Dangerous Situations Research Articles

Risk-aware controller for autonomous vehicles using model-based collision prediction and reinforcement learning

Autonomous Vehicles (AVs) have the potential to save millions of lives and increase the efficiency of transportation services. However, the successful deployment of AVs requires tackling multiple challenges related to modeling and certifying safety. State-of-the-art decision-making methods usually rely on end-to-end learning or imitation learning approaches, which still pose significant safety risks. Hence the necessity of risk-aware AVs that can better predict and handle dangerous situations. Furthermore, current approaches tend to lack explainability due to their reliance on end-to-end Deep Learning, where significant causal relationships are not guaranteed to be learned from data.This paper introduces a novel risk-aware framework for training AV agents using a bespoke collision prediction model and Reinforcement Learning (RL). The collision prediction model is based on Gaussian Processes and vehicle dynamics, and is used to generate the RL state vector. Using an explicit risk model increases the post-hoc explainability of the AV agent, which is vital for reaching and certifying the high safety levels required for AVs and other safety-sensitive applications. Experimental results obtained with a simulator and state-of-the-art RL algorithms show that the risk-aware RL framework decreases average collision rates by 15%, makes AVs more robust to sudden harsh braking situations, and achieves better performance in both safety and speed when compared to a standard rule-based method (the Intelligent Driver Model). Moreover, the proposed collision prediction model outperforms other models in the literature.

Autonomous overtaking control considering the velocity change of obstacle vehicles on two-lane highways

Aiming at the speed change of obstacle vehicles in the process of autonomous ground vehicle (AGV) overtaking on two-lane highways, an overtaking control method including trajectory planning and trajectory control is proposed. First, the reference overtaking trajectory of the AGV is planned in stages (lane changing, passing and merging) using fifth order polynomials satisfying stability and comfort constraints. The initial planning trajectory may be unsafe due to the speed change of obstacle vehicles. So a collision prediction algorithm is built to predict two kinds of dangerous situations in real time: AGV collides with the leading vehicle during lane changing, and AGV collides with the oncoming vehicle during merging. Secondly, the potential field of the obstacle vehicles embedded with the relative speed is established and combined with the road potential field as the risk indicator. Then, combining the nonlinear model predictive control algorithm and risk index, the overtaking trajectory is replanned in real time with the initial trajectory as reference. Finally, a lateral path tracking controller based on model predictive control and a longitudinal speed tracking controller based on sliding mode control are established to track the planned trajectory rapidly and stably. To verify the performance of the proposed method, overtaking in three cases is simulated, and vehicle responding curves are obtained. The results show that the proposed control method can achieve safe and stable overtaking.

Murderinos

Consumption of true crime podcasts has shot up in recent years, with a majority of listeners being female. There are many things which might draw women to these pieces of media. Women who previously reported elevated levels of fear of violent crime have taken to viewing true crime as a safety measure, wanting to become educated on the stories of others in an effort to not become victims themselves. Many women report feeling more confident that they would know how to avoid or get out of a dangerous situation after learning the details of previous crimes. Production and consumption of true crime media can also lead to increased feelings of justice and a desire to speak out about personal experience or societal and systemic issues. Studies have also been done on the effect of hearing women’s voices on true crime podcasts and its relation to the visualization of mutilation in true crime stories. While earlier studies argue that audiences of true crime media have more anxiety and trepidation around violent crime, I argue that true crime podcasts have many psychological benefits, specifically for women. In their podcast, Karen Kilgariff and Georgia Hardstark tell murder stories with comedic commentary, including advice about how to avoid meeting the same fate as the victims they speak of. Following the success of the podcast, the hosts published Stay Sexy and Don’t Get Murdered: The Definitive How-To Guide, condensing their tips into book form and sharing their own experiences with true crime. Using select episodes from the podcast as well as the book, I will prove that participating in these audio spaces improves feelings that women have agency over their lives, giving them an awareness of what’s happening around them that makes them savvier, helps them notice patterns, and allows them to break out of the dangerous mold society has placed them in.

