MOSCOW, 7 Nov. Scientists from King Abdul-Aziz University (Jeddah, Saudi Arabia) and South Ural State University (SUSU) agreed to jointly refine and implement a system for monitoring pollutant emissions from transport at intersections, developed at the Russian university. Scientists plan to carry out work within the framework of an international grant, reported the SUSU press service.
The AIMS-Eco system, created by a team of SUSU scientists under the leadership of Associate Professor of the Department of Automotive Transport Vladimir Shepelev, uses artificial intelligence to analyze emission plumes left by various types of vehicles at the crossroads.
According to the authors, monitoring does not require complex measuring instruments; the system is built on the use of existing telecommunications and meteorological infrastructure of cities and allows remote connection and maintenance of digital environmental monitoring posts anywhere in the world. In addition, South Ural scientists “taught” artificial intelligence to predict emissions taking into account meteorological forecasts.
Currently existing foreign methods that determine the amount of emissions on roads work based on statistical data, he said Shepelev.
“These systems collect information and then use it in calculations. But this approach does not allow managing environmental risks in real time. The fact is that depending on changes in temperature, pressure or wind during the day, emissions can change significantly. It is the continuous analysis of this information that allows us to promptly influence road traffic in order to reduce emissions harmful to people.”
Vladimir Shepelevdocent, Department of Automotive Transport, SUSU
The system created at SUSU has already been tested in Russian cities – Perm, Chelyabinsk, Magnitogorsk, St. Petersburg and can successfully complement the developments of scientists from Saudi Arabia, the researchers believe.
“Our colleagues from Saudi Arabia have created a technology that allows cars to transmit information to the outside world while driving about where they are and what emission class they have. Integrating this technology with our platform will significantly expand the capabilities of both. The fact is that now our cameras only see the situation at the intersection, but this is not enough to analyze the situation on a section of the road network of significant duration,” explained Vladimir Shepelev.
Conventional traffic cameras that are aimed at lanes cannot be used for integration with the system, since they only see vehicle numbers, but do not see the depth of the queue of cars and do not allow one to estimate how long traffic has stopped, said co-author of the development, associate professor Department of System Programming at SUSU Olga Ivanova. According to her, it is the delay of vehicles in traffic jams or at traffic lights that causes the main damage to the environment.
“Tracking a queue of cars is a very difficult task, since cars can stand far from the camera and overlap each other. With the help of foreign colleagues, we can solve this problem. We will be able to identify vehicles, determine how long they move, how much they cost and how much they form emissions when driving and when idle. Such a tool as we will create together has never existed before,” she noted.
The result of the work will allow traffic lights to be configured in such a way as to ensure non-stop passage of vehicles from intersection to intersection. In this case, priority will be given not just to those areas where more cars have accumulated, but to those where more buses and trucks have accumulated, which cause the main damage to the air basin of the road network. All these measures are designed to significantly reduce environmental air pollution in cities, said Alexander Glushkov, associate professor of the SUSU Department of Mathematical and Computer Modeling.
“With the help of the developments of our colleagues from Saudi Arabia, we will also be able to assess air pollution even when driving electric vehicles. It is important to note that the transition to electric vehicles will eliminate harmful emissions from fuel combustion, but there will remain in the air a suspended fine suspension of dust rising when vehicles pass. Therefore it is important to learn how to model and control the movement of flow, calculating air turbulence in urban areas surrounding transport routes,” he said.
Scientists plan to create a new multi-component system architecture that reflects the state of the level of pollution by various air substances in real time. They are developing models of air movement in urban canyons, methods for classifying road accidents, taking into account their impact on the formation of congestion.
Plans include creating digital passports of the busiest intersections, assessing the potential of transport infrastructure and its existing capabilities, identifying driver behavior patterns, and creating models ecological trails for pedestrians.