A UAV environmental monitoring and disaster warning system and method

Abstract

The present invention discloses a UAV environment monitoring and disaster warning system and method, which relates to the field of environmental monitoring, including: using UAVs to integrate corresponding monitoring equipment and MEC servers as mobile carriers for edge computing of environmental monitoring; cleaning and standardizing data; extracting the characteristics of abnormal events, and establishing an environmental abnormality feature library; calculating the static interference offset coefficient of environmental data in the monitoring area; determining the error factors brought to the detection results by different environments and UAV dynamic factors; establishing an environmental monitoring dynamic-static input matrix; judging whether there are abnormalities in the monitoring area environment based on a neural network model, and judging the abnormality level; establishing a ground receiving platform for receiving, displaying and storing environmental monitoring data. The advantages of the present invention are: by establishing a dynamic-static monitoring mode of a UAV, the accuracy and reliability of environmental monitoring data are effectively improved, thereby greatly improving the accuracy of environmental abnormality judgment.

Description

Technical Field

[0001] The present invention relates to the field of environmental monitoring, and in particular to an unmanned aerial vehicle (UAV) environmental monitoring and disaster early warning system and method. Background Art

[0002] Environmental monitoring and disaster warning mainly involves collecting environmental data from the monitored area through various monitoring equipment, and through comprehensive analysis of the collected environmental data, discovering possible environmental problems in the current area, making timely warnings, and taking solutions to reduce or avoid the occurrence and expansion of environmental hazards. With the development of drone technology, drones are combined with environmental monitoring equipment to effectively improve the efficiency and convenience of environmental monitoring, taking advantage of their strong maneuverability and convenient environmental monitoring. It also provides the possibility for environmental monitoring in complex terrain and sparsely populated areas, and realizes large-scale, no-dead-angle environmental monitoring. Therefore, it is of great significance to combine drones with environmental monitoring equipment to conduct large-scale, no-dead-angle monitoring of the environment.

[0003] Existing UAV environmental monitoring and disaster warning technologies mainly focus on using UAVs to carry monitoring equipment for collection, and using ground receiving equipment to receive the collected environmental data using wireless transmission technology. However, long-distance data transmission and the influence of the external environment and the UAV itself result in low quality and accuracy of the collected environmental data. When the system performs analysis and comprehensive evaluation, misjudgment and low monitoring efficiency are prone to occur, resulting in a waste of manpower and material resources and untimely response to environmental problems, which in turn causes the further expansion of environmental problems and is not conducive to the long-term development of environmental monitoring and disaster warning. Summary of the invention

[0004] In order to solve the above technical problems, a UAV environmental monitoring and disaster warning system and method are provided. This technical solution solves the problems raised in the above background technology, such as long-distance data transmission and the influence of the external environment and the UAV itself, resulting in low quality and accuracy of the collected environmental data. When the system performs analysis and comprehensive evaluation, misjudgment and low monitoring efficiency are prone to occur, resulting in waste of manpower and material resources and untimely response to environmental problems, thereby causing further expansion of environmental problems, which is not conducive to the long-term development of environmental monitoring and disaster warning.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A method for unmanned aerial vehicle environment monitoring and disaster warning, characterized by comprising:

[0007] According to the requirements of the monitoring environment, drones are used to integrate corresponding monitoring equipment and MEC servers as mobile carriers for environmental monitoring edge computing;

[0008] Clean and standardize the environmental data collected by monitoring equipment;

[0009] Based on the field detection and historical data of abnormal events in the monitoring area, the characteristics of abnormal events are extracted and an environmental abnormality feature library is established;

[0010] Calculate the static interference deviation coefficient of the environmental data in the monitoring area based on the normal standard value of the environment and the test results of the actual test flight;

[0011] According to the monitoring environment and UAV dynamic analysis, determine the error factors caused by different environmental and UAV dynamic factors to the detection results;

[0012] Based on the static interference offset coefficient and error factor, the dynamic-static input matrix of environmental monitoring is established;

[0013] According to the dynamic-static input matrix of environmental monitoring, based on the neural network model, it is determined whether there is any abnormality in the environment of the monitoring area, and the abnormality level is determined;

[0014] Establish a ground receiving platform to receive, display and store environmental monitoring data, and indicate the level of environmental abnormality through small lights of different colors.

[0015] Preferably, the cleaning and standardization of the environmental data collected by the monitoring equipment specifically includes:

[0016] Based on the first-order Kalman filter algorithm, the received data is subjected to real-time denoising and filtering processing;

[0017] Perform bit reduction processing on the denoising filtered data;

[0018] For the data after decimal point reduction, standardization is performed to remove decimal points;

[0019] The drop processing expression is:

[0020] In the formula, It is the environmental monitoring data after downgrading. is the environmental monitoring data after denoising and filtering. The part with the same value as the higher digit of the environmental monitoring data;

[0021] The decimal point removal standardization processing expression is: In the formula, This is the environmental monitoring data after decimal point standardization. It is the environmental monitoring data after downgrading. is a constant term.

