Intellectual transport systems - problems on the way of introduction in Russia
More than a year has passed since my last publication on ITS . It's time to continue the interrupted conversation. Today we will touch on the topic of correct understanding of the subject of ITS. The problem of understanding the subject is inextricably linked with the understanding of the role and place of automation in general, not only in the field of transport. We dabble around the definition of ETS, from which rather curious logical connectives will stretch. We will touch on the topic of traffic flow modeling and the conceptual role of ITS in traffic management. Fiction turned out not entertaining. But as on Habré (and indeed in RuNet) silence reigns on this topic (except for the occasional opus of any “humanities”), it is necessary to bear the heavy burden of enlightenment and accurately state the basic things.
In matters of creating intelligent transport systems, we have not yet filled up enough cones - this area is relatively new, there are practically no specialists of ours, the implementations are rare, and even those can be considered bold experiments at public expense, and not something of practical value. Even the term "ITS" mean that it is not clear what. The official interpretation does not exist, despite the fact that in the field of ITS there is already the first standard GOST R ISO 14813-1 “Recommended architecture model for the ITS sector”. That is, there is a standard, but there is no definition of ITS in it.
Russian Wikipedia translates the definition of ITS from English, which we will use (I apologize for bold italics, but the definition will not be lost in an array of text):
')
ITS is an intelligent system that uses innovative developments in the modeling of transport systems and the regulation of traffic flows, providing end users with greater information and safety, as well as qualitatively increasing the level of interaction of traffic participants in comparison with conventional transport systems. ( Link to article )
Intelligent transport systems are the point of contact between the automotive industry and the information technology industry and are based on two “pillars” - modeling transport systems and regulating traffic flows .
The definition of ITS gives us an idea of ​​the main objectives:
The above definition contains everything necessary for a proper understanding of the issue. The only thing that prevents us from understanding it correctly and doing the right thing is our traditional perception. Please take this thought seriously: we have everything we need for business, except for the right way of thinking ! In this context, the " right " way of thinking means a way of thinking sufficient to understand the Western approach to the subject and to use the available tools for solving problems, nothing more. For universal truths we are not going to chase.
The Western engineer thinks with functions, he is primarily focused on what the system should do. In our thinking, an object view of the world is “wired up”, real objects are important to us, that is, we think first of all about how the system will work. This distinction is not as subtle as it may seem at first glance.
I will give an example. The word "server" for a Western engineer means something that provides services, service. That is, the function. For our engineer, the “server” is first of all an iron box with light bulbs, that is, an object. To make sense, we have to use a variety of crutches: "server application", "mail server", "server queues", etc. And all the same, even with crutches we have a hard time - at the words “mail server” we still see a box with light bulbs that sends mail.
All this is not a joke. Thinking objects of the real world, of course, possible. But it is the privilege of the highest professionals who master the functional decomposition so masterly that it becomes unnoticeable to an outside observer. Looking at the jugglers in the circus, it may also seem that throwing and catching objects is easy and simple. But only complete idiots can sincerely believe that they can repeat the juggler's tricks without training and training. Unfortunately, what is obvious to everyone in the circus is far from obvious to everyone in technology.
Here ITS are powerless (photo from personal mobile)
One of the most painful problems in the design of information systems in our country is the dominance of objects and tools over functionality. Many customers sincerely believe that information systems solve problems. While information systems actually allow you to solve problems. We say "electric drill bore hole". And in fact, "electric drill allows you to drill a hole." Getting into the semantic trap, we subconsciously believe that the purchase of an electric drill is equal to a hole in the wall. And then it turns out that you need to be able to use a drill, that you need electricity for a drill, that you need a tempered drill of a certain diameter, that there will be noise and dust, etc. And if, in the example with a drill, we approximately imagine the process of work and can guess what is needed besides buying a tool, then in the case of more complex systems we can stay in a sweet illusion until the very end of the project.
We now return to the definition of ITS and consider it in a new light. ITS, I repeat, is based on the modeling of transport systems and the regulation of traffic flows. “Our person”, after reading the definition, immediately makes a conclusion that he needs:
"Our man" writes the TZ, where he describes the detailed requirements for the modeling system and means of regulating traffic flows. He can study well the systems available on the market and describe them in detail. These systems will bring, deploy and connect. Maybe even on time.