Vehicle Speed Controller in Restricted Area

Speeding violations are one of the leading causes of traffic accidents in Malaysia. This project proposes traffic monitoring and control using RFID technology to reduce crime. AES cameras are installed at fixed points to capture vehicles exceeding the speed limit in Malaysia. After a while, the driver will remember the location of the AES camera. This can lead to a very dangerous situation. For example, when drivers approach a police station, they slow down and do not accelerate until they pass the control area. RFID readers using Ultra High Frequency (UHF) can easily scan tags up to 100 meters long. Therefore, RFID readers can be installed away from the road, making it difficult for drivers to see the police in the area. This is why I love RFID technology after discussion and research. In this project, he will show how to collect data from the RFID reader every time a car passes through the RFID reader. Start the time when the vehicle passes the starting RFID reader and stop the time when the vehicle stops the RFID reader. The speed be calculated according to the time difference. If the measured speed exceeds the speed limit, an audible warning is given and a message is sent to the driver. Key Words: RFID, AES Cameras, Ultra High Frequency .

Traffic Risk Image in The Light of the Road Situation Assessment Questionnaire

Introduction: Drivers often lack reflection on their behaviour. For predictive purposes and educational interventions, it is useful to know the individual traffic risk picture of the driver under investigation.
 Objective: This paper aimed to show the psychometric properties of the questionnaire and to establish a hierarchy of traffic situations in terms of perceptions of associated danger.
 Material and Methods: A total of 548 people aged between 18 and 68 years participated in the study. A self-administered questionnaire method was used to measure situational (road) awareness and the ability to anticipate the consequences of engaging in unsafe road traffic actions. The tool's accuracy was verified with the 'Controversy' questionnaire, which is used to measure the severity of drivers' normative beliefs about breaking traffic laws.
 Results: Satisfactory psychometric values of the questionnaire were obtained. After taking into account the correlation of measurement errors, an acceptable fit was obtained: χ2 /df = 2.48; GFI = 0.95; AGFI = 0.92; CFI = 0.96; RMSEA = 0.06 (0.048-0.072). Mean values and standard deviations for 14 descriptions of unsafe traffic situations were shown.
 Conclusions: The satisfactory reliability and accuracy of the measurement allow the method to be used in research and individual diagnosis in the areas of transport psychology and education. Overtaking at pedestrian crossings, on a bend, uphill, and continuing to drive despite fatigue are the most dangerous traffic situations associated with the risk of traffic accidents.

REQUIREMENTS FOR HAZARD ANALYSIS MODELS AND CALCULATIONS DEGREE OF RISK

Based on statistical data, the main reasons for the low level of labor protection organization in Ukraine are formulated. In recent years, there has been a sharp increase in cases of industrial injuries. UAH 334 million was spent on monthly insurance payments to families who lost breadwinners. for temporary incapacity for work due to an accident or occupational disease - UAH 138.1 million. The data presented in the article indicate that the main causes of fatal accidents are organizational, psychophysical, technogenic, and technical. The main reasons for the low level of labor protection organization in Ukraine are given by the authors, who say that in today's conditions, the labor protection management system should be based not only on measures from the state, but also on the interest of the subjects of labor relations in maintaining the proper physical condition of the employee. A real health risk assessment is impossible without a sufficient information base on the quantitative and qualitative characteristics of risk factors and data on the health status of those contingents of the population that are exposed to these factors. The effect of potentiating with the combined influence of professional and non-professional factors has been established. Therefore, there was a need to develop new methods of analysis and modeling of the conditional probability of the occurrence of an industrial injury, acceptable for practical engineering calculations. The proposed indicators of the level of ensuring the safety of people at the objects are the probability of preventing exposure to dangerous factors of the production environment and work process, which can be determined by the full risk of an accident. Thus, on the basis of the application of the probabilistic method for modeling the conditional probability of injury to an employee during technological operations, an assessment of the probability of an accident and the development of various variants of dangerous situations was carried out. This will make it possible to get an assessment of the circumstances and conditions that contribute to the occurrence and development of traumatic situations at work. Keywords: labor protection, occupational injuries, industrial environment, accident, risk, danger, harmfulness, probability.

Simultaneous Estimation of Vehicle Sideslip and Roll Angles Using an Integral-Based Event-Triggered H$_{\infty }$ Observer Considering Intravehicle Communications

In recent years, several technological advances have been incorporated into vehicles to ensure their safety and ride comfort. Most of these driver-assistance technologies aim to prevent skidding, whereas less attention has been paid to the avoidance of other dangerous situations such as a rollover. Since knowledge of slip and roll angles is critical to the control and safety of vehicle handling, their estimation remains of great interest when addressing emerging constraints in modern technologies involving networked communications and distributed computing. This paper presents an integral-based event-triggered H <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$_{\infty }$</tex-math></inline-formula> observer to simultaneously estimate the sideslip and roll angles, considering intravehicle communications with a networked-induced delay. As the longitudinal velocity and tire cornering stiffness of a vehicle can vary significantly during driving and have a strong influence on vehicle lateral stability, these time-varying parameter uncertainties are considered in the design of the observer. The simulation and experimental results demonstrate the effectiveness of the proposed observer.