[0022] Preferably, extracting the features of abnormal events based on field detection and historical data of abnormal events in the monitoring area and establishing an environmental abnormality feature library specifically includes:

[0023] Based on the environmental field detection values ​​and external appearances when abnormal events occur in the monitoring area, a large number of environmental abnormal environment detection values ​​and external appearances similar to those when abnormal events occur in the monitoring area are obtained through big data;

[0024] Through horizontal comparison, the common characteristic items of environmental abnormal events are extracted, and the characteristic items are digitized and standardized;

[0025] The processed environmental anomaly feature items are classified and divided according to their nature, the disaster risk level is set, and an environmental anomaly feature database is established.

[0026] Preferably, the step of calculating the static interference offset coefficient of the environmental data in the monitoring area according to the normal standard value of the environment and the detection result of the actual test flight specifically includes:

[0027] Through professional equipment, we can comprehensively detect the environmental data of the monitoring area and compare it with the environmental abnormality feature library, establish the normal standard value of the environment, and set the stable range;

[0028] While conducting professional inspections, use drones equipped with monitoring equipment to statically collect environmental data at the same location and area;

[0029] Calculate the static interference deviation coefficient of the environmental data in the monitoring area based on the normal standard value of the environment and the test results of the actual test flight;

[0030] The static interference offset coefficient expression of the monitoring area environmental data is: In the formula, To monitor the static interference offset coefficient of regional environmental data, is the error balance factor, is a constant, For Moment Normal standard value of the environment, For Moment UAV environmental monitoring values, is the number of data groups.

[0031] Preferably, the determination of the error factors caused by different environmental and UAV dynamic factors to the detection results based on the monitoring environment and the UAV dynamic analysis specifically includes:

[0032] Based on historical data or relevant literature, through big data analysis, obtain the relevant influencing factors of drones affecting environmental monitoring in a dynamic state;

[0033] According to the types of influencing factors, the linear regression equation of the error factor of UAV dynamic environment monitoring is established;

[0034] Through field test experiments or simulation software, the dynamic monitoring of UAVs can be simulated to obtain the dynamic environmental monitoring data of UAVs;

[0035] According to the normal standard value of the environment and the dynamic environmental monitoring data of the UAV, an error factor evaluation model is established based on the least squares method and gradient descent algorithm;

[0036] According to the error factor evaluation model, the acceptable range of monitoring data is obtained through big data analysis, and the threshold is set;

[0037] Through the error factor evaluation model, the received UAV dynamic environment monitoring data is evaluated in real time to determine whether the calculated error value exceeds the threshold.

[0038] If yes, it means that the environmental monitoring data of the drone in the current state is distorted. If continuous distortion occurs, it means that the current drone cannot complete the environmental monitoring task and needs to return for adjustment. If no, it means that the drone meets the environmental monitoring requirements in the current state.

[0039] The linear regression equation of the error factor of the UAV dynamic environment monitoring is: In the formula, is the calculated value of the error factor of the UAV dynamic environment monitoring, To influence environmental monitoring The parameters of the influencing factors are: is the error term, is a constant, To influence environmental monitoring Influencing factors, among which The value of is equal to 1, is the number of factors affecting environmental monitoring;

[0040] The least squares method and gradient descent algorithm expressions are: In the formula, is the optimal value of the error factor calculated by the least squares method, is the number of environmental monitoring data sample groups, is the error factor of the UAV dynamic environment monitoring Calculated values, For the Normal standard value of the environment, To influence environmental monitoring The update parameters of the impact factors are To influence environmental monitoring The parameters of the influencing factors, is the learning rate of gradient descent, For the Impact Factor Environmental monitoring data.

[0041] Preferably, the establishment of the environment monitoring dynamic-static input matrix based on the static interference offset coefficient and the error factor specifically includes:

[0042] According to the operation status of the UAV, establish a dynamic-static monitoring mode for environmental monitoring;

[0043] According to the change of the speed of the drone in the hovering state, set the threshold of whether the drone is moving;

[0044] The dynamic and static state of the drone is determined according to the speed monitoring value of the drone. By comparing whether the speed monitoring value of the drone is greater than the threshold, if so, it means that the drone is in motion and the dynamic mode of environmental monitoring is adopted. If not, it means that the drone is in hovering state and the static mode of environmental monitoring is adopted.

[0045] According to the static interference offset coefficient and error factor, the dynamic-static input matrix of environmental monitoring is established;

[0046] The environmental monitoring dynamic-static input matrix expression is: In the formula, Dynamic-static input matrix for environmental monitoring, Screening matrix for dynamic-static mode of environmental monitoring, is the correction value for the UAV static environment monitoring, Correction value for UAV dynamic environment monitoring.

[0047] Preferably, the step of judging whether there is an abnormality in the environment of the monitoring area based on the dynamic-static input matrix of the environmental monitoring and the neural network model, and judging the abnormality level specifically includes:

[0048] According to the dynamic-static input matrix of environmental monitoring, based on the neural network model, the received data is processed to determine whether there is any abnormality in the environment of the monitoring area;

[0049] By comparing the environmental monitoring data determined as abnormal by the neural network model with the environmental abnormality feature library, the type of environmental abnormal behavior is queried and the abnormality level is determined;

[0050] Based on whether there is an abnormality and the abnormality level, a decision is made whether to send abnormal information to the ground receiving platform. If so, an abnormal alarm and abnormality type are issued to the ground platform, and the abnormal information is recorded through a storage device. If not, it means that there is no environmental problem in the area and no abnormal alarm is issued.

