Air pollution and nuisance investigation equipment, systems and methods

Abstract

This application relates to the field of air pollutant or nuisance source identification and detection technology, and discloses an air pollution and nuisance investigation device, system, and method. The investigation device includes: at least two air intake units and corresponding sensor units, a central processing unit, and a motion unit; the at least two air intake units are tilted in different directions to collect at least two air samples from different directions and positions; the sensor units detect the air samples to obtain concentration data of air pollutants and / or nuisance sources in the air samples; the central processing unit generates motion signals based on the differences in concentration data; and the motion unit controls the investigation device to orient and move according to the motion signals. The investigation device of this application can collect at least two air samples, accurately and efficiently detect the concentration data of multiple air pollutants, and use this concentration data to control the device to effectively move towards the pollution source, quickly locating the pollution source.

Description

Technical Field

[0001] This application relates to the field of air pollutant or nuisance source identification and detection technology, and in particular to an air pollution and nuisance investigation device, system and method. Background Technology

[0002] Environmental protection departments and agencies around the world receive numerous complaints about air pollution, the source of which is often unclear, causing many affected individuals to experience symptoms such as discomfort, nausea, and even dizziness. In some cases, identifying the source of pollution is crucial to quickly stop the disturbance, especially when vulnerable children are involved.

[0003] Traditionally, investigators have relied primarily on their sense of smell to track and identify pollution sources. Prolonged investigations can pose a threat to the safety and health of frontline workers, especially when dealing with environmental disturbances such as Mysterious Gas Incidents (MGIs). While gas detectors can be used to measure the concentration of certain known pollutants, they provide limited information, only measuring the concentration of some identified or suspected pollutants. Further manual selection and analysis based on the concentration data are then required. Therefore, current technology has significant limitations in predicting and tracing pollution sources, and investigations remain highly dependent on manual searching and the experience and judgment of investigators. Summary of the Invention

[0004] This application provides an air pollution and nuisance investigation device, system, and method that can identify and investigate pollution or nuisance sources in real time without human intervention.

[0005] In a first aspect, embodiments of this application provide an air pollution and nuisance investigation device, comprising:

[0006] At least two air intake units and a sensor unit connected to each air intake unit;

[0007] A central processing unit; and

[0008] A motion unit;

[0009] The at least two air intake units are tilted in different directions to collect at least two air samples from different directions and positions. The sensor unit detects the air samples collected by its corresponding air intake unit to obtain concentration data of one or more air pollutants and / or nuisance sources in the air samples. The central processing unit generates a motion signal based on the difference in concentration data of the at least two air samples and sends the motion signal to the motion unit. The motion unit controls the air pollution and nuisance investigation equipment to orient and move according to the motion signal.

[0010] Optionally, the central processing unit generates motion signals based on the differences in concentration data from at least two air samples, including:

[0011] The central processing unit obtains the concentration difference value of the at least two air samples based on the concentration data of the at least two air samples, and compares the concentration difference value with a preset difference threshold.

[0012] If the concentration difference value is greater than or equal to the difference threshold, a new direction of movement is determined based on the air sample with a higher concentration, and the motion signal is generated according to the new direction of movement.

[0013] If the concentration difference value is less than the difference threshold, the current movement direction is obtained, and the motion signal is generated based on the current movement direction.

[0014] Optionally, the central processing unit uses a trained AI prediction model to analyze the concentration data of some or all of the air samples in the at least two air samples to predict the type of pollution or nuisance source.

[0015] Optionally, each type of pollution or nuisance source corresponds to a "fingerprint" concentration distribution map, which is generated based on the typical concentrations of various air pollutants or nuisance sources corresponding to the type of pollution or nuisance source. The central processing unit generates motion signals based on the differences in concentration data from at least two air samples, including:

[0016] The central processing unit obtains the concentration distribution maps corresponding to the at least two air samples based on the concentration data of the at least two air samples;

[0017] From the concentration distribution maps corresponding to the at least two air samples, determine the concentration distribution map that is most similar to the "fingerprint" concentration distribution map corresponding to the predicted type of pollution or nuisance source;

[0018] A new direction of movement is determined based on the air sample corresponding to the most similar concentration distribution map, and the motion signal is generated based on the new direction of movement.

[0019] Optionally, the central processing unit counts the number of times the air sample is detected, and when the count reaches a preset number, it controls the air pollution and nuisance investigation equipment to stop its investigation.