Research on coupled dynamic characteristics of Floating Wind Turbine under complex environment

The fully coupled aerodynamic-hydrodynamic-mooring numerical model of floating offshore wind turbine(FOWT) has been established. The complex conditions such as second-order wave force, full-field turbulent wind, and mooring break have been taking into account. Using OrcaFlex and FAST to calculate the dynamic response of FOWT under the combined action of wind, wave and current. The time-domain response of floating foundation, mooring tension, blade root contact force, aerodynamic power, and wind rotor thrust are obtained. And the effect of the turbulent wind and mooring break at different positions on the FOWT are studied. The results show that compared with the platform motion and mooring tension, the aerodynamic power and rotor thrust are more sensitive to the turbulent wind, and there are fluctuations in amplitude and multiple peaks in the frequency spectrum. Compared with the leeward side, the mooring break on the windward side is a more dangerous situation, and it is easy to cause the collapse of platform and the break of the remaining mooring. The mooring break on the leeward side is relatively safe, and it will produce a drift in the sway direction, and have little impact on the safety of the wind turbine.

Street-crimes modelled arms recognition technique employing deep learning and quantum deep learning (SMARTED)

An &lt;span&gt;increase in population causes loopholes in controlling law and order situations. One of the threatening aspects of peace is the availability of weapons to the general public. As a result, many dangerous situations arise, most notably street crimes. Traditional methods are not sufficient to deal with such situations. Consequently, the police and other security concerns need serious technological reforms to prevent such situations. In modern technology tools, deep learning has made great improvements in various areas of daily life, especially in object detection. This paper presents an efficient technique for detecting weapons from closed-circuit television (CCTV) cameras, videos, or images. Upon the detection of the weapon, the concerned person is automatically informed to take the necessary action; without human intervention. For the first time, RetinaNet has been employed to detect weapons in real-time scenarios. RetinaNet has shown remarkable improvement in this domain, by achieving an average of 90% accuracy in real-time scenarios. With the emergence of quantum computing, many computer environments saw a revolution. Thus, we have also utilized quantum computing technology for real-time weapons detection using quantum deep learning. In this paper, quantum inspired RetinaNet (QIR-Net) is developed for weapons detection and amazing results are observed.&lt;/span&gt;

Visually Impaired People Positioning Assistance System Using Artificial Intelligence

Blindness and visual impairment are commonly associated with social and functional limitations, almost 45 million people in the world have blindness, and 135 million have any visual impairment. This condition has a significant impact on the quality of life and brings many challenges to the individual, one of which is the navigation and positioning tasks. Although there are already apps capable of helping visually impaired people for mobility purposes, most of them focus on detecting obstacles and, therefore, on avoiding dangerous situations. However, mobility of visually impaired people involves many more tasks, such as knowing their exact position and staying informed along an entire route. For this purpose, a standalone and customizable solution is proposed that uses traditional visual recognition of landmarks to process the surroundings of the current location of the visually impaired person using a smartphone and informing about the nearby places assuring the user a sense of the site. For feature detection, it was used the Oriented FAST and Rotated BRIEF (ORB) algorithm, and for feature matching, it was used the brute-force method with the k-nearest neighbors (KNN) algorithm. Results show that the proposed solution can analyze pictures in fractions of a second with satisfactory accuracy.

Precautions For Jakarta-Bandung High-Speed Rail Infrastructure Project Cooperation: Construction At Night And In Storm

The Jakarta-Bandung high-speed rail infrastructure project delays cause the financial burden in the form of interest during construction to swell KCIC had to improve project performance by carrying out construction at night and must improve strategies for dealing with Indonesian extreme weather. Before construction at night, strict inspection shall also be focused on lighting lamps. Construction personnel shall prepare flashlights, searchlights, all parts and components shall be carefully inspected in strict accordance with relevant regulations to ensure the normal operation of the equipment. During the flood season, the construction personnel of large-scale railway maintenance machinery shall strictly implement the measures for safe driving in the flood season issued by the Ministry of railways. the construction staff shall report the disaster and dangerous situation to the attendant of the nearest station through the train dispatching station in time. When large-scale railway maintenance machinery encounters strong wind, rainstorm, Blizzard and other difficulties or unknown conditions during construction and operation, it shall immediately stop construction and slow down, and timely report the situation to the station.&#x0D; Keyword: KCIC, Construction, Railway, Maintenance.