[0051] Furthermore, a UAV environment monitoring and disaster warning system is proposed, which is used to implement the above-mentioned UAV environment monitoring and disaster warning method, including:

[0052] A hardware setting module, which is used to integrate corresponding monitoring equipment and MEC servers using drones as mobile carriers for edge computing of environmental monitoring according to the requirements of the monitoring environment;

[0053] A data processing module, which is used to clean and standardize the environmental data collected by the monitoring equipment; extract the characteristics of abnormal events based on the field detection and historical data of abnormal events in the monitoring area, and establish an environmental abnormality feature library;

[0054] A dynamic-static modeling module, which is used to calculate the static interference offset coefficient of the environmental data in the monitoring area according to the normal standard value of the environment and the test results of the actual test flight; and to determine the error factors caused by different environmental and UAV dynamic factors to the test results according to the monitoring environment and the dynamic analysis of the UAV;

[0055] An anomaly detection module, the anomaly detection module is used to establish an environmental monitoring dynamic-static input matrix based on the static interference offset coefficient and the error factor; according to the environmental monitoring dynamic-static input matrix, based on the neural network model, determine whether there is an abnormality in the monitoring area environment, and determine the abnormality level;

[0056] A ground receiving module is used to establish a ground receiving platform for receiving, displaying and storing environmental monitoring data, and indicating the level of environmental anomalies through small lights of different colors.

[0057] Preferably, the data processing module specifically includes:

[0058] A data processing unit, the data processing unit is used to clean and standardize the environmental data collected by the monitoring equipment;

[0059] An abnormal feature library unit is established, and the abnormal feature library unit is used to extract the characteristics of abnormal events and establish an environmental abnormal feature library based on field detection and historical data of abnormal events in the monitoring area.

[0060] Preferably, the dynamic-static modeling module specifically includes:

[0061] A static modeling unit, wherein the static modeling unit is used to calculate a static interference offset coefficient of the environmental data of the monitoring area according to a normal standard value of the environment and a detection result of an actual test flight;

[0062] A dynamic modeling unit, which is used to determine the error factors caused by different environmental and UAV dynamic factors to the detection results according to the monitoring environment and the UAV dynamic analysis;

[0063] The anomaly detection module specifically includes:

[0064] Establishing an input matrix unit, wherein the input matrix establishing unit is used to establish an environmental monitoring dynamic-static input matrix based on a static interference offset coefficient and an error factor;

[0065] The abnormality judgment unit is used to judge whether there is an abnormality in the monitoring area environment and to judge the abnormality level based on the dynamic-static input matrix of the environmental monitoring and the neural network model.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] The drone is integrated with corresponding monitoring equipment and MEC server as a mobile carrier of environmental monitoring edge computing. When controlling the drone for environmental monitoring, the monitoring data close to the source of data is automatically processed and analyzed, which effectively improves the reliability and accuracy of the data and prevents the loss and damage of data during the transmission process, thereby wasting manpower and material resources. Secondly, by analyzing the static and dynamic environmental monitoring errors of the drone during the environmental monitoring process, a dynamic-static mode of drone environmental monitoring is established, and through the environmental monitoring dynamic-static mode screening matrix, according to the speed detection of the drone, a suitable drone environmental monitoring mode is screened out, and an environmental monitoring dynamic-static mode screening matrix is ​​established. The static input matrix uses the environmental monitoring dynamic-static input matrix and the neural network model to determine whether there are any abnormalities in the environment of the monitoring area and the level of abnormalities. Finally, the ground receiving platform is used to receive, display and store environmental monitoring data. Small lights of different colors are used to indicate the level of environmental abnormalities, thereby achieving the purpose of environmental monitoring and disaster warning for the monitored area by drones. It effectively improves the efficiency, convenience, accuracy and reliability of environmental monitoring, greatly improves the accuracy of environmental anomaly judgment, ensures timely warnings and solutions, and provides the possibility for environmental monitoring in complex terrain and sparsely populated areas, realizing large-scale and no-dead-angle environmental monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a flow chart of a method for UAV environment monitoring and disaster warning of the present invention;

[0069] Figure 2 A flow chart of cleaning and standardizing the environmental data collected by the monitoring equipment according to the present invention;

[0070] Figure 3The present invention is a flow chart for calculating the static interference deviation coefficient of the monitoring area environmental data according to the normal standard value of the environment and the detection results of the actual test flight;

[0071] Figure 4 This is a flow chart of the present invention for determining the error factors caused by different environmental and UAV dynamic factors to the detection results based on the monitoring environment and UAV dynamic analysis;

[0072] Figure 5 The invention provides a flow chart for establishing a dynamic-static input matrix for environmental monitoring based on static interference offset coefficient and error factor;

