburg L.K., Koval V.E., Feklin V.I. and etc.
The problem of construction in cramped urban conditions is becoming more and more acute in Odessa.