Do we now have ITS? Our man will definitely say yes. A Westerner will definitely say no. Because our person assesses the availability of equipment, and the Western person assesses the performance of relevant functions.
Ask our person exactly how the equipment purchased by him will contribute to the achievement of goals (see the definition of ITS): to increase the information content, security and improve information interaction? Most likely, there will be no answer. Because the answer lies in the field of functional decomposition, which allows to move from the set goals to the functions of future systems, simultaneously catching all the necessary from adjacent areas.
The question of the use of certain elements of ITS in the city is closely related to understanding how exactly we plan to achieve the goals. And we need to move on to the technical characteristics of the equipment only after we have determined the main ways of solving problems.
Returning to the definition of ITS, we read that ITS " is an intelligent system that uses innovative developments in modeling transport systems and regulating traffic flows ."
Under the words “innovative developments in the modeling of transport systems” anything can be hidden, but if you rely on logic and technical knowledge, you can assume what is at stake.
Any automated control system, to which ITS fully applies, does one simple thing: it collects information about the control object, analyzes it and exerts a direct or indirect control effect on this object.
The control object for ITS are traffic flows. The source of information about the control object is sensors and detectors on the road, related information systems and operator input.
But in order to analyze information about the control object, it is necessary to put in the system a certain idea about this object, which is called a model . The detail and accuracy of the model is determined solely by the tasks facing ITS.
Transport models are divided into mathematical and imitation . The first operate with the known laws of traffic, presented in the form of formulas, systems of equations, etc. The second imitate the movement of individual vehicles, the behavior of drivers, the work of traffic lights, etc. In practice, a certain mixture of mathematical and simulation models is more often used.
For example, transport modeling systems at the macro level (country, city, microdistrict) operate on demographic data, the concepts of “graph of roads”, “zone of attraction”, “transport demand and supply”. They contain data on the percentage of use of cars by the population, on the capacity of streets, on the number of parking spaces at shopping centers. The macro model uses mainly mathematical modeling methods and tries to answer the questions: “why and where are everyone going?”, “Is there enough street capacity to serve everyone?”, “And what will happen if this street is blocked?” etc.
An example of a software package for macro modeling PTV Visum ( source )
Micro models operate with specific objects from the “real world” - an adjustable intersection, a traffic intersection, a street network, a car. At the same time, the micro model “knows” about the number of lanes, the presence of rises / descents, the characteristics of car engines (how quickly they can start), the rules of movement and stopping. In order for the micro model to work at full capacity, it is necessary to input information from the macro model at the entrance: the number and composition of vehicles at certain points in time (how many cars and how many trucks, how many buses, trams, etc.), behavioral features Drivers (are they often rebuilt, how often do signs follow the board and do they follow the parking rules). If the macro-level data is correct, the micro-level allows you to simulate the actual traffic flow with high accuracy.
An example of the Aimsun micro-modeling software package ( source )
The main purpose of transport models is to conduct experiments . We can check how these or other changes in the organization of the movement will affect traffic. We can set up traffic lights, make decisions on the expansion of the street, on the prohibition or permission of turns, on the organization of one-way traffic. The model will help to develop temporary plans for the organization of movement for the period of major events - competitions, street parades, etc. At the city level, transport modeling will make it possible to decide on the implications for the transport situation of the construction of another shopping center or a new neighborhood. In other words, the transport model is an indispensable tool for improving the city without serious consequences.
The more accurate the model, the more diverse information it stores. Keeping the model up to date means reflecting in it all changes in the real world — traffic overlaps, road repairs, new roads, traffic lights, lanes, residential areas, schools, offices and retail space. Keeping the model up to date is a time-consuming and responsible process that places high demands on the qualifications of personnel, on the organization of internal processes, on the quality and stability of information channels.
You have to agree that very few people initially think about what really stands behind the words innovative developments in the modeling of transport systems . After all, to organize a process of this level, train people, pay for their work, and agree on the provision of high-quality source data with different departments is tantamount to a civil feat in our country! And this is certainly not the same as the purchase and installation of a simulation system on a computer.
When we figured out the modeling and models, we can move on to the ETS functions.
Generally speaking, the need for ITS with such a question is not at all obvious. It is likely that most of the problems will be solved by competent use of the available technical means of organizing the movement. But when the available technical equipment is not enough, the question arises about the use of ITS.