[0020] Optionally, the air intake unit includes a directional air intake port and a forced air intake module. The directional air intake port is tilted in a preset direction, and the forced air intake module controls the air sample to enter the corresponding sensor unit at a preset flow rate.

[0021] Optionally, the motion unit includes any one of a propulsion motor, a drone motor, or a marine motor.

[0022] Optionally, the air pollution and nuisance investigation device is implemented in the form of a robot dog.

[0023] Secondly, embodiments of this application provide an air pollution and nuisance investigation system, including the air pollution and nuisance investigation equipment described above and a control terminal communicatively connected to the air pollution and nuisance investigation equipment.

[0024] Thirdly, embodiments of this application provide a method for investigating air pollution and nuisance, the method being applied to an air pollution and nuisance investigation device, including:

[0025] Collect at least two air samples from different directions and locations;

[0026] The at least two air samples are tested separately to obtain concentration data of one or more air pollutants and / or sources of disturbance in the at least two air samples;

[0027] A motion signal is generated based on the difference in concentration data from the at least two air samples;

[0028] The air pollution and nuisance investigation equipment is controlled to orient and move according to the motion signal.

[0029] Optionally, generating a motion signal based on the difference in concentration data from the at least two air samples includes:

[0030] The concentration difference value of the at least two air samples is obtained based on the concentration data in the at least two air samples, and the concentration difference value is compared with a preset difference threshold.

[0031] If the concentration difference value is greater than or equal to the difference threshold, a new direction of movement is determined based on the air sample with a higher concentration, and the motion signal is generated according to the new direction of movement.

[0032] If the concentration difference value is less than the difference threshold, the current movement direction is obtained, and the motion signal is generated based on the current movement direction.

[0033] Optionally, after obtaining the concentration data of one or more air pollutants and / or nuisance sources in the at least two air samples, the method further includes:

[0034] The concentration data of some or all of the air samples in the at least two air samples are analyzed using a trained AI prediction model to predict the type of pollution or nuisance source.

[0035] Optionally, each type of pollution or nuisance source corresponds to a "fingerprint" concentration distribution map, which is generated based on the typical concentrations of various air pollutants or nuisance sources corresponding to the type of pollution or nuisance source. Generating a motion signal based on the difference in concentration data from the at least two air samples includes:

[0036] Based on the concentration data of the at least two air samples, obtain the concentration distribution maps corresponding to the at least two air samples respectively;

[0037] From the concentration distribution maps corresponding to the at least two air samples, determine the concentration distribution map that is most similar to the "fingerprint" concentration distribution map corresponding to the predicted type of pollution or nuisance source;

[0038] A new direction of movement is determined based on the air sample corresponding to the most similar concentration distribution map, and the motion signal is generated based on the new direction of movement.

[0039] Optionally, the process of collecting at least two air samples from different directions and locations may include:

[0040] Air samples were collected and tested from various known types of pollution or nuisance sources to obtain concentration data of air pollutants or nuisance sources in various known types of pollution or nuisance sources;

[0041] The AI ​​prediction model, constructed using machine learning algorithms, is trained based on concentration data of various known types of pollution or nuisance sources to obtain the trained AI prediction model.

[0042] This application provides an air pollution and nuisance investigation device, including at least two air intake units and a sensor unit connected to each air intake unit; a central processing unit; and a motion unit. The at least two air intake units are tilted in different directions, allowing them to collect at least two air samples from different directions and locations. The sensor units detect the air samples collected by their corresponding air intake units to obtain concentration data of one or more air pollutants and / or nuisance sources in the air samples. The central processing unit generates a motion signal based on the difference in concentration data between the at least two air samples. The motion unit controls the orientation and movement of the air pollution and nuisance investigation device based on the motion signal. This air pollution and nuisance investigation device can collect at least two air samples from different directions and locations, accurately and efficiently detect the concentration data of multiple air pollutants, and use this concentration data to control the device to move effectively toward the pollution source without having to traverse different directions, thus quickly locating the pollution source and improving investigation efficiency. This application also uses an AI prediction model to predict the type of pollution or nuisance source, replacing humans in performing dangerous tasks and eliminating potential occupational safety and health risks. In addition, the device is capable of continuous searching, which greatly improves the efficiency of finding pollution or nuisance sources. Attached Figure Description

[0043] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on the drawings without creative effort.