Имитационные модели оценки опасности возникновения подземных аварий в угольных шахтах

The article is related to the safety issues of the coal mines considering not only the standard conditions of their operation, but also the influence of external effects, personnel errors during mining operations and operation of controls and automation. The relationship between the causes of accidents and sources of hazards is shown. The analysis of mine safety research methods was carried out, which allow to identify the most dangerous causes and conditions of the underground accidents to one degree or another. It is shown that at the mining sites, despite the extremely small, but non-zero probability, accidents still occur due to the probability densities of rare dangerous situations leading to severe consequences. The expediency of logical-probabilistic methods for modeling scenarios of possible accidents, which are interpreted by the cause-and-effect diagrams of the occurrence of a chain of incidents, is asserted. Based on the normalization of the frequencies of occurrence of emergency situations in coal mining, a rationale is given for the transition from the weights of various initiating conditions to their contributions to emergency situations. Simulation models of decision-making for operational substantive analysis of emergency situations were developed, including an assessment of the weights and contributions of mining and geological, mining and technical factors, and personnel errors to the risk of methane explosion occurrence, fires and gasification of workings in case of violations in the degassing systems. It is shown that the existing hazards must be able to measure, i.e. have numerical scales. These hazardrs can be assessed based on the construction of hazard trees. Using them, as well as assessments of emergency changes in the state of the technosphere of coal mines, which are based on the analysis of cause-and-effect relationships of hazard sources, simulation models of the hazard of underground accidents were formed and implemented. These models can be considered as functional models of any emergency situation, allowing to obtain analytical estimates of variations in emergency situations by their weights and contributions to the hazard of emergency situations.

Grand Theft Auto-Based Cycling Simulator for Cognitive Enhancement Technologies in Dangerous Traffic Situations.

While developing traffic-based cognitive enhancement technology (CET), such as bike accident prevention systems, it can be challenging to test and evaluate them properly. After all, the real-world scenario could endanger the subjects' health and safety. Therefore, a simulator is needed, preferably one that is realistic yet low cost. This paper introduces a way to use the video game Grand Theft Auto V (GTA V) and its sophisticated traffic system as a base to create such a simulator, allowing for the safe and realistic testing of dangerous traffic situations involving cyclists, cars, and trucks. The open world of GTA V, which can be explored on foot and via various vehicles, serves as an immersive stand-in for the real world. Custom modification scripts of the game give the researchers control over the experiment scenario and the output data to be evaluated. An off-the-shelf bicycle equipped with three sensors serves as a realistic input device for the subject's movement direction and speed. The simulator was used to test two early-stage CET concepts enabling cyclists to sense dangerous traffic situations, such as trucks approaching from behind the cyclist. Thus, this paper also presents the user evaluation of the cycling simulator and the CET used by the subjects to sense dangerous traffic situations. With the knowledge of the first iteration of the user-centered design (UCD) process, this paper concludes by naming improvements for the cycling simulator and discussing further research directions for CET that enable users to sense dangerous situations better.

Forgotten Victims: Understanding the Victimisation Vulnerability of the Homeless

Having undergone a radical political transformation from apartheid to democracy, South Africa is commonly seen as having one of the most intriguing transformation stories in the world. Furthermore, it is often suggested that the aftershock of apartheid has forced many individuals to the periphery of society. The homeless form part of those burdened by the past and as a result are often confronted by adversity. Existing literature as well as various victimological perspectives indicate that the homeless are prone to high rates of violence and victimisation. As a result of their exposure and high levels of vulnerability, the homeless often find themselves in unfavourable conditions, often typified by inherently dangerous people, places and situations. A study exploring the experiences of victimisation of the homeless - guided by a qualitative methodological approach - identified various victimogenic risk factors such as childhood adversity, a devalued social status, the use/abuse of alcohol and narcotics, and conditions associated with psychological and physical well-being, which could potentially prove useful in explaining the victimisation vulnerability of members of this vulnerable population. It is therefore important to assess these factors, to generate a better understanding of this complex phenomenon, thereby increasing the level of sensitivity around the plight of the homeless victim – which will in turn not only acknowledge their role as victims but also better inform and improve the victim assistance measures created and offered.