[0073] Figure 6 This is a flow chart of the present invention for judging whether there is any abnormality in the environment of the monitoring area and judging the abnormality level based on the dynamic-static input matrix of the environment monitoring and the neural network model. DETAILED DESCRIPTION

[0074] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0075] Reference Figure 1 As shown, a method for UAV environment monitoring and disaster warning includes:

[0076] According to the requirements of the monitoring environment, drones are used to integrate corresponding monitoring equipment and MEC servers as mobile carriers for environmental monitoring edge computing;

[0077] Clean and standardize the environmental data collected by monitoring equipment;

[0078] Based on the field detection and historical data of abnormal events in the monitoring area, the characteristics of abnormal events are extracted and an environmental abnormality feature library is established;

[0079] Calculate the static interference deviation coefficient of the environmental data in the monitoring area based on the normal standard value of the environment and the test results of the actual test flight;

[0080] According to the monitoring environment and UAV dynamic analysis, determine the error factors caused by different environmental and UAV dynamic factors to the detection results;

[0081] Based on the static interference offset coefficient and error factor, the dynamic-static input matrix of environmental monitoring is established;

[0082] According to the dynamic-static input matrix of environmental monitoring, based on the neural network model, it is determined whether there is any abnormality in the environment of the monitoring area, and the abnormality level is determined;

[0083] Establish a ground receiving platform to receive, display and store environmental monitoring data, and indicate the level of environmental abnormality through small lights of different colors.

[0084] It can be explained that this solution uses drones to integrate corresponding monitoring equipment and MEC servers as mobile carriers for environmental monitoring edge computing. When controlling drones for environmental monitoring, it automatically processes and analyzes monitoring data close to the source of data generation, effectively improving the reliability and accuracy of data, and preventing data loss and damage during transmission, thereby wasting manpower and material resources. Secondly, by analyzing the static and dynamic environmental monitoring errors of drones during environmental monitoring, a dynamic-static mode of drone environmental monitoring is established, and through the environmental monitoring dynamic-static mode screening matrix, according to the speed detection of the drone, a suitable drone environmental monitoring mode is screened out, and an environmental monitoring dynamic-static input matrix is ​​established. Through the environmental monitoring dynamic-static input matrix, based on the neural network model, it is determined whether there is an abnormality in the monitoring area environment and the level of abnormality. Finally, the ground receiving platform is used to receive, display and store environmental monitoring data, and the level of environmental abnormality is indicated by small lights of different colors.

[0085] Reference Figure 2 As shown, the cleaning and standardization of the environmental data collected by the monitoring equipment specifically includes:

[0086] Based on the first-order Kalman filter algorithm, the received data is subjected to real-time denoising and filtering processing;

[0087] Perform bit reduction processing on the denoising filtered data;

[0088] For the data after decimal point reduction, standardization is performed to remove decimal points;

[0089] The drop processing expression is:

[0090] In the formula, It is the environmental monitoring data after downgrading. is the environmental monitoring data after denoising and filtering. The part with the same value as the higher digit of the environmental monitoring data;

[0091] The decimal point removal standardization processing expression is: In the formula, This is the environmental monitoring data after decimal point standardization. It is the environmental monitoring data after downgrading. is a constant term.

[0092] It can be explained that, due to the non-uniform units of environmental data collected by different environmental monitoring devices during the environmental monitoring process, as well as the influence of decimal points, the processing of environmental data at the software level is more complicated. This solution uses a first-order Kalman filtering algorithm to reduce noise and filter the collected environmental monitoring data to reduce the impact of abnormal peaks, and reduces the digit and decimal point problems of the environmental monitoring data through digit reduction and decimal point standardization processing, thereby improving the system's processing efficiency of environmental monitoring data.

[0093] Reference Figure 3 As shown, the calculation of the static interference offset coefficient of the environmental data in the monitoring area according to the normal standard value of the environment and the test results of the actual test flight specifically includes:

[0094] Through professional equipment, we can comprehensively detect the environmental data of the monitoring area and compare it with the environmental abnormality feature library, establish the normal standard value of the environment, and set the stable range;

[0095] While conducting professional inspections, use drones equipped with monitoring equipment to statically collect environmental data at the same location and area;

[0096] Calculate the static interference deviation coefficient of the environmental data in the monitoring area based on the normal standard value of the environment and the test results of the actual test flight;

[0097] The static interference offset coefficient expression of the monitoring area environmental data is: In the formula, To monitor the static interference offset coefficient of regional environmental data, is the error balance factor, is a constant, For Moment Normal standard value of the environment, For Moment UAV environmental monitoring values, is the number of data groups.

[0098] It can be explained that by analyzing the differences between the environmental monitoring data collected by the UAV in a hovering state and the environmental data professionally detected at the same time, the static interference offset coefficient of the environmental data in the monitoring area is calculated using the variance expression. This coefficient can be used to effectively counter-optimize the static environmental monitoring data of the UAV, thereby improving the accuracy and reliability of the environmental monitoring data of the UAV in static mode. In particular, in order to ensure the accuracy of the static interference offset coefficient, one or more test experiments are required when implementing environmental monitoring data collection, and the static interference offset coefficient value is adjusted according to the test results.

