{"id":204,"date":"2024-01-25T08:53:56","date_gmt":"2024-01-25T05:53:56","guid":{"rendered":"http:\/\/cues.group\/?page_id=204"},"modified":"2025-03-13T14:52:51","modified_gmt":"2025-03-13T11:52:51","slug":"results","status":"publish","type":"page","link":"http:\/\/cues.group\/?page_id=204&lang=en","title":{"rendered":"Results"},"content":{"rendered":"\n

Our results for the second half of 2024:<\/strong><\/h3>\n

In accordance with subtasks of calendar plan 1.2 (WP1.2) and 1.3 (WP1.3), the following work was completed:<\/p>\n

1\u00a0\u00a0\u00a0\u00a0 A software and hardware platform for an unmanned aerial vehicle complex for performing monitoring tasks within the city economy (PABLKG) has been developed.<\/strong><\/p>\n

1.2 \u00a0Prototypes of the PAPBLKG (heavier-than-air) hardware have been developed (WP1.2)<\/p>\n

Composite drone parts manufactured. Components selected. Motor and battery controllers selected. Drone assembled and payload installed.<\/p>\n

1.2.1\u00a0A universal drone for eco-monitoring based on a hexacopter has been developed and assembled.<\/strong><\/p>\n

\"\"
Universal drone for eco monitoring based on hexacopter<\/figcaption><\/figure>\n

The drone frame was designed using the SolidWorks software product.<\/p>\n

\"\"
Drone Frame Design in SolidWorks<\/figcaption><\/figure>\n

A drone with six engines (hexacopter) was selected based on the requirements of payload capacity, flight time, compactness and the ability to continue a safe flight in emergency mode if one of the engines fails. The production of master models of the drone body parts was carried out on a CNC machine. The manufacturing process is shown in the figure.\u00a0<\/p>\n

\"\"
Manufacturing master models of parts on a CNC machine<\/figcaption><\/figure>\n

1.2.2 Components for the manufactured drone have been selected<\/strong><\/p>\n

An electric motor, propellers, controllers, speed controller, sensors, receiver, and batteries have been selected.<\/p>\n

1.2.3 The drone is assembled with all components<\/p>\n

\"\"
Installing the motor and controller Installing the central board Assembled drone (bottom view)<\/figcaption><\/figure>\n
\"\"
Installing the central board<\/figcaption><\/figure>\n
\"\"
Assembled drone (bottom view)<\/figcaption><\/figure>\n

1.3 Flight planning models and algorithms have been developed (WP1.3)<\/strong><\/p>\n

The following steps were performed: requirements analysis, optimization algorithms, creation of mathematical models, development of a program for converting the obtained waypoints, testing and validation.<\/p>\n

2 A set of on-board mounted systems has been developed to solve the problems of monitoring urban infrastructure (detection of garbage, air and water pollution) (subtask WP2.2)<\/strong><\/p>\n

Atmospheric air analysis sensors have been developed that are installed on an unmanned aerial vehicle. The developed sensors are designed to measure pollutant indicators such as NO2, CO, NH3, CO2, VOCs, PM2.5, PM10.<\/p>\n

The 3D model of the sensors was designed using the Onshape CAD application. Models of all components were created at a scale of 1:1. The appearance of the housing is shown in the figure.<\/p>\n

\"\"
3D model of the sensor for a small UAV, rear view. IMU version with MQ135<\/figcaption><\/figure>\n
\"\"
Ready-made model of the sensor on the drone. IMU version with MQ135<\/figcaption><\/figure>\n

3 Software tools have been developed to solve problems of classification and recognition of images and data obtained from UAVs to solve problems of mapping environmental disturbances in the city’s air and ground environment, heat losses (subtasks WP3.1, WP3.2, WP3.3)<\/strong><\/p>\n

3.1 Overflights were conducted and data on pollution within the urban agglomeration was collected (WP3.1)<\/strong><\/p>\n