Effects of hyperprotein diet on anxiety, haemodynamics and morphofunctional aspects of the heart of Wistar rats.

What is the central question of this study? The consumption of a high-protein diet has been associated with an anxiogenic factor that can influence anxiety and possible cardiovascular changes: does the consumption of a high-protein diet interfere with anxiety, haemodynamics and morphofunctional aspects of the heart of Wistar rats? What is the main finding and its importance? Our study showed that the high-protein diet did not interfere with anxiety and haemodynamics. The animals in the hyperproteic group showed positive heart adaptations characterized by less work and lower heart rate without impairing ejection fraction and systemic blood pressure. Anxiety is a mechanism preparatory to a response in situations of threat and danger, involving behavioural, affective and physiological factors. Protein-based foods have a high concentration of amino acids which perform multiple functions, including in the biosynthesis of excitatory transmitters for the central nervous system. In recent years, adherence to high-protein diets has been gaining ground in society, on the basis that it brings benefits to the musculoskeletal system and cardiovascular health. The aim of the present study was to investigate the effect of a high-protein diet in a state of anxiety and to investigate morphofunctional cardiovascular effects of a high-protein diet in Wistar rats. The experiment lasted 8weeks and two groups of male rats were submitted to either a normoproteic or a hyperproteic diet. Anxiety was assessed using the plus maze test and cardiovascular morphofunctional aspects using transthoracic echocardiography and invasive measurements of femoral blood pressure. There was no statistically significant difference in the anxiety test, but the hyperproteic group was more agitated, with greater displacement during the test. Changes were found in systolic and end-diastolic volume, left ventricular diameter in systole and heart rate, which were significantly lower in the hyperproteic group, and there was an increase in the thickness of the interventricular septum in diastole. The results showed no influence of the higher protein diet on the animals' anxiety, body weight and haemodynamics.

Silné konvektivní bouře a vydatný déšť v období 24. 6. až 1. 7. 2022

The storm season of 2022 was rich in dangerous and notable storm situations. Although dangerous stormy weather occurred also in other months of summer, especially in August, the longest period of more than a week with storms occurred in the last week of June, when the high intensity of precipitation led to a repeated announcement of even an extreme warning level. This period is evaluated in detail in the article. In that season, storm situations were continuously communicated on social networks having elicited a positive response from the professional and lay public.

Safety of digital learning environment for the health of high school and university students in distance learning

The development of technologies using the electromagnetic field (EMF) of the radio frequency range has become a leading trend related to the need for wireless data transmission. This made it possible for distance learning of schoolchildren and students in a dangerous epidemic situation. Accumulated data on the sensitivity of the human body to the action of EMF. The purpose of the work is to assess the conditions of teaching schoolchildren in computer science classrooms, as well as the impact of online learning on the health of high school students and university students. Based on the results of an online survey of schoolchildren and students according to the author’s questionnaire, a database was formed and an assessment of the conditions of distance learning was carried out. With the use of instrumental research, an assessment of the learning conditions in computer science classrooms was carried out. The materials were processed using the reliability criterion χ2 and the health risk assessment RR. Control data in computer science classrooms showed non-compliance of temperature, air humidity, air ionization, illumination, noise with regulatory requirements, except for EMF levels. With online learning, a high duration of stay of high school students and students in a digital environment has been established, including the learning process, homework, and leisure time on the Internet. In the structure of complaints related to studying online and doing homework, there were: visual impairment over the past year, headaches, back pain, irritability, eye fatigue, a feeling of dryness in the eyes. The time for sleep and physical activity decreased. Data have been obtained confirming the need to protect the health of students in a digital environment with the provision of safe conditions for visual work, limiting the time of use of digital devices, increasing the literacy of all users.