[0099] Reference Figure 4As shown, according to the monitoring environment and the UAV dynamic analysis, the error factors caused by different environmental and UAV dynamic factors to the detection results are determined, which specifically include:

[0100] Based on historical data or relevant literature, through big data analysis, obtain the relevant influencing factors of drones affecting environmental monitoring in a dynamic state;

[0101] According to the types of influencing factors, the linear regression equation of the error factor of UAV dynamic environment monitoring is established;

[0102] Through field test experiments or simulation software, the dynamic monitoring of UAVs can be simulated to obtain the dynamic environmental monitoring data of UAVs;

[0103] According to the normal standard value of the environment and the dynamic environmental monitoring data of the UAV, an error factor evaluation model is established based on the least squares method and gradient descent algorithm;

[0104] According to the error factor evaluation model, the acceptable range of monitoring data is obtained through big data analysis, and the threshold is set;

[0105] Through the error factor evaluation model, the received UAV dynamic environment monitoring data is evaluated in real time to determine whether the calculated error value exceeds the threshold.

[0106] If yes, it means that the environmental monitoring data of the drone in the current state is distorted. If continuous distortion occurs, it means that the current drone cannot complete the environmental monitoring task and needs to return for adjustment. If no, it means that the drone meets the environmental monitoring requirements in the current state.

[0107] The linear regression equation of the error factor of the UAV dynamic environment monitoring is: In the formula, is the calculated value of the error factor of the UAV dynamic environment monitoring, To influence environmental monitoring The parameters of the influencing factors are: is the error term, is a constant, To influence environmental monitoring Influencing factors, among which The value of is equal to 1, is the number of factors affecting environmental monitoring;

[0108] The least squares method and gradient descent algorithm expressions are: In the formula, is the optimal value of the error factor calculated by the least squares method, is the number of environmental monitoring data sample groups, is the error factor of the UAV dynamic environment monitoring Calculated values, For the Normal standard value of the environment, To influence environmental monitoring The update parameters of the impact factors are To influence environmental monitoring The parameters of the influencing factors, is the learning rate of gradient descent, For the Impact Factor Environmental monitoring data.

[0109] It can be explained that when using drones for dynamic data collection, the accuracy of the drone's dynamic environmental monitoring data is not high due to the influence of the drone's aerodynamics, movement and noise, as well as the temperature, humidity, pressure, wind speed, bad weather and other factors of the environment. This scheme establishes a linear regression equation for the error factor of the drone's dynamic environmental monitoring, relates all influencing factors in a linear way, and reduces the gap between the calculated value and the actual value of the error factor through the least squares method and gradient descent algorithm, thereby improving the accuracy of the error factor calculation value and determining the optimal value in the best way, thereby effectively counter-complementing and optimizing the drone's dynamic environmental monitoring data through the drone's dynamic environmental monitoring error factor.

[0110] Reference Figure 5 As shown, the establishment of the environmental monitoring dynamic-static input matrix based on the static interference offset coefficient and the error factor specifically includes:

[0111] According to the operation status of the UAV, establish a dynamic-static monitoring mode for environmental monitoring;

[0112] According to the change of the speed of the drone in the hovering state, set the threshold of whether the drone is moving;

[0113] The dynamic and static state of the drone is determined according to the speed monitoring value of the drone. By comparing whether the speed monitoring value of the drone is greater than the threshold, if so, it means that the drone is in motion and the dynamic mode of environmental monitoring is adopted. If not, it means that the drone is in hovering state and the static mode of environmental monitoring is adopted.

[0114] According to the static interference offset coefficient and error factor, the dynamic-static input matrix of environmental monitoring is established;

[0115] The environmental monitoring dynamic-static input matrix expression is: In the formula, Dynamic-static input matrix for environmental monitoring, Screening matrix for dynamic-static mode of environmental monitoring, is the correction value for the UAV static environment monitoring, Correction value for UAV dynamic environment monitoring.

[0116] What can be explained is that by analyzing the static and dynamic environmental monitoring errors of the UAV during the environmental monitoring process, a dynamic-static mode of UAV environmental monitoring is established. The operating state of the UAV is judged according to the speed change threshold of the UAV in the hovering and shaking state. By comparing whether the UAV speed monitoring value is greater than the threshold, if so, it means that the UAV is in motion and the environmental monitoring dynamic mode is adopted. If not, it means that the UAV is in a hovering state and the environmental monitoring static mode is adopted. The UAV environmental monitoring mode is screened and confirmed through the environmental monitoring dynamic-static mode screening matrix, thereby screening out the environmental monitoring mode suitable for the operating state of the UAV, and establishing the environmental monitoring dynamic-static input matrix.