We will not deviate from the definition of ITS, and remember that ITS is not only " innovative means of regulation ", but also a system " providing end users with more information and safety ."
Under the “innovative means of regulation” in the city, the most commonly understood is network coordinated control of traffic lights (the so-called “smart traffic lights”) and the placement of digital information boards at the forks.
Also to the means of informing are Internet sites for planning trips (like the well-known Yandex traffic jam service) and information support services for drivers during a trip (various navigation services). All of this is in fact also ITS subsystems, and in Western countries they are part of a single information space.
About "smart traffic lights" I once wrote on Habré ( link ), here we restrict ourselves to a fast dotted line. The combination of traffic lights into a network is in itself obvious and useful, given the cheapness of electronics in our time. In the presence of an outdoor video surveillance system, this will allow at least to adjust the traffic lights manually, sitting in a warm office, rather than standing on a dirty roadside with a remote control.
The traffic lights become “smarter” if the intersection is supplied with a system of transport detectors, and a special algorithm begins to work in the center. The need for a “smart” traffic light, as well as the settings of the control algorithm, are determined using a transport model and a special “traffic light” module, which allows to calculate the initial parameters of the control cycle and determine the limits of automatic control.
An example of the interface of the software package for configuring the smart traffic lights TRANSYT (source - "TRANSYT 14 User Guide")
Similarly, the location of the installation of digital boards and the information that will be displayed in them in one or another case is determined.
It is obvious that the ITS elements installed on city streets must be entered into the model, and the model must “know” about the algorithms of the adaptive traffic lights, board, etc. For example, for a board that recommends choosing Street A instead of Street B, the model has a rule that 80% of motorists will follow the advice, while 20% will traditionally ignore what will immediately affect transport traffic. Modern simulation systems are able to simulate the readings of detectors placed on virtual streets, the impact of electronic scoreboards and variable signs of speed limits on traffic flows, allow you to create complex control scenarios in a form suitable for use in ITS. Example of a response scenario for ITS: “If detector X detects a flux density of 70%, then display“ Y ”on the board Y sign M at the traffic lights Z mode N”.
There may be several hundred control scenarios, while the transport simulation system can automate the process of their generation.
That is, ITS is not only equipment on poles and a control center with a huge screen. ITS is primarily intelligence - control algorithms based on modeling real traffic situations, as well as the processes of their compilation, testing and implementation.
OUR MAIN TROUBLE
In matters of creating intelligent transport systems, we have not yet filled up enough cones - this area is relatively new, there are practically no specialists of ours, the implementations are rare, and even those can be considered bold experiments at public expense, and not something of practical value. Even the term "ITS" mean that it is not clear what. The official interpretation does not exist, despite the fact that in the field of ITS there is already the first standard GOST R ISO 14813-1 “Recommended architecture model for the ITS sector”. That is, there is a standard, but there is no definition of ITS in it.
Russian Wikipedia translates the definition of ITS from English, which we will use (I apologize for bold italics, but the definition will not be lost in an array of text):
')
ITS is an intelligent system that uses innovative developments in the modeling of transport systems and the regulation of traffic flows, providing end users with greater information and safety, as well as qualitatively increasing the level of interaction of traffic participants in comparison with conventional transport systems. ( Link to article )
Intelligent transport systems are the point of contact between the automotive industry and the information technology industry and are based on two “pillars” - modeling transport systems and regulating traffic flows .
The definition of ITS gives us an idea of ​​the main objectives:
- Informative and safety;
- Qualitatively new level of information interaction of road users
The above definition contains everything necessary for a proper understanding of the issue. The only thing that prevents us from understanding it correctly and doing the right thing is our traditional perception. Please take this thought seriously: we have everything we need for business, except for the right way of thinking ! In this context, the " right " way of thinking means a way of thinking sufficient to understand the Western approach to the subject and to use the available tools for solving problems, nothing more. For universal truths we are not going to chase.
The Western engineer thinks with functions, he is primarily focused on what the system should do. In our thinking, an object view of the world is “wired up”, real objects are important to us, that is, we think first of all about how the system will work. This distinction is not as subtle as it may seem at first glance.
I will give an example. The word "server" for a Western engineer means something that provides services, service. That is, the function. For our engineer, the “server” is first of all an iron box with light bulbs, that is, an object. To make sense, we have to use a variety of crutches: "server application", "mail server", "server queues", etc. And all the same, even with crutches we have a hard time - at the words “mail server” we still see a box with light bulbs that sends mail.
