A method and system for monitoring the ecological environment along the expressway road area

Abstract

The present application relates to the technical field of ecological environment monitoring, in particular to a kind of ecological environment monitoring method and system along the expressway road area, the present application is according to vehicle speed, vehicle type and environmental parameter, the similarity of environment at different time is analyzed, and the driving similarity of different vehicles of the same vehicle type is obtained Predicted speed sequence and predicted traffic flow sequence of each vehicle in each monitoring area in prediction period;Further, in combination with neural network model, the predicted greenhouse effect gas content sequence in each monitoring area in prediction period is obtained;Then, by analyzing the consistency of the predicted greenhouse effect gas content in each monitoring area, the importance of each monitoring area is obtained, and the density of unmanned aerial vehicle in each monitoring area is adjusted. Through the present application, the density of unmanned aerial vehicle in each monitoring area of expressway can be timely regulated, thereby effectively improving the monitoring accuracy of ecological environment in each monitoring area of the whole line, while avoiding the monitoring error caused by sudden environmental change.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ecological environment monitoring, in particular to a method and system for monitoring ecological environment along the road area of a highway. BACKGROUND

[0002] As an important part of China's modern transportation network system, highways have greatly promoted the vigorous development of the economy of the areas along the highways. However, the operation of highways is accompanied by the emission of pollutants, among which carbon dioxide and nitrogen oxides are the main sources of pollution. These gases directly affect the climate change and air quality of the areas along the highways, and the greenhouse effect caused by carbon dioxide has a significant impact on the health of the ecological system. Therefore, it is necessary to monitor the content of greenhouse gases along the road area of the highway in order to evaluate the ecological environment along the road area of the highway. However, due to the large range of the road area along the highway, it is difficult to achieve complete monitoring, so the density of monitoring points in each region along the road area of the highway is dynamically adjusted according to the relevant state of the content of greenhouse gases in each region, so as to achieve ecological environment monitoring along the highway.

[0003] In the prior art, when dynamically adjusting the density of monitoring points along the road area of the highway, the content change of greenhouse gases in the prediction period is first predicted according to the content change of greenhouse gases in each region in the past, and when the content change exceeds the expected value, the density of monitoring points in the corresponding region is increased. However, on the highway, vehicle exhaust is the main source of pollutants (carbon dioxide, nitrogen oxides, etc.) in the highway, and the traffic volume changes with the weather and time, which makes the traditional LSTM model-based prediction of the content change of greenhouse gases less effective, and thus the density of monitoring points in each region of the highway cannot be adjusted in a timely and accurate manner, making it difficult to accurately monitor the ecological environment along the road area of the highway. SUMMARY

[0004] In order to solve the above technical problems, the purpose of the present application is to provide a method and system for monitoring the ecological environment along the road area of a highway.

[0005] According to the first aspect of the embodiment of the present application, a method for monitoring the ecological environment along the road area of a highway is provided, and the technical solution is as follows:

[0006] The road area of the highway is divided into a plurality of monitoring regions, and the content of greenhouse gases, traffic volume, vehicle speed, vehicle type and environmental parameters in each monitoring region are collected;

[0007] According to the vehicle speed, in combination with the vehicle type and the environmental parameter, the environmental similarity at different time points and the driving similarity of different vehicles of the same vehicle type are analyzed to obtain a predicted vehicle speed sequence and a predicted vehicle flow sequence of each vehicle in each monitoring area within a prediction period;

[0008] According to the vehicle flow, the vehicle speed and the greenhouse gas content within a historical period, a neural network model for predicting the greenhouse gas content in each monitoring area is trained;

[0009] Based on the neural network model, in combination with the predicted vehicle speed sequence and the predicted vehicle flow sequence, a predicted greenhouse gas content sequence in each monitoring area within a prediction period is obtained;

[0010] According to the greenhouse gas content sequence within a historical period and the predicted greenhouse gas content sequence, the prediction consistency of the greenhouse gas content in each monitoring area is analyzed to obtain the importance degree of each monitoring area;

[0011] According to the importance degree, the density of the unmanned aerial vehicle in each monitoring area is adjusted.