[0117] Reference Figure 6 As shown, the dynamic-static input matrix of environmental monitoring is used to determine whether there is an abnormality in the monitoring area environment based on the neural network model, and the abnormality level is determined specifically including:

[0118] According to the dynamic-static input matrix of environmental monitoring, based on the neural network model, the received data is processed to determine whether there is any abnormality in the environment of the monitoring area;

[0119] By comparing the environmental monitoring data determined as abnormal by the neural network model with the environmental abnormality feature library, the type of environmental abnormal behavior is queried and the abnormality level is determined;

[0120] Based on whether there is an abnormality and the abnormality level, a decision is made whether to send abnormal information to the ground receiving platform. If so, an abnormal alarm and abnormality type are issued to the ground platform, and the abnormal information is recorded through a storage device. If not, it means that there is no environmental problem in the area and no abnormal alarm is issued.

[0121] It can be explained that, based on the neural network model, this scheme determines whether there is any abnormality in the environment of the monitoring area by confirming the dynamic-static input matrix of environmental monitoring, and processing the received dynamic-static input matrix data of environmental monitoring through the neural network model, and compares the environmental abnormality feature library based on the abnormal data to determine the level of environmental abnormality, and decides whether to send abnormal information to the ground receiving platform based on the judgment of whether it is abnormal and the level of abnormality. If so, an abnormal alarm and abnormal type are issued to the ground platform, and the abnormal information is recorded through a storage device. If not, it means that there is no environmental problem in the area and no abnormal alarm is issued.

[0122] Further, based on the same inventive concept as the above-mentioned drone environment monitoring and disaster warning method, this solution also proposes a drone environment monitoring and disaster warning system, including:

[0123] A hardware setting module, which is used to integrate corresponding monitoring equipment and MEC servers using drones as mobile carriers for edge computing of environmental monitoring according to the requirements of the monitoring environment;

[0124] A data processing module, which is used to clean and standardize the environmental data collected by the monitoring equipment; extract the characteristics of abnormal events based on the field detection and historical data of abnormal events in the monitoring area, and establish an environmental abnormality feature library;

[0125] A dynamic-static modeling module, which is used to calculate the static interference offset coefficient of the environmental data in the monitoring area according to the normal standard value of the environment and the test results of the actual test flight; and to determine the error factors caused by different environmental and UAV dynamic factors to the test results according to the monitoring environment and the dynamic analysis of the UAV;

[0126] An anomaly detection module, the anomaly detection module is used to establish an environmental monitoring dynamic-static input matrix based on the static interference offset coefficient and the error factor; according to the environmental monitoring dynamic-static input matrix, based on the neural network model, determine whether there is an abnormality in the monitoring area environment, and determine the abnormality level;

[0127] A ground receiving module, which is used to establish a ground receiving platform for receiving, displaying and storing environmental monitoring data, and indicating the level of environmental anomalies through small lights of different colors;

[0128] The data processing module includes:

[0129] A data processing unit, the data processing unit is used to clean and standardize the environmental data collected by the monitoring equipment;

[0130] Establishing an abnormal feature library unit, the abnormal feature library establishing unit is used to extract the features of abnormal events according to the field detection and historical data of abnormal events in the monitoring area, and establish an environmental abnormal feature library;

[0131] The dynamic-static modeling module includes:

[0132] A static modeling unit, wherein the static modeling unit is used to calculate a static interference offset coefficient of the environmental data of the monitoring area according to a normal standard value of the environment and a detection result of an actual test flight;

[0133] A dynamic modeling unit, which is used to determine the error factors caused by different environmental and UAV dynamic factors to the detection results according to the monitoring environment and the UAV dynamic analysis;

[0134] The anomaly detection module includes:

[0135] Establishing an input matrix unit, wherein the input matrix establishing unit is used to establish an environmental monitoring dynamic-static input matrix based on a static interference offset coefficient and an error factor;

[0136] The abnormality judgment unit is used to judge whether there is an abnormality in the monitoring area environment and to judge the abnormality level based on the dynamic-static input matrix of the environmental monitoring and the neural network model.

[0137] In summary, the advantages of the present invention are: using drones to integrate corresponding monitoring equipment and MEC servers as mobile carriers for environmental monitoring edge computing, automatically processing and analyzing monitoring data close to the source of data generation when controlling drones for environmental monitoring, effectively improving data reliability and accuracy, and preventing data loss and damage during transmission, thereby wasting manpower and material resources; secondly, by analyzing the static and dynamic environmental monitoring errors of drones during environmental monitoring, a dynamic-static mode of drone environmental monitoring is established, and through the environmental monitoring dynamic-static mode screening matrix, according to the speed detection of the drone, a suitable drone environmental monitoring mode is screened out, and a Establish an environmental monitoring dynamic-static input matrix, and through the environmental monitoring dynamic-static input matrix, based on the neural network model, judge whether there is any abnormality in the monitoring area environment, and judge the level of abnormality. Finally, receive, display and store environmental monitoring data through the ground receiving platform, and use small lights of different colors to indicate the level of environmental abnormality, so as to achieve the purpose of environmental monitoring and disaster warning of the monitoring area by UAV, effectively improve the efficiency, convenience, accuracy and reliability of environmental monitoring, greatly improve the accuracy of environmental anomaly judgment, ensure timely warning and take solutions, and provide possibilities for environmental monitoring in complex terrain and sparsely populated areas, and realize large-scale and no-dead-angle environmental monitoring.