All this is not a joke. Thinking objects of the real world, of course, possible. But it is the privilege of the highest professionals who master the functional decomposition so masterly that it becomes unnoticeable to an outside observer. Looking at the jugglers in the circus, it may also seem that throwing and catching objects is easy and simple. But only complete idiots can sincerely believe that they can repeat the juggler's tricks without training and training. Unfortunately, what is obvious to everyone in the circus is far from obvious to everyone in technology.
Here ITS are powerless (photo from personal mobile)
One of the most painful problems in the design of information systems in our country is the dominance of objects and tools over functionality. Many customers sincerely believe that information systems solve problems. While information systems actually allow you to solve problems. We say "electric drill bore hole". And in fact, "electric drill allows you to drill a hole." Getting into the semantic trap, we subconsciously believe that the purchase of an electric drill is equal to a hole in the wall. And then it turns out that you need to be able to use a drill, that you need electricity for a drill, that you need a tempered drill of a certain diameter, that there will be noise and dust, etc. And if, in the example with a drill, we approximately imagine the process of work and can guess what is needed besides buying a tool, then in the case of more complex systems we can stay in a sweet illusion until the very end of the project.
We now return to the definition of ITS and consider it in a new light. ITS, I repeat, is based on the modeling of transport systems and the regulation of traffic flows. “Our person”, after reading the definition, immediately makes a conclusion that he needs:
- System for transport simulation;
- Means of regulation of traffic flows.
"Our man" writes the TZ, where he describes the detailed requirements for the modeling system and means of regulating traffic flows. He can study well the systems available on the market and describe them in detail. These systems will bring, deploy and connect. Maybe even on time.
Do we now have ITS? Our man will definitely say yes. A Westerner will definitely say no. Because our person assesses the availability of equipment, and the Western person assesses the performance of relevant functions.
Ask our person exactly how the equipment purchased by him will contribute to the achievement of goals (see the definition of ITS): to increase the information content, security and improve information interaction? Most likely, there will be no answer. Because the answer lies in the field of functional decomposition, which allows to move from the set goals to the functions of future systems, simultaneously catching all the necessary from adjacent areas.
The question of the use of certain elements of ITS in the city is closely related to understanding how exactly we plan to achieve the goals. And we need to move on to the technical characteristics of the equipment only after we have determined the main ways of solving problems.
TRANSPORT MODELING
Returning to the definition of ITS, we read that ITS " is an intelligent system that uses innovative developments in modeling transport systems and regulating traffic flows ."
Under the words “innovative developments in the modeling of transport systems” anything can be hidden, but if you rely on logic and technical knowledge, you can assume what is at stake.
Any automated control system, to which ITS fully applies, does one simple thing: it collects information about the control object, analyzes it and exerts a direct or indirect control effect on this object.
The control object for ITS are traffic flows. The source of information about the control object is sensors and detectors on the road, related information systems and operator input.
But in order to analyze information about the control object, it is necessary to put in the system a certain idea about this object, which is called a model . The detail and accuracy of the model is determined solely by the tasks facing ITS.
Transport models are divided into mathematical and imitation . The first operate with the known laws of traffic, presented in the form of formulas, systems of equations, etc. The second imitate the movement of individual vehicles, the behavior of drivers, the work of traffic lights, etc. In practice, a certain mixture of mathematical and simulation models is more often used.
For example, transport modeling systems at the macro level (country, city, microdistrict) operate on demographic data, the concepts of “graph of roads”, “zone of attraction”, “transport demand and supply”. They contain data on the percentage of use of cars by the population, on the capacity of streets, on the number of parking spaces at shopping centers. The macro model uses mainly mathematical modeling methods and tries to answer the questions: “why and where are everyone going?”, “Is there enough street capacity to serve everyone?”, “And what will happen if this street is blocked?” etc.