[0044] Figure 1 An exemplary block diagram of an air pollution and nuisance investigation device is shown;

[0045] Figure 2 An exemplary schematic diagram of a robotic dog for investigating air pollution and nuisance is shown;

[0046] Figure 3 An exemplary diagram illustrates the training and prediction process of an AI prediction model;

[0047] Figure 4 An exemplary diagram illustrates data annotation by placing a multi-parameter gas sensor at a known pollution source in multiple scenarios;

[0048] Figure 5 An example diagram of an integrated user interface for air pollution and nuisance investigations is shown.

[0049] Figure 6An exemplary flowchart of an air pollution and nuisance investigation method is shown;

[0050] Figure 7 An exemplary flowchart of another method for investigating air pollution and nuisance is shown;

[0051] Figure 8 An example diagram is shown comparing the concentration distribution maps of two air samples with a “fingerprint” concentration distribution map. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0053] Please refer to Figure 1 and Figure 2 , Figure 1 An exemplary structure of an air pollution and nuisance investigation device 100 is shown. Figure 2 An exemplary diagram illustrates a robotic dog used for investigating air pollution and nuisance. Figure 1 and Figure 2 As shown, the air pollution and nuisance investigation device 100 includes at least two air intake units 101 and a sensor unit 102, a central processing unit 103, and a motion unit 104 connected to each air intake unit 101. The at least two air intake units 101 are positioned at different locations and tilted in different directions, collecting at least two air samples from different directions and locations. The sensor unit 102 detects the air sample collected by its corresponding air intake unit to obtain concentration data of one or more air pollutants and / or nuisance sources in the air sample. The central processing unit 103 is connected to each sensor unit 102 and the motion unit 104, generating a motion signal based on the difference in concentration data between the at least two air samples and sending the motion signal to the motion unit 104. The motion unit 104 controls the air pollution and nuisance investigation device 100 to orient and move according to the motion signal.

[0054] In one embodiment, the air intake unit 101 includes a directional air intake and a forced air intake module. The directional air intake is tilted in a preset direction, and the forced air intake module controls the air sample to enter the corresponding sensor unit 102 at a preset flow rate. For example... Figure 2The air pollution and nuisance investigation robot includes two directional air inlets located on the left and right sides of a camera. The left directional air inlet faces left relative to the robot's front, and the right directional air inlet faces right relative to the robot's front. In one embodiment, the preset flow rate is 3 liters per minute. In another embodiment, the tilt direction and angle of the directional air inlets can be adjusted to more effectively infer the direction of the target air pollutant and / or nuisance source.

[0055] In one embodiment, sensor unit 102 includes a multi-parameter gas sensor capable of simultaneously measuring the concentration of multiple parameters, including: (1) hydrogen sulfide (H2S), (2) volatile organic compounds (VOCs), (3) particulate matter PM1.0, (4) particulate matter PM2.5, (5) particulate matter PM10, (6) nitrogen dioxide (NO2), (7) ozone (O3), (8) carbon monoxide (CO), (9) hydrocarbons (CxHy), and (10) ammonia (NH3).

[0056] In another embodiment, sensor unit 102 may simultaneously measure the concentration of one or more air pollutants and / or nuisance sources, including: (1) hydrogen sulfide (H2S), (2) volatile organic compounds (VOCs), (3) particulate matter (PM1.0, PM2.5, PM10), (4) nitrogen dioxide (NO2), (5) ozone (O3), (6) carbon monoxide (CO), (7) hydrocarbons (CxHy), (8) ammonia (NH3), (9) sulfur dioxide (SO2), (10) carbon dioxide (CO2), and (11) hydrogen chloride (HCl).

[0057] In one embodiment, sensor unit 102 is mounted on a stabilizer to reduce interference with the sensor, thereby enabling more accurate collection of concentration data of air pollutants and / or nuisance sources that can be compared among multiple sensors.

[0058] In one embodiment, the motion unit 104 may be implemented as a propulsion motor with a robotic arm, wheels, tracks, or other means, enabling the air pollution and nuisance investigation device 100 to move toward a pollution or nuisance source on land or water, across different types of terrain. In a particular embodiment, the air pollution and nuisance investigation device 100 is implemented in the form of a robotic dog, such as... Figure 2 As shown.

[0059] In another embodiment, the motion unit 104 is implemented in the form of a drone motor, enabling the air pollution and nuisance investigation device 100 to move in the air toward the pollution or nuisance source.

[0060] In yet another embodiment, the motion unit is implemented in the form of a ship's motor, enabling the air pollution and nuisance investigation device 100 to move on water toward the source of pollution or nuisance.