[0012] In some embodiments of the present application, according to the vehicle speed, in combination with the vehicle type and the environmental parameter, the environmental similarity at different time points and the driving similarity of different vehicles of the same vehicle type are analyzed to obtain a predicted vehicle speed sequence and a predicted vehicle flow sequence of each vehicle in each monitoring area within a prediction period, including:

[0013] According to the environmental parameter, the environmental similarity of each monitoring point in each monitoring area at different time points is analyzed to obtain a reference period of the current time point and the environmental reference degree of the reference period of each monitoring point in the monitoring area;

[0014] Based on the vehicle speed and the environmental parameter of all vehicles of the same vehicle type at each monitoring point within the current time point and its reference period, in combination with the environmental reference degree, a preliminary predicted vehicle speed sequence of each vehicle of each vehicle type in each monitoring area within a prediction period is obtained;

[0015] According to the vehicle speed, a known vehicle speed change sequence of each vehicle of each vehicle type in each monitoring area within a historical period is obtained;

[0016] The difference of the vehicle speed corresponding to different vehicles of the same vehicle type is analyzed, and in combination with the environmental similarity, the driving similarity of different vehicles of the same vehicle type in each monitoring area is obtained;

[0017] In combination with the known vehicle speed change sequence and the driving similarity, a predicted vehicle speed change amount of each vehicle at each monitoring point in each monitoring area is obtained;

[0018] According to the predicted vehicle speed change amount, a predicted vehicle speed sequence of each vehicle in each monitoring area in a prediction period is obtained in combination with the preliminary predicted vehicle speed sequence;

[0019] Based on the predicted vehicle speed sequence, a predicted vehicle flow sequence in each monitoring area in a prediction period is obtained.

[0020] In some embodiments of the application, according to the environmental parameters, the environmental similarity of each monitoring point at different times in each monitoring area is analyzed to obtain a reference period of each monitoring point at the current time in the monitoring area and the environmental reference degree of the reference period, including:

[0021] According to the environmental parameters, the difference of the same kind of environmental parameters corresponding to each monitoring point at different times in each monitoring area is analyzed, and all kinds of environmental parameters are traversed to obtain the environmental similarity of each monitoring point at different times in each monitoring area;

[0022] A preset environmental similarity threshold is set, and if the environmental similarity of all times in the past period is greater than or equal to the environmental similarity threshold, the past period is recorded as the reference period of the current time, and multiple reference periods of each monitoring point at the current time in the monitoring area are obtained.

[0023] The average of the environmental similarity of each time in the reference period and the current time is calculated to obtain the environmental reference degree of the reference period.

[0024] In some embodiments of the application, based on the vehicle speed and environmental parameters of all vehicles of the same vehicle type at each monitoring point in the current time and the reference period thereof, in combination with the environmental reference degree, a preliminary predicted vehicle speed sequence of each vehicle of each vehicle type in each monitoring area in a prediction period is obtained, including:

[0025] Based on the LSTM prediction model, the vehicle speed and environmental parameters of all vehicles of the same vehicle type at each monitoring point in the current time and the multiple reference periods thereof are input, wherein the environmental reference degree is used as the reference weight of the vehicle speed and environmental parameters of each reference period, and the preliminary predicted vehicle speed sequence of each vehicle of each vehicle type in each monitoring area in a prediction period is output.

[0026] In some embodiments of the application, the difference of the vehicle speed corresponding to different vehicles of the same vehicle type is analyzed, and in combination with the environmental similarity, the driving similarity of different vehicles of the same vehicle type in each monitoring area is obtained, including:

[0027] Calculate absolute values of differences of the corresponding vehicle speeds of different vehicles of the same vehicle model passing through the same monitoring point, combine the corresponding environment similarities of different time points of different vehicles of the same vehicle model passing through the same monitoring point, traverse all monitoring points commonly passed through by different vehicles of the same vehicle model, and obtain driving similarities of different vehicles of the same vehicle model in each monitoring area.