[0138] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for unmanned aerial vehicle environment monitoring and disaster warning, characterized in that: include: According to the requirements of the monitoring environment, drones are used to integrate corresponding monitoring equipment and MEC servers as mobile carriers for environmental monitoring edge computing; Clean and standardize the environmental data collected by monitoring equipment; Based on the field detection and historical data of abnormal events in the monitoring area, the characteristics of abnormal events are extracted and an environmental abnormality feature library is established; Calculate the static interference deviation coefficient of the environmental data in the monitoring area based on the normal standard value of the environment and the test results of the actual test flight; According to the monitoring environment and the dynamic analysis of the UAV, the relevant influencing factors that affect environmental monitoring under different environments and dynamic states of the UAV are determined, and the linear regression equation of the error factor of the dynamic environmental monitoring of the UAV is constructed. At the same time, based on the least squares method and gradient descent algorithm, the error factor brought to the detection result by the current monitoring environment and the dynamic state of the UAV is determined; Based on the static interference offset coefficient and error factor, the dynamic-static input matrix of environmental monitoring is established; According to the dynamic-static input matrix of environmental monitoring, based on the neural network model, it is determined whether there is any abnormality in the environment of the monitoring area, and the abnormality level is determined; Establish a ground receiving platform to receive, display and store environmental monitoring data, and indicate the level of environmental anomalies through small lights of different colors; The static interference offset coefficient expression of the monitoring area environmental data is: In the formula, To monitor the static interference offset coefficient of regional environmental data, is the error balance factor, is a constant, For Moment Normal standard value of the environment, For Moment UAV environmental monitoring values, is the number of data groups; The establishment of the environmental monitoring dynamic-static input matrix based on the static interference offset coefficient and the error factor specifically includes: According to the operation status of the UAV, establish a dynamic-static monitoring mode for environmental monitoring; According to the change of the speed of the drone in the hovering state, set the threshold of whether the drone is moving; The dynamic and static state of the drone is determined according to the speed monitoring value of the drone. By comparing whether the speed monitoring value of the drone is greater than the threshold, if so, it means that the drone is in motion and the dynamic mode of environmental monitoring is adopted. If not, it means that the drone is in hovering state and the static mode of environmental monitoring is adopted. According to the static interference offset coefficient and error factor, the dynamic-static input matrix of environmental monitoring is established; The environmental monitoring dynamic-static input matrix expression is: In the formula, Dynamic-static input matrix for environmental monitoring, Screening matrix for dynamic-static mode of environmental monitoring, is the correction value for the UAV static environment monitoring, Correction value for UAV dynamic environment monitoring.

2. The method for environmental monitoring and disaster warning using a drone according to claim 1, characterized in that: The cleaning and standardization of the environmental data collected by the monitoring equipment specifically includes: Based on the first-order Kalman filter algorithm, the received data is subjected to real-time denoising and filtering processing; Perform bit reduction processing on the denoising filtered data; For the data after decimal point reduction, standardization is performed to remove decimal points; The drop processing expression is: In the formula, It is the environmental monitoring data after downgrading. is the environmental monitoring data after denoising and filtering. The part with the same value as the higher digit of the environmental monitoring data; The decimal point removal standardization processing expression is: In the formula, This is the environmental monitoring data after decimal point standardization. It is the environmental monitoring data after downgrading. is a constant term.

3. The method for environmental monitoring and disaster warning using a drone according to claim 2, characterized in that: The extracting of abnormal event features based on field detection and historical data of abnormal events in the monitoring area and establishing an environmental abnormality feature library specifically includes: Based on the on-site detection values ​​and external appearances of the environment when an abnormal event occurs in the monitoring area, a large number of abnormal environment detection values ​​and external appearances of the environment similar to the abnormal event in the monitoring area are obtained through big data; Through horizontal comparison, the common feature items of environmental abnormal events are extracted, and the feature items are digitized and standardized; The processed environmental anomaly feature items are classified and divided according to their nature, the disaster risk level is set, and an environmental anomaly feature database is established.

4. The method for environmental monitoring and disaster warning using a drone according to claim 3, characterized in that: The calculation of the static interference offset coefficient of the environmental data in the monitoring area according to the normal standard value of the environment and the test results of the actual test flight specifically includes: Through professional equipment, we can comprehensively detect the environmental data of the monitoring area and compare it with the environmental abnormality feature library, establish the normal standard value of the environment, and set the stable range; While conducting professional inspections, use drones equipped with monitoring equipment to statically collect environmental data at the same location and area; According to the normal standard value of the environment and the detection results of the actual test flight, the static interference deviation coefficient of the environmental data in the monitoring area is calculated.