[0061] In one embodiment, the central processing unit 103 acquires concentration data of at least two air samples from at least two sensor units 102, calculates the concentration difference value of the at least two air samples, compares the concentration difference value with a preset difference threshold, and if the concentration difference value is greater than or equal to the difference threshold, determines a new direction of movement based on the air sample with the larger concentration and generates a motion signal based on the new direction of movement; if the concentration difference value is less than the difference threshold, acquires the current direction of movement and generates a motion signal based on the current direction of movement.

[0062] The method for calculating the concentration difference value can be set according to requirements. In one embodiment, the concentration difference value is calculated based on the concentrations of all types of air pollutants or nuisance sources. Specifically, a weighting coefficient is pre-set for each type of air pollutant or nuisance source. First, the sum of the concentrations of each air sample across all types of air pollutants or nuisance sources based on the weighting coefficient is calculated. Then, the difference between the sums of the concentrations of two air samples is calculated, and this difference is used as the concentration difference value between the two air samples. In this embodiment, a new direction of movement is determined based on the air sample with the higher concentration, i.e., a new direction of movement is determined based on the air sample with the higher concentration.

[0063] In another embodiment, the concentration difference value is calculated based on the concentration of a certain type of air pollutant or nuisance source. Specifically, the concentration difference between the two air samples for each type of air pollutant or nuisance source is first calculated, and then the largest concentration difference is selected as the concentration difference value between the two air samples. In this embodiment, the new direction of movement is determined based on the air sample with the larger concentration, i.e., the new direction of movement is determined based on the air sample with the larger concentration among the largest concentration differences. In a further embodiment, a certain type of air pollutant or nuisance source can also be pre-selected to calculate the concentration difference value.

[0064] In another embodiment, the central processing unit 103 has a pre-trained AI (Artificial Intelligence) prediction model built in. This AI prediction model can predict the type of pollution or nuisance source based on the concentration data of air pollutants and / or nuisance sources in air samples from unknown sources detected by the sensor unit 102. In this embodiment, for each type of pollution or nuisance source, a "fingerprint" concentration distribution map can be generated based on the typical concentrations of various air pollutants or nuisance sources. When using the air pollution and nuisance investigation device 100, the central processing unit 103 can generate a concentration distribution map based on the concentration of air pollutants or nuisance sources measured by each sensor unit 102, compare it with the "fingerprint" concentration distribution map corresponding to the pollution or nuisance source type predicted by the AI ​​prediction model, determine the concentration distribution map most similar to the "fingerprint" concentration distribution map, and determine a new direction of movement based on the air sample corresponding to the most similar concentration distribution map. For example, Figure 8 The concentration distribution map of air sample 1 is most similar to the "fingerprint" concentration distribution map, therefore the new direction of movement is the direction of air sample 1. In this embodiment, the air pollution and nuisance investigation device 100 can iteratively move towards the target pollution or nuisance source location based on the type of pollution or nuisance source predicted by AI.

[0065] When the concentration difference reaches the difference threshold, the central processing unit 103 controls the motion unit 104 to adjust the direction of the air pollution and nuisance investigation device 100, moving it towards the pollution source. When the concentration difference does not reach the difference threshold, the central processing unit 103 controls the motion unit 104 to control the air pollution and nuisance investigation device 100 to continue moving forward until the difference threshold is reached, at which point necessary directional adjustments are made. It is understood that the distance of each movement can be preset in the air pollution and nuisance investigation device 100.

[0066] In a further embodiment, the central processing unit 103 counts the number of times the air sample is detected. When the count reaches a preset number, it controls the air pollution and nuisance investigation device 100 to stop its investigation. After the air pollution and nuisance investigation device 100 completes a preset number of iterations and movements, it can perform manual analysis and adjustments before proceeding with further investigation. This is particularly useful when the air pollution and nuisance investigation device 100 reaches a dead end or needs to bypass obstacles to reach the pollution or nuisance source.

[0067] In one embodiment, the air pollution and nuisance investigation device 100 further includes a computer vision unit, which includes a 4G or 5G camera, a night vision camera, an infrared camera, and / or a lidar (LiDAR) system, etc. Figure 2The air pollution and nuisance investigation robot shown includes cameras and lidar. The central processing unit 103 is also used for autonomous navigation based on image data acquired by the computer vision unit, enabling the device to adapt to various terrains and avoid obstacles to reach the target location.

[0068] In one embodiment, the air pollution and nuisance investigation device 100 also includes a communication unit to allow seamless communication between devices or between devices and a control terminal equipped with a central command system. For example... Figure 2 The air pollution and nuisance investigation robot shown includes a 5G router and a 5G antenna. High-speed 5G connectivity supports the real-time transmission of large amounts of data, which is crucial for real-time monitoring and assessment.