[0028] In some embodiments of the present application, according to the predicted vehicle speed change amount, the preliminary predicted vehicle speed sequence is combined to obtain a predicted vehicle speed sequence of each vehicle in each monitoring area in a prediction period, including:

[0029] According to the predicted vehicle speed change amount, the preliminary predicted vehicle speed sequence is combined to obtain a corrected predicted vehicle speed sequence of each vehicle in each monitoring area in a prediction period;

[0030] Based on the corrected predicted vehicle speed sequence, Lagrange interpolation processing is performed to obtain a fitted vehicle speed of each position of the vehicle in the monitoring area on the expressway;

[0031] The fitted vehicle speed is arranged according to the position sequence of the vehicle on the expressway to obtain a predicted vehicle speed sequence of each vehicle in each monitoring area in a prediction period.

[0032] In some embodiments of the present application, according to the historical period greenhouse gas content and the predicted greenhouse gas content, the prediction past consistency of the greenhouse gas content in each monitoring area is analyzed to obtain the importance degree of each monitoring area, including:

[0033] Calculate the DTW similarity of the historical period greenhouse gas content and the predicted greenhouse gas content to obtain the prediction past consistency of the greenhouse gas content in each monitoring area;

[0034] According to the prediction past consistency, the importance degree of each monitoring area is obtained.

[0035] In some embodiments of the present application, according to the importance degree, the density of the unmanned aerial vehicle in each monitoring area is adjusted, including:

[0036] The importance degree threshold value includes a large threshold value, a medium threshold value and a lower threshold value;

[0037] If the importance degree of the monitoring area is greater than or equal to the large threshold value, the density of the unmanned aerial vehicle in the monitoring area is doubled;

[0038] If the importance degree of the monitoring area is greater than or equal to the medium threshold value and less than the large threshold value, the density of the unmanned aerial vehicle in the monitoring area is increased by 0.5 times;

[0039] If the importance of the monitoring area is greater than or equal to the lower threshold and less than the middle threshold, the density of the unmanned aerial vehicle in the monitoring area is increased by 0.25 times.

[0040] According to a second aspect of the embodiments of the present application, a kind of ecological environment monitoring system of highway along the line is provided, comprising: memory and processor, wherein:

[0041] The memory is used to store program code.

[0042] The processor is used to read the program code stored in the memory, and execute the method described in the first aspect of the embodiments of the present application.

[0043] In some embodiments of the present application, the processor includes:

[0044] The data acquisition module is used to divide the highway into several monitoring areas, and collect the greenhouse gas content, traffic flow, vehicle speed, vehicle type and environmental parameters in each monitoring area.

[0045] The road condition prediction module is used to analyze the environmental similarity at different times and the driving similarity of different vehicles of the same vehicle type according to the vehicle speed, combined with the vehicle type and the environmental parameters, to obtain the predicted vehicle speed and predicted traffic flow of each vehicle in each monitoring area in the prediction period.

[0046] The greenhouse gas content prediction module is used to train a neural network model for predicting the greenhouse gas content in each monitoring area according to the traffic flow, vehicle speed and greenhouse gas content in the historical period, and obtain the predicted greenhouse gas content in each monitoring area in the prediction period based on the neural network model, combined with the predicted vehicle speed and the predicted traffic flow.

[0047] The unmanned aerial vehicle density adjustment module is used to analyze the prediction consistency of the greenhouse gas content in each monitoring area according to the greenhouse gas content and the predicted greenhouse gas content in the historical period, to obtain the importance of each monitoring area, and adjust the density of unmanned aerial vehicles in each monitoring area according to the importance.

[0048] Compared with the prior art, the ecological environment monitoring method and system provided by the present application have the following advantages:

[0049] The present application firstly predicts the vehicle speed of each vehicle in the prediction period based on the environment and the vehicle position, then obtains the position information of each vehicle in the prediction period according to the vehicle speed of each vehicle in the prediction period, and obtains the corresponding traffic flow and speed in each monitoring area in the prediction period; then, the greenhouse gas content in the prediction period is predicted according to the past speed and traffic flow of each monitoring area and the corresponding greenhouse gas content, combined with the neural network model; finally, the dynamic attention degree of each monitoring area is constructed by combining the greenhouse gas content of each monitoring area on the whole line, and the density of the unmanned aerial vehicle in each monitoring area is adjusted. The density of the unmanned aerial vehicle monitoring point in each monitoring area of the expressway is timely regulated by the present application, thereby effectively improving the monitoring accuracy of the ecological environment in each monitoring area on the whole line, and avoiding the monitoring error caused by sudden changes in the environment. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, a brief introduction will be given to the drawings needed in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0051] Figure 1 A basic flowchart of an ecological environment monitoring method for an expressway along the road area provided by an embodiment of the present application;

[0052] Figure 2 A basic flowchart of a method for obtaining a predicted speed sequence and a predicted traffic flow sequence of each vehicle in each monitoring area in the prediction period provided by an embodiment of the present application;

[0053] Figure 3 A basic composition diagram of an ecological environment monitoring system for an expressway along the road area provided by an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the specific implementation, structure, features and effects of the ecological environment monitoring method and system for an expressway along the road area according to the present application are described in detail as follows by combining the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The use of the terms "at least" and "one or more of should be taken as

[0056] The application provides a method for monitoring an ecological environment along a highway region.

[0057] Please refer to Figure 1 , which shows the basic flow of a method for monitoring an ecological environment along a highway region according to an embodiment of the application.

[0058] As shown in Figure 1 , the method for monitoring an ecological environment along a highway region according to an embodiment of the application specifically includes the following steps.

[0059] S100: Divide the highway region into a plurality of monitoring areas, and collect the greenhouse gas content, traffic volume, vehicle speed, vehicle type, and environmental parameters in each monitoring area.

[0060] The highway region is divided into a plurality of monitoring areas, and the greenhouse gas content, traffic volume, vehicle speed, vehicle type, and environmental parameters in each monitoring area are collected, wherein the environmental parameters include air visibility and road surface humidity. Specifically, the grid point method is used along the highway region to divide the highway region into a plurality of monitoring areas. An initial monitoring point density is set for these monitoring areas, i.e., the number of monitoring drones in each monitoring area. The initial monitoring point density is preset to be 4 monitoring drones per square kilometer, and an electrochemical and semiconductor gas sensor is arranged on each monitoring drone to detect the greenhouse gas content in the air in real time, wherein the preset detection frequency is 10 times per minute. In the embodiment of the application, the greenhouse gas can be carbon dioxide.

[0061] Meanwhile, a fixed monitoring point is set every certain distance (100 km by default) along the highway, and a high-definition camera is arranged at the monitoring point. High-precision vehicle detection (vehicle speed and vehicle type detection) is realized by combining the YOLO and Mask R-CNN models based on deep learning. In addition, an induction coil and a geomagnetic sensor are arranged to realize real-time vehicle flow detection. The vehicle and vehicle flow detection frequency is set to 1 time / second, and the vehicle flow data, vehicle speed, and vehicle type classification at each monitoring point at each time are obtained. A scattering visibility meter and a non-contact road surface condition detector (such as the TH-LM1 series) are also arranged at the monitoring point to monitor environmental parameters such as air visibility and road surface humidity.

[0062] The preset data collection time of the greenhouse gas content, vehicle flow, vehicle speed, vehicle type, and environmental parameters in each monitoring area is the past year to the present, a total of 365 days.

[0063] At this point, the information of several parameters at each monitoring point in each monitoring area is collected to form an information library.

[0064] On the highway, the driving of vehicles is greatly affected by the environment. When the visibility or road surface humidity changes, the vehicle speed needs to be reduced to ensure safety during high-speed driving. When the vehicle flow does not change and the vehicle speed is reduced, the vehicles will stay in some monitoring areas for a longer time, which will gradually increase the greenhouse gas content in the monitoring area. Although the vehicle flow of a monitoring area on the highway is difficult to predict, there is basically no vehicle flow convergence or divergence on the highway, which also leads to the fact that vehicles will always move forward on the highway. Therefore, the monitoring area of these vehicles in the prediction period can be determined according to the vehicle flow of each monitoring area on the highway in the current period of time, combined with the speed of the vehicle under the current environment. Further, the vehicle flow information of each monitoring area in the prediction period can be obtained, and then the change of the greenhouse gas content of each monitoring area along the line in the future period of time can be predicted according to the corresponding relationship between the vehicle flow, vehicle speed, and oxide content in the past information library. The specific steps include S200 to S400.

[0065] S200: According to the vehicle speed, combined with the vehicle type and environmental parameters, the similarity of the environment at different times and the similarity of the driving of vehicles of the same type are analyzed to obtain the predicted vehicle speed sequence and the predicted vehicle flow sequence of each vehicle in each monitoring area in the prediction period.

[0066] Please refer to Figure 2 which shows the basic flow of a method for obtaining the predicted vehicle speed sequence and the predicted vehicle flow sequence of each vehicle in each monitoring area in the prediction period provided by an embodiment of the present application.

[0067] AsFigure 2 As shown, according to the vehicle speed, in combination with the vehicle type and the environmental parameter, the environmental similarity at different time is analyzed, and the driving similarity of different vehicles of the same vehicle type is analyzed, to obtain the predicted vehicle speed sequence and the predicted vehicle flow sequence of each vehicle in each monitoring area in the prediction period. Further comprising:

[0068] S201: According to the environmental parameter, the environmental similarity of each monitoring point in each monitoring area at different time is analyzed, to obtain the reference period of the current time at each monitoring point in the monitoring area and the environmental reference degree of the reference period.

[0069] On the highway, various vehicles need to drive at high speed, but due to the difficulty of responding to sudden conditions when the vehicle drives at high speed, especially in low visibility and wet road surface, the vehicle speed needs to be reduced when the environment is poor, which leads to that the driving speed of the vehicle on the highway is greatly affected by the environment, and different vehicle types are also affected by the environment.

[0070] Based on the above analysis, in some embodiments of the present application, according to the environmental parameter, the environmental similarity of each monitoring point in each monitoring area at different time is analyzed, to obtain the reference period of the current time at each monitoring point and the environmental reference degree of the reference period. Specifically comprising:

[0071] First, according to the environmental parameter, the difference of the same kind of environmental parameter corresponding to each monitoring point in each monitoring area at different time is analyzed, all kinds of environmental parameters are traversed, and the environmental similarity of each monitoring point in each monitoring area at different time is obtained. Since the more similar the environment is, the more likely the vehicle speed is the same as the current one, the environmental similarity between each reference time and the current time is calculated first, wherein an environmental monitoring point (referred to as the current position ) is taken as an example, the environmental similarity between the past time and the current time is calculated as:

[0072] In the formula, represents the environmental similarity between the past time and the current time at the monitoring point (the current position ); represents the parameter value of the environmental parameter at the past time at the current position ; represents the parameter value of the environmental parameter at the past time at the current position ;​ represents the total number of environmental parameter types participating in the ecological environment detection; represents taking the absolute value; represents the exponential function with the natural constant as the base; represents the linear normalization function.

[0073] The smaller the difference in the numerical values of all kinds of environmental parameters at two time points, the greater the environmental similarity between the two time points.

[0074] Similarly, the environmental similarity of each monitoring point in each monitoring area at different time points is obtained.

[0075] Then, a preset environmental similarity threshold (the value can be 0.6) is set, and if the environmental similarity between all time points in the past period and the current time point is greater than or equal to the environmental similarity threshold, that is, it is considered that the environment in the monitoring point at the time period is basically the same as the current environment, and the environmental change in the prediction period from the current time point is also basically the same, and the past period is recorded as the reference period of the current time point, that is, the past period is the period in which the current time point is located (the preset backward extension is half an hour, that is, the prediction length), and the multiple reference periods of each monitoring point at the current time point (the period in which the current time point is located) are obtained.

[0076] Finally, the average of the environmental similarity between each time point in the reference period and the current time point is calculated to obtain the environmental reference degree of the reference period.

[0077] S202: Based on the vehicle speed and environmental parameters of all vehicles of the same vehicle type at each monitoring point in the current time point and its reference period, and combined with the environmental reference degree, a preliminary prediction speed sequence of each vehicle of each vehicle type in each monitoring area in the prediction period is obtained.

[0078] Under the demand for safe driving, the reaction of the vehicle in the same environment will inevitably have some similar trends, so the possible speed of the vehicle in the monitoring area in the prediction period can be predicted according to the speed change trend of the vehicle in the past similar environment.

[0079] Based on the above analysis, in some embodiments of the present application, based on the vehicle speed and environmental parameters of all vehicles of the same vehicle type at each monitoring point in the current time point and its reference period, and combined with the environmental reference degree, a preliminary prediction speed sequence of each vehicle of each vehicle type in each monitoring area in the prediction period is obtained. Specifically, it includes:

[0080] Firstly, based on the LSTM prediction model, the speed and environmental parameters of all vehicles of the same vehicle type at each monitoring point in the current time and multiple reference periods are input, wherein the environmental reference degree is taken as the reference weight of the speed and environmental parameters in each reference period, and a preliminary predicted speed sequence of each vehicle of each vehicle type in each monitoring area in the prediction period is output. It should be noted that when the vehicle is normally driving on the highway, the time and position are in a one-to-one correspondence, that is, when the time advances, the position of the vehicle also advances accordingly, so the preliminary predicted speed sequence is not only a time sequence corresponding to the prediction period, but also a position sequence of each monitoring point in each monitoring area corresponding to the prediction period.

[0081] S203: Obtain a known speed change sequence of each vehicle of each vehicle type in each monitoring area in the history period according to the speed.

[0082] For the highway, the driving conditions of vehicles in different monitoring areas are different, such as straight driving of vehicles in a straight road without the need to decelerate, but in a curve, the speed needs to be reduced to ensure the stability of the vehicle, and the same type of vehicles have certain similarity in the time and condition of acceleration and deceleration in the same road condition, so the speed change demand of the target vehicle in the prediction period in a monitoring area can be obtained according to the known speed change of numerous vehicles of the same type in the monitoring area.

[0083] Based on the above analysis, in some embodiments of the present application, firstly, a known speed change sequence of each vehicle of each vehicle type in each monitoring area in the history period is obtained according to the speed. Specifically, according to the obtained speed data, a known speed sequence of each vehicle of each vehicle type in each monitoring area in the history period is obtained, a difference sequence of the known speed sequence is calculated, and a known speed change sequence of each vehicle of each vehicle type in each monitoring area in the history period is obtained. Taking a vehicle in a monitoring area on the highway as an example, denoted as a target vehicle in a target monitoring area , a plurality of vehicles of the same type as the target vehicle are selected as reference vehicles , the speed change of the reference vehicles in the target monitoring area is analyzed, that is, the difference sequence of the known speed sequence of the reference vehicles at each speed monitoring point in the target monitoring area is calculated, denoted as the known speed change sequence of the reference vehicles at each speed monitoring point in the target monitoring area , that is, the known speed change sequence of the reference vehicles in the target monitoring area in the history period is obtained, denoted as ​,in, Indicates reference vehicle In the target monitoring area Internal monitoring point The known speed changes at each location are obtained. Similarly, the known speed change sequences for each vehicle type in each monitoring area during the historical period are obtained.

[0084] S204: Analyze the differences in vehicle speeds among different vehicles of the same model, and combine this with environmental similarity to obtain the driving similarity of different vehicles of the same model within each monitoring area.

[0085] Although vehicles of the same type have certain similarities in acceleration, deceleration, and driving behavior under similar road conditions, there are still differences in vehicle performance and driver habits. Therefore, it is necessary to assess the similarity of vehicle performance and driver driving status based on the similarity of driving behavior of different vehicles of the same model on past routes, and to construct the similarity of driving status between the two vehicles on past highway sections.

[0086] Based on the above analysis, in some embodiments of the present invention, the driving similarity of different vehicles of the same model is obtained by analyzing the differences in vehicle speeds and combining this with environmental similarity. Specifically, the absolute value of the difference in vehicle speeds at the same monitoring point for different vehicles of the same model is calculated, and combined with the environmental similarity at different times corresponding to the same monitoring point for different vehicles of the same model, all monitoring points commonly passed by different vehicles of the same model are traversed to obtain the driving similarity of different vehicles of the same model. (Taking the target vehicle as an example...) Its corresponding reference vehicle For example, the target vehicle Its corresponding reference vehicle driving similarity The calculation formula is as follows:

[0087] In the formula, Indicates the target vehicle Its corresponding reference vehicle Driving similarity; Indicates the reference vehicle With the target vehicle After the first on the highway Environmental similarity at different times corresponding to each vehicle speed monitoring point; Indicates reference vehicle Passing through vehicle speed monitoring point The vehicle speed at that time; Indicates the target vehicle Passing through vehicle speed monitoring point The vehicle speed at that time; for taking absolute value; denotes exponential function with natural constant as base; denotes target vehicle and reference vehicle total number of vehicle speed monitoring points commonly passed in highway journey.

[0088] The smaller the value, i.e. the smaller the speed difference of the target vehicle and the reference vehicle when passing the same vehicle speed monitoring point, the more similar the performance and driving habits of the two vehicles are; in addition, since the time of the two vehicles passing the same vehicle speed monitoring point can be different, the environmental conditions of the same vehicle speed monitoring point can also be different, the greater the environmental similarity, the the higher the reliability, therefore the environmental similarity between the corresponding different time points when the two vehicles pass the same vehicle speed monitoring point is used to correct , and finally the driving similarity of the target vehicle and the reference vehicle is obtained.

[0089] Similarly, the driving similarity of different vehicles of the same vehicle type in each monitoring area is obtained.

[0090] S205: Obtain the predicted vehicle speed change amount of each vehicle at each monitoring point in each monitoring area by combining the known vehicle speed change sequence and the driving similarity.

[0091] Obtain the predicted vehicle speed change amount of each vehicle at each monitoring point by combining the known vehicle speed change sequence and the driving similarity. Take the target vehicle in the target monitoring area as an example, the more similar the driving conditions of the two vehicles in the past journey are, the more similar the driving speeds of the two vehicles in the following monitoring area will be, and when predicting the driving conditions of the target vehicle by the driving conditions of numerous vehicles of the same type passing each monitoring area, the reference will be higher, and the driving actions of these vehicles of the same type passing the same monitoring area in the same environment will also be more similar. Accordingly, to construct the predicted vehicle speed sequence of the target vehicle in the target monitoring area , first, obtain the predicted vehicle speed change amount of the target vehicle at the monitoring point (current position ) in the target monitoring area by combining the known vehicle speed change sequence and the driving similarity, which is:

[0092] In the formula, denotes the target vehicle​ In the target monitoring area Internal monitoring point (Current location) The predicted change in vehicle speed; Indicates reference vehicle In the target monitoring area Internal monitoring point The known change in vehicle speed at the location; Indicates the target vehicle Its corresponding reference vehicle Driving similarity; Indicates the target monitoring area Internal target vehicle The total number of corresponding reference vehicles.

[0093] By referring to the known changes in vehicle speed By combining the driving similarity between the reference vehicle and the target vehicle, the change in the target vehicle's speed is predicted, which is the known change in the reference vehicle's speed. The larger the value, the greater the similarity between the driving characteristics of the reference vehicle and the target vehicle. The larger the value, the greater the predicted change in the target vehicle's speed. The larger.

[0094] S206: Based on the predicted change in vehicle speed and combined with the preliminary predicted vehicle speed sequence, the predicted vehicle speed sequence for each vehicle in each monitoring area during the prediction period is obtained.

[0095] In summary, the speed of a vehicle on a highway varies under the combined influence of the environment and road conditions. That is, when the environment is fixed, the speed of a vehicle will be adjusted according to the local road conditions as it travels to different locations.

[0096] Therefore, in some embodiments of the present invention, based on the predicted vehicle speed change and combined with the preliminary predicted vehicle speed sequence, the predicted vehicle speed sequence for each vehicle within each monitoring area during the prediction period is obtained. Specifically:

[0097] First, based on the predicted vehicle