5. The method for environmental monitoring and disaster warning using a drone according to claim 4, characterized in that: The error factors brought to the detection results by the current monitoring environment and the dynamic state of the UAV specifically include: Based on historical data or relevant literature, through big data analysis, obtain the relevant influencing factors of drones affecting environmental monitoring in a dynamic state; According to the types of influencing factors, the linear regression equation of the error factor of UAV dynamic environment monitoring is established; Through field test experiments or simulation software, the dynamic monitoring of UAVs can be simulated to obtain the dynamic environmental monitoring data of UAVs; According to the normal standard value of the environment and the dynamic environmental monitoring data of the UAV, an error factor evaluation model is established based on the least squares method and gradient descent algorithm; According to the error factor evaluation model, the acceptable range of monitoring data is obtained through big data analysis, and the threshold is set; Through the error factor evaluation model, the received UAV dynamic environment monitoring data is evaluated in real time to determine whether the calculated error value exceeds the threshold. If yes, it means that the environmental monitoring data of the drone in the current state is distorted. If continuous distortion occurs, it means that the current drone cannot complete the environmental monitoring task and needs to return for adjustment. If no, it means that the drone meets the environmental monitoring requirements in the current state. The linear regression equation of the error factor of the UAV dynamic environment monitoring is: In the formula, is the calculated value of the error factor of the UAV dynamic environment monitoring, To influence environmental monitoring The parameters of the influencing factors are: is the error term, is a constant, To influence environmental monitoring Influencing factors, among which The value of is equal to 1, is the number of factors affecting environmental monitoring; The least squares method and gradient descent algorithm expressions are: In the formula, is the optimal value of the error factor calculated by the least squares method, is the number of environmental monitoring data sample groups, is the error factor of the UAV dynamic environment monitoring Calculated values, For the Normal standard value of the environment, To influence environmental monitoring The update parameters of the impact factors are To influence environmental monitoring The parameters of the influencing factors, is the learning rate of gradient descent, For the Impact Factor Environmental monitoring data.

6. The method for environmental monitoring and disaster warning using a drone according to claim 5, characterized in that: The method of judging whether there is an abnormality in the environment of the monitoring area based on the dynamic-static input matrix of the environmental monitoring and the neural network model and judging the abnormality level specifically includes: According to the dynamic-static input matrix of environmental monitoring, based on the neural network model, the received data is processed to determine whether there is any abnormality in the environment of the monitoring area; By comparing the environmental monitoring data judged as abnormal by the neural network model with the environmental anomaly feature library, the type of environmental abnormal behavior is queried and the abnormality level is determined; based on whether it is abnormal and the abnormality level, a decision is made whether to send abnormal information to the ground receiving platform. If so, an abnormal alarm and the abnormality type are issued to the ground platform, and the abnormal information is recorded through a storage device. If not, it means that there is no environmental problem in the area and no abnormal alarm is issued.

7. An unmanned aerial vehicle environment monitoring and disaster warning system, used to implement the unmanned aerial vehicle environment monitoring and disaster warning method according to any one of claims 1 to 6, characterized in that: include: A hardware setting module, which is used to integrate corresponding monitoring equipment and MEC servers using drones as mobile carriers for edge computing of environmental monitoring according to the requirements of the monitoring environment; A data processing module, which is used to clean and standardize the environmental data collected by the monitoring equipment; Based on the field detection and historical data of abnormal events in the monitoring area, the characteristics of abnormal events are extracted and an environmental abnormality feature library is established; A dynamic-static modeling module, which is used to calculate the static interference offset coefficient of the environmental data in the monitoring area according to the normal standard value of the environment and the test results of the actual test flight; and to determine the error factors caused by different environmental and UAV dynamic factors to the test results according to the monitoring environment and the dynamic analysis of the UAV; An anomaly detection module, the anomaly detection module is used to establish an environmental monitoring dynamic-static input matrix based on the static interference offset coefficient and the error factor; according to the environmental monitoring dynamic-static input matrix, based on the neural network model, determine whether there is an abnormality in the monitoring area environment, and determine the abnormality level; A ground receiving module is used to establish a ground receiving platform for receiving, displaying and storing environmental monitoring data, and indicating the level of environmental anomalies through small lights of different colors.

8. The UAV environment monitoring and disaster warning system according to claim 7, characterized in that: The data processing module specifically includes: A data processing unit, the data processing unit is used to clean and standardize the environmental data collected by the monitoring equipment; An abnormal feature library unit is established, and the abnormal feature library unit is used to extract the characteristics of abnormal events and establish an environmental abnormal feature library based on field detection and historical data of abnormal events in the monitoring area.

9. The UAV environment monitoring and disaster warning system according to claim 8, characterized in that: The dynamic-static modeling module specifically includes: A static modeling unit, wherein the static modeling unit is used to calculate a static interference offset coefficient of the environmental data of the monitoring area according to a normal standard value of the environment and a detection result of an actual test flight; A dynamic modeling unit, which is used to determine the error factors caused by different environmental and UAV dynamic factors to the detection results according to the monitoring environment and the UAV dynamic analysis; The anomaly detection module specifically includes: Establishing an input matrix unit, wherein the input matrix establishing unit is used to establish an environmental monitoring dynamic-static input matrix based on a static interference offset coefficient and an error factor; The abnormality judgment unit is used to judge whether there is an abnormality in the monitoring area environment and to judge the abnormality level based on the dynamic-static input matrix of the environmental monitoring and the neural network model.

Images