An example of a software package for macro modeling PTV Visum ( source )
Micro models operate with specific objects from the “real world” - an adjustable intersection, a traffic intersection, a street network, a car. At the same time, the micro model “knows” about the number of lanes, the presence of rises / descents, the characteristics of car engines (how quickly they can start), the rules of movement and stopping. In order for the micro model to work at full capacity, it is necessary to input information from the macro model at the entrance: the number and composition of vehicles at certain points in time (how many cars and how many trucks, how many buses, trams, etc.), behavioral features Drivers (are they often rebuilt, how often do signs follow the board and do they follow the parking rules). If the macro-level data is correct, the micro-level allows you to simulate the actual traffic flow with high accuracy.
An example of the Aimsun micro-modeling software package ( source )
The main purpose of transport models is to conduct experiments . We can check how these or other changes in the organization of the movement will affect traffic. We can set up traffic lights, make decisions on the expansion of the street, on the prohibition or permission of turns, on the organization of one-way traffic. The model will help to develop temporary plans for the organization of movement for the period of major events - competitions, street parades, etc. At the city level, transport modeling will make it possible to decide on the implications for the transport situation of the construction of another shopping center or a new neighborhood. In other words, the transport model is an indispensable tool for improving the city without serious consequences.
The more accurate the model, the more diverse information it stores. Keeping the model up to date means reflecting in it all changes in the real world — traffic overlaps, road repairs, new roads, traffic lights, lanes, residential areas, schools, offices and retail space. Keeping the model up to date is a time-consuming and responsible process that places high demands on the qualifications of personnel, on the organization of internal processes, on the quality and stability of information channels.
You have to agree that very few people initially think about what really stands behind the words innovative developments in the modeling of transport systems . After all, to organize a process of this level, train people, pay for their work, and agree on the provision of high-quality source data with different departments is tantamount to a civil feat in our country! And this is certainly not the same as the purchase and installation of a simulation system on a computer.
ITS FUNCTIONS
When we figured out the modeling and models, we can move on to the ETS functions.
Generally speaking, the need for ITS with such a question is not at all obvious. It is likely that most of the problems will be solved by competent use of the available technical means of organizing the movement. But when the available technical equipment is not enough, the question arises about the use of ITS.
We will not deviate from the definition of ITS, and remember that ITS is not only " innovative means of regulation ", but also a system " providing end users with more information and safety ."
Under the “innovative means of regulation” in the city, the most commonly understood is network coordinated control of traffic lights (the so-called “smart traffic lights”) and the placement of digital information boards at the forks.
Also to the means of informing are Internet sites for planning trips (like the well-known Yandex traffic jam service) and information support services for drivers during a trip (various navigation services). All of this is in fact also ITS subsystems, and in Western countries they are part of a single information space.
About "smart traffic lights" I once wrote on Habré ( link ), here we restrict ourselves to a fast dotted line. The combination of traffic lights into a network is in itself obvious and useful, given the cheapness of electronics in our time. In the presence of an outdoor video surveillance system, this will allow at least to adjust the traffic lights manually, sitting in a warm office, rather than standing on a dirty roadside with a remote control.
The traffic lights become “smarter” if the intersection is supplied with a system of transport detectors, and a special algorithm begins to work in the center. The need for a “smart” traffic light, as well as the settings of the control algorithm, are determined using a transport model and a special “traffic light” module, which allows to calculate the initial parameters of the control cycle and determine the limits of automatic control.
An example of the interface of the software package for configuring the smart traffic lights TRANSYT (source - "TRANSYT 14 User Guide")
Similarly, the location of the installation of digital boards and the information that will be displayed in them in one or another case is determined.
It is obvious that the ITS elements installed on city streets must be entered into the model, and the model must “know” about the algorithms of the adaptive traffic lights, board, etc. For example, for a board that recommends choosing Street A instead of Street B, the model has a rule that 80% of motorists will follow the advice, while 20% will traditionally ignore what will immediately affect transport traffic. Modern simulation systems are able to simulate the readings of detectors placed on virtual streets, the impact of electronic scoreboards and variable signs of speed limits on traffic flows, allow you to create complex control scenarios in a form suitable for use in ITS. Example of a response scenario for ITS: “If detector X detects a flux density of 70%, then display“ Y ”on the board Y sign M at the traffic lights Z mode N”.
There may be several hundred control scenarios, while the transport simulation system can automate the process of their generation.
That is, ITS is not only equipment on poles and a control center with a huge screen. ITS is primarily intelligence - control algorithms based on modeling real traffic situations, as well as the processes of their compilation, testing and implementation.
Source: https://habr.com/ru/post/175497/
All Articles