\"\"
Location of overflights on the map<\/figcaption><\/figure>\n

The flights were carried out by two drones: the first was equipped with a Micasense RedEdge-MX multispectral camera, an RGB camera and a FLIR thermal imager, the second drone was a DJI Mavic Mini equipped with an RGB camera.<\/p>\n

\"\"
Waste dump captured by drone camera within city limits\u00a0<\/figcaption><\/figure>\n
\"\"
Thermal leaks on the roof of the EPC Group building<\/figcaption><\/figure>\n

During the work carried out under the program, air quality measurements were taken using the AtmoTube Pro sensor.<\/p>\n

The AtmoTube Pro wearable sensor measures air quality (PM1, PM2.5, PM10), as well as volatile organic compounds (VOCs), temperature, humidity and atmospheric pressure.<\/p>\n

During the measurements, the air quality is displayed in the sensor’s mobile application as an air quality index (AQI), calculated based on the above-mentioned PM and VOC indicators.<\/p>\n

Next, an air quality map was built based on the measurements taken. Also, based on the air quality index, threshold values \u200b\u200bare displayed, summarizing the information and indicating the level of air pollution.\u00a0<\/p>\n

\"\"
AtmoTube Pro Sensor Mobile App Interface<\/figcaption><\/figure>\n

3.2 Labeled datasets of pollution within the urban agglomeration were generated for tuning machine learning models (WP3.2)<\/strong><\/p>\n

A comparison was made between the weather data services Weatherbit, OpenWeather, Yandex, stormglass and AQICN. Preference was given to OpenWeather and the Yandex.Weather service, since Stormglass and WeatherBit have a low update frequency and spatial resolution, as well as a limit on the number of free requests.<\/p>\n

3.3. A Telegram bot has been developed to generate data sets about garbage using the crowdsourcing method.<\/strong><\/p>\n

\"\"
Example of sending images and coordinates via the bot interface<\/figcaption><\/figure>\n

3.3 Methods have been developed for solving problems of classification and recognition of images and data obtained from UAVs to solve problems of mapping environmental disturbances in the city’s air and ground environment, heat loss (WP3.3)<\/strong><\/p>\n

The possibilities of satellite monitoring of the spectral characteristics of the water surface of Lake Sorbulak (55 km2, wastewater storage facility for Almaty) in the Almaty region of Kazakhstan, which has been accumulating wastewater from Almaty since 1975, are considered.<\/p>\n

The activities of self-purification processes of the Almaty air basin due to mountain-valley circulation (MVC) in stable weather conditions were analyzed.<\/p>\n

\"\"
Example of wind field map format<\/figcaption><\/figure>\n
\"\"
Frequency of recording mountain breeze episodes in the period December 2023 — January 2024 within the city of Almaty<\/figcaption><\/figure>\n

The activities of self-purification processes of the Almaty air basin due to mountain-valley circulation (MVC) under stable weather conditions were analyzed.<\/p>\n

\"\"
An example of a wind scene (at a height of 10 m) for the Almaty region in the Yandex Weather informer at 2 a.m. on August 15, 2024<\/figcaption><\/figure>\n

An analysis of the state of vegetation was carried out. Using Sentinel-2 data, available since 2016, with a spatial resolution of 10 m, made it possible to characterize the vegetation of urbanized areas, parks, squares, alleys and other plantings.<\/p>\n

A comparison of the results of the trend analysis of the state of vegetation in southern Kazakhstan was built on the basis of a linear regression analysis of the time series of the MOD09A1 product (2002-2020) against estimates of the average direction of long-term monotonic trends (Mann-Kendall test) based on MOD13Q1A1_NDVI (2000-2022).\u00a0<\/p>\n

\"\"
Comparison of the results of trend analysis of the state of vegetation in the south of Kazakhstan<\/figcaption><\/figure>\n

Change in atmospheric smoke in Almaty from January 27, 2025 to January 29, 2025 (result of a stationary camera)<\/strong><\/p>\n