Legal regulation of the will of a person in relation to the conduct of criminal procedural actions in the conditions of martial law

The article is devoted to the study of the legal regulation of a person's expression of will regarding the conduct of criminal procedural actions under martial law. When posing the problem, it is emphasized that the will of a person in criminal proceedings directly affects the activities of the subjects of the criminal process in terms of the realization of human rights and freedoms. In the process of revealing the state of development of the relevant issues, it is noted that scientists did not consider problematic issues regarding the legal regulation of the will of a person during criminal proceedings under martial law. In the course of the presentation of the main material, it is indicated that during the legal regime of martial law, the lack of consent to conduct criminal procedural actions can in exceptional cases be compensated by prosecutorial supervision (control). It is noted that the procedure for obtaining a court or prosecutor's (under martial law) permission is quite cumbersome, formalized and sometimes long, which determines the desire of the subjects of the prosecution to obtain in any way the consent of a person to conduct criminal procedural actions. Attention is drawn to the fact that the regulatory requirement regarding the written consent will not solve all the problems regarding its voluntary nature. It is determined that the subjects of the criminal process, abusing their powerful powers, through the use of coercion, may illegally obtain a person's "voluntary" consent to conduct procedural actions and record it through the drafting of a written document. It is noted that during the period of martial law, the situation with the "voluntariness" of written consent may deteriorate due to the conduct of military operations. the dangerous situation in certain territories, the limitation of procedural tools of proof during martial law creates a significant risk of ignoring the true will of a person regarding the implementation of criminal procedural actions. It is concluded that in order to prevent an irreparable violation of the rights and freedoms of a person who voluntarily consents to the implementation of criminal procedural actions during the period of martial law, it is advisable to provide for such a person at the legislative level the necessary procedural guarantees of rights and freedoms. It is noted that the scope of procedural guarantees should be determined depending on the degree of risk of violation of human rights and freedoms.

CNN-Based Crosswalk Pedestrian Situation Recognition System Using Mask-R-CNN and CDA

Researchers are studying CNN (convolutional neural networks) in various ways for image classification. Sometimes, they must classify two or more objects in an image into different situations according to their location. We developed a new learning method that colored objects from images and extracted them to distinguish the relationship between objects with different colors. We can apply this method in certain situations, such as pedestrians in a crosswalk. This paper presents a method for learning pedestrian situations on CNN using Mask R-CNN (Region-based CNN) and CDA (Crosswalk Detection Algorithm). With this method, we classified the location of the pedestrians into two situations: safety and danger. We organized the process of preprocessing and learning images into three stages. In Stage 1, we used Mask R-CNN to detect pedestrians. In Stage 2, we detected crosswalks with the CDA and placed colors on detected objects. In Stage 3, we combined crosswalks and pedestrian objects into one image and then, learned the image to CNN. We trained ResNet50 and Xception using images in the proposed method and evaluated the accuracy of the results. When tested experimentally, ResNet50 exhibited 96.7% accuracy and Xception showed 98.7% accuracy. We then created an image that simplified the situation with two colored boxes of crosswalks and pedestrians. We confirmed that the learned CNN with the images of colored boxes could classify the same test images applied in the previous experiment with 96% accuracy by ResNet50. This result indicates that the proposed system is suitable for classifying pedestrian safety and dangerous situations by accurately dividing the positions of the two objects.

Using convolutional neural networks for acoustic-based emergency vehicle detection

Background: A siren is a special signal given by emergency vehicles such as fire trucks, police cars and ambulances to warn drivers or pedestrians on the road. However, drivers sometimes may not hear the siren due to the sound insulation of a modern car, the noise of city traffic, or their own inattention. This problem can lead to a delay in the provision of emergency services or even to traffic accidents.&#x0D; Aim: develop an acoustic method for detecting the presence of emergency vehicles on the road through the use of convolutional neural networks.&#x0D; Materials and Methods: The algorithm of work is based on the conversion of sound from the external environment into its spectrogram, for analysis by a convolutional neural network. An open dataset (Emergency Vehicle Siren Sounds) from sources available on Internet sites such as Google and Youtube, saved in the .wav audio format, was used as a dataset for siren sounds and city traffic. The code was developed on the Google.Colab platform using cloud storage.&#x0D; Results: The conducted experiments showed that the proposed method and model of the neural network make it possible to achieve an average efficiency of determining the type of sound with an accuracy of 93.3 % and a speed recognition of 0.00045 % of a second.&#x0D; Conclusion: The use of the developed technology for recognizing siren sounds in city noize will improve traffic safety and increase the chances of preventing a dangerous situation. Also, this system can be an additional assistant for hearing-impaired people while driving and everyday life for timely notification of the presence of emergency services nearby.