[0069] In one embodiment, the central command system corresponds to an integrated user interface. For example... Figure 5 As shown, the integrated user interface synchronously displays real-time environmental images captured by air pollution and nuisance investigation equipment. It also displays the concentration data of air pollutants and / or nuisance sources detected in real-time by two multi-parameter gas sensors (left and right), prediction results from the AI ​​prediction model based on the concentration data, and multiple operation buttons on the real-time environmental images. Specifically, the directional operation button in the lower left corner controls the robot dog's movement direction; the function operation button in the bottom center is used to set automatic tasks to start or stop, view videos, view photos, or browse, etc.; the directional operation button in the lower right corner controls the shooting angle; and the function operation button in the upper left corner is used to set the robot dog's posture, speed, and movement mode.

[0070] The solution of this application can also be implemented as an air pollution and nuisance investigation system, including an air pollution and nuisance investigation device 100 and a control terminal. The air pollution and nuisance investigation device 100 and the control terminal establish a communication connection, and the air pollution and nuisance investigation device transmits the concentration data of at least two air samples to the control terminal in real time. The control terminal implements the aforementioned integrated user interface. Through this integrated user interface, the user can monitor and evaluate the environment in which the air pollution and nuisance investigation device is located, the detected concentration data, and the prediction results in real time, and manually adjust the air pollution and nuisance investigation device through corresponding operation buttons. In one embodiment, the control terminal is a tablet computer. Figure 2 Taking the air pollution and nuisance investigation robot dog as an example, the air pollution and nuisance investigation system consists of a robot unit, a central control platform (mounted on the robot dog, including an air intake unit, sensor unit, computer vision unit, communication unit, and central processing unit), and a control terminal with a central command system. The devices in the system are interconnected to form an Internet of Things network, and the units work together in the network, which improves the overall efficiency of the system.

[0071] like Figure 3 The diagram shows the training and prediction process of an AI prediction model. The training process includes:

[0072] Step 1: Collect, test, and label air samples from various known types of pollution or nuisance sources to obtain labeled data.

[0073] Specifically, devices equipped with air sensors or the air pollution and nuisance investigation equipment described in this application can be deployed to various known pollution source locations to collect concentration data containing one or more air pollutants or nuisance sources, and these concentration data can be labeled to obtain labeled data. For example... Figure 4 The diagram illustrates data annotation by placing multi-parameter gas sensors at known pollution sources in multiple scenarios. In one embodiment, pollution or nuisance sources used to train the AI ​​prediction model may include small-scale private wastewater treatment plants, sewage sewers, construction sites, pig farms, poultry farms, and ambient air (i.e., the baseline environment). In another embodiment, pollution or nuisance sources used to train the AI ​​prediction model may further include large-scale public wastewater treatment plants, drainage culverts, drainage outlets, dog kennels, large construction sites, road construction sites, chemical waste storage areas, chemical waste collection points, chemical waste treatment facilities, landfills, remediated landfills, metal plating plants, chimneys, restaurant exhaust vents, and recycling stations.

[0074] Step 2: Input the labeled data into the AI ​​prediction model built using machine learning algorithms for training to obtain the trained AI prediction model.

[0075] In one embodiment, a gradient boosting decision tree machine learning method is used to train the AI ​​prediction model. During training, the AI ​​prediction model analyzes the patterns and concentration characteristics of various types of pollution or nuisance sources based on labeled data, thereby enabling the AI ​​prediction model to identify and distinguish different pollution or nuisance sources.

[0076] The prediction process includes:

[0077] Step 1: Collect and detect air samples from unknown types of pollution or nuisance sources to obtain concentration data of unknown types of pollution or nuisance sources, and input the concentration data into the trained AI prediction model.

[0078] Step 2: The AI ​​prediction model analyzes the concentration data of unknown types of pollution or nuisance sources to predict the type of pollution or nuisance source.

[0079] Specifically, the air pollution and nuisance investigation device 100 of this application can be used to collect and detect air samples from unknown types of pollution or nuisance sources, obtaining concentration data of air pollutants or nuisance sources in at least two air samples. The concentration data of all or part of the at least two air samples are input into a trained AI prediction model. The AI ​​prediction model analyzes the concentration data of all or part of the at least two air samples to predict the type of pollution or nuisance source. In particular, the concentration data of the air sample with the highest concentration among the at least two air samples is input into the trained AI prediction model. The AI ​​prediction model analyzes the concentration data of the air sample with the higher concentration to predict the type of pollution or nuisance source. Figure 3 As shown, the type of pollution or nuisance source is sewage sewers, and the suggested source is the Tai Wai culvert in Hong Kong.

[0080] In one embodiment, the concentration data of air pollutants or nuisance sources collected by the air pollution and nuisance investigation device 100, as well as the types of pollution or nuisance sources confirmed by humans, can be further used as annotation data for the AI ​​prediction model, thereby improving the AI ​​prediction model for future use.

[0081] In one embodiment, a geofence can be constructed in the patrol area of ​​the air pollution and nuisance investigation device 100 or a search range can be set in the autonomously navigable air pollution and nuisance investigation device 100 to narrow the area for searching the location of pollution or nuisance sources.

[0082] In one embodiment, the air pollution and nuisance investigation device 100 or the air pollution and nuisance investigation system can access key information from multiple sources, including: (1) known air emission inventories from environmental protection departments and agencies; (2) real-time meteorological data; and (3) location or topographic data. This accessibility of the data enables AI predictive models to better identify and classify pollution or nuisance sources (such as sites, factories, or public facilities) and their names and addresses, thereby further improving the efficiency and effectiveness of navigating to the location of pollution or nuisance sources and identifying the types of pollution or nuisance sources.

[0083] The air pollution and nuisance investigation device of this application can collect at least two air samples from different directions and locations, accurately and efficiently detect the concentration data of multiple air pollutants, and use this concentration data to control the device to effectively move towards the pollution source without having to turn in different directions to traverse the area, thus quickly locating the pollution source and improving investigation efficiency. When combined with AI analysis methods for determining the type of pollution or nuisance source and directional air intake methods for determining the location of pollution or nuisance sources, comprehensive and holistic analysis of pollution or nuisance sources can be achieved without human intervention, replacing humans in performing dangerous tasks and eliminating potential occupational safety and health risks. In addition, the device can perform continuous searches, greatly improving the search efficiency for pollution or nuisance sources.

[0084] Please refer to Figure 6 , Figure 6 An exemplary flowchart of an air pollution and nuisance investigation method is shown, which is applied to an air pollution and nuisance investigation device. In some embodiments, the air pollution and nuisance investigation device here can be used as... Figures 1-2 The structure shown is implemented in detail in the above embodiments, and will not be repeated here.

[0085] like Figure 6 As shown, the methods for investigating air pollution and nuisance include:

[0086] Step S601: Collect at least two air samples from different directions and different locations.

[0087] Specifically, at least two air samples are collected simultaneously through at least two air intake units tilted in different directions on the air pollution and nuisance investigation equipment.

[0088] Step S602: Detect at least two air samples to obtain concentration data of one or more air pollutants and / or nuisance sources in the at least two air samples.

[0089] Specifically, at least two air samples are detected by at least two sensor units on the air pollution and nuisance investigation equipment to obtain concentration data of one or more air pollutants and / or nuisance sources in the at least two air samples.

[0090] Step S603: Generate a motion signal based on the difference in concentration data from at least two air samples.

[0091] In one embodiment, step S603 is implemented as follows: obtain the concentration difference value of at least two air samples based on the concentration data in at least two air samples, compare the concentration difference value with a preset difference threshold; if the concentration difference value is greater than or equal to the difference threshold, determine a new direction of movement based on the air sample with the larger concentration, and generate a motion signal based on the new direction of movement; if the concentration difference value is less than the difference threshold, obtain the current direction of movement, and generate a motion signal based on the current direction of movement.

[0092] The method for calculating the concentration difference value can be set according to requirements. In one embodiment, the concentration difference value is calculated based on the concentrations of all types of air pollutants or nuisance sources. Specifically, a weighting coefficient is pre-set for each type of air pollutant or nuisance source. First, the sum of the concentrations of each air sample across all types of air pollutants or nuisance sources based on the weighting coefficient is calculated. Then, the difference between the sums of the concentrations of two air samples is calculated, and this difference is used as the concentration difference value between the two air samples. In this embodiment, a new direction of movement is determined based on the air sample with the higher concentration, i.e., a new direction of movement is determined based on the air sample with the higher concentration.

[0093] In another embodiment, the concentration difference value is calculated based on the concentration of a certain type of air pollutant or nuisance source. Specifically, the concentration difference between the two air samples for each type of air pollutant or nuisance source is first calculated, and then the largest concentration difference is selected as the concentration difference value between the two air samples. In this embodiment, the new direction of movement is determined based on the air sample with the larger concentration, i.e., the new direction of movement is determined based on the air sample with the larger concentration among the largest concentration differences.

[0094] Step S604: Control the air pollution and nuisance investigation equipment to orient and move according to the motion signal.

[0095] Specifically, the air pollution and nuisance investigation equipment is first oriented according to the direction of movement in the motion signal, and then the air pollution and nuisance investigation equipment is controlled to move a preset distance in that direction of movement.

[0096] Please refer to Figure 7 , Figure 7 An exemplary flowchart of another method for investigating air pollution and nuisance is shown, which is applied to an air pollution and nuisance investigation device. In some embodiments, the air pollution and nuisance investigation device here can be used as... Figures 1-2 The structure shown is implemented in detail in the above embodiments, and will not be repeated here.

[0097] like Figure 7 As shown, the methods for investigating air pollution and nuisance include:

[0098] Step S701: Collect air samples from various known types of pollution or nuisance sources and test them to obtain concentration data of air pollutants or nuisance sources in various known types of pollution or nuisance sources.

[0099] Step S702: Train the AI ​​prediction model constructed using machine learning algorithms based on the concentration data of various known types of pollution or nuisance sources to obtain the trained AI prediction model.

[0100] Please refer to the foregoing description for specific data collection and training methods. The trained AI prediction model can identify the characteristics of pollution or nuisance sources based on the concentration data of air pollutants or nuisance sources in unknown types of pollution or nuisance sources, thereby determining the type of pollution or nuisance source.

[0101] Step S703: Collect at least two air samples from different directions and different locations.

[0102] Step S704: Detect at least two air samples to obtain concentration data of one or more air pollutants and / or nuisance sources in the at least two air samples.

[0103] Step S705: Use the trained AI prediction model to analyze the concentration data of some or all of the air samples in at least two air samples to predict the type of pollution or nuisance source.

[0104] like Figure 5 As shown, the prediction results obtained based on the trained AI prediction model are: the probability that the pollution or nuisance source is sewage sewer is 78%, the probability that it is sewage treatment plant is 22%, and the suggested source is the Tai Wai culvert in Hong Kong.

[0105] Step S706: Generate a motion signal based on the difference in concentration data from at least two air samples.

[0106] Step S707: Control the air pollution and nuisance investigation equipment to orient and move according to the motion signal.

[0107] Step S708: Determine whether the number of detections has reached the preset number. If not, proceed to step S703; if yes, proceed to step S709.

[0108] In one embodiment, the method further includes counting the number of times the air sample is detected before step S708.

[0109] Step S709: Control air pollution and stop the investigation equipment from working.

[0110] The air pollution and nuisance investigation method provided in this application can collect at least two air samples from different directions and locations, accurately and efficiently detect the concentration data of multiple air pollutants, and use this concentration data to control the equipment to effectively move towards the pollution source without having to turn to different directions to traverse the area, thus quickly locating the pollution source and improving investigation efficiency. This application also uses an AI prediction model to predict the type of pollution or nuisance source, replacing humans in performing dangerous tasks, thereby eliminating potential occupational safety and health risks.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them; under the concept of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in a corresponding order, and there are many other variations of different aspects of this application as described above, which are not provided in detail for the sake of brevity; although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. An air pollution and nuisance investigation device, characterized in that, include: At least two air intake units and a sensor unit connected to each air intake unit; A central processing unit; as well as A motion unit; The at least two air intake units are tilted in different directions to collect at least two air samples from different directions and positions. The sensor unit detects the air samples collected by its corresponding air intake unit to obtain concentration data of one or more air pollutants and / or nuisance sources in the air samples. The central processing unit generates a motion signal based on the difference in concentration data of the at least two air samples and sends the motion signal to the motion unit. The motion unit controls the air pollution and nuisance investigation equipment to orient and move according to the motion signal.

2. The air pollution and nuisance investigation equipment according to claim 1, characterized in that, The central processing unit generates motion signals based on the differences in concentration data from at least two air samples, including: The central processing unit obtains the concentration difference value of the at least two air samples based on the concentration data of the at least two air samples, and compares the concentration difference value with a preset difference threshold. If the concentration difference value is greater than or equal to the difference threshold, a new direction of movement is determined based on the air sample with a higher concentration, and the motion signal is generated according to the new direction of movement. If the concentration difference value is less than the difference threshold, the current movement direction is obtained, and the motion signal is generated based on the current movement direction.

3. The air pollution and nuisance investigation equipment according to claim 1, characterized in that, The central processing unit uses a trained AI prediction model to analyze the concentration data of some or all of the air samples in the at least two air samples to predict the type of pollution or nuisance source.

4. The air pollution and nuisance investigation equipment according to claim 3, characterized in that, Each type of pollution or nuisance source corresponds to a "fingerprint" concentration distribution map, which is generated based on the typical concentrations of various air pollutants or nuisance sources corresponding to that type of pollution or nuisance source. The central processing unit generates motion signals based on the differences in concentration data from at least two air samples, including: The central processing unit obtains the concentration distribution maps corresponding to the at least two air samples based on the concentration data of the at least two air samples; From the concentration distribution maps corresponding to the at least two air samples, determine the concentration distribution map that is most similar to the "fingerprint" concentration distribution map corresponding to the predicted type of pollution or nuisance source; A new direction of movement is determined based on the air sample corresponding to the most similar concentration distribution map, and the motion signal is generated based on the new direction of movement.

5. The air pollution and nuisance investigation equipment according to claim 1, characterized in that, The central processing unit counts the number of times the air sample is detected, and when the count reaches a preset number, it controls the air pollution and nuisance investigation equipment to stop its investigation.

6. The air pollution and nuisance investigation equipment according to any one of claims 1 to 5, characterized in that, The air intake unit includes a directional air intake port and a forced air intake module. The directional air intake port is tilted in a preset direction, and the forced air intake module controls the air sample to enter the corresponding sensor unit at a preset flow rate.

7. The air pollution and nuisance investigation equipment according to any one of claims 1 to 5, characterized in that, The motion unit includes any one of a propulsion motor, a drone motor, or a marine motor.

8. The air pollution and nuisance investigation equipment according to any one of claims 1 to 5, characterized in that, The air pollution and nuisance investigation equipment is implemented in the form of a robot dog.

9. An air pollution and nuisance investigation system, characterized in that, It includes the air pollution and nuisance investigation device as described in any one of claims 1 to 8, and a control terminal that is communicatively connected to the air pollution and nuisance investigation device.

10. A method for investigating air pollution and nuisance, characterized in that, The method is applied to air pollution and nuisance investigation equipment, including: Collect at least two air samples from different directions and locations; The at least two air samples are tested separately to obtain concentration data of one or more air pollutants and / or sources of disturbance in the at least two air samples; A motion signal is generated based on the difference in concentration data from the at least two air samples; The air pollution and nuisance investigation equipment is controlled to orient and move according to the motion signal.

11. The method according to claim 10, characterized in that, The generation of motion signals based on the differences in concentration data from the at least two air samples includes: The concentration difference value of the at least two air samples is obtained based on the concentration data in the at least two air samples, and the concentration difference value is compared with a preset difference threshold. If the concentration difference value is greater than or equal to the difference threshold, a new direction of movement is determined based on the air sample with a higher concentration, and the motion signal is generated according to the new direction of movement. If the concentration difference value is less than the difference threshold, the current movement direction is obtained, and the motion signal is generated based on the current movement direction.

12. The method according to claim 10, characterized in that, After obtaining the concentration data of one or more air pollutants and / or nuisance sources in the at least two air samples, the method further includes: The concentration data of some or all of the air samples in the at least two air samples are analyzed using a trained AI prediction model to predict the type of pollution or nuisance source.

13. The method according to claim 12, characterized in that, Each type of pollution or nuisance source corresponds to a "fingerprint" concentration distribution map, which is generated based on the typical concentrations of various air pollutants or nuisance sources corresponding to the type of pollution or nuisance source. Generating motion signals based on the differences in concentration data from at least two air samples includes: Based on the concentration data of the at least two air samples, obtain the concentration distribution maps corresponding to the at least two air samples respectively; From the concentration distribution maps corresponding to the at least two air samples, determine the concentration distribution map that is most similar to the "fingerprint" concentration distribution map corresponding to the predicted type of pollution or nuisance source; A new direction of movement is determined based on the air sample corresponding to the most similar concentration distribution map, and the motion signal is generated based on the new direction of movement.

14. The method according to claim 12 or 13, characterized in that, The process of collecting at least two air samples from different directions and locations also includes: Air samples were collected and tested from various known types of pollution or nuisance sources to obtain concentration data of air pollutants or nuisance sources in various known types of pollution or nuisance sources; The AI ​​prediction model, constructed using machine learning algorithms, is trained based on concentration data of various known types of pollution or nuisance sources to obtain the trained AI prediction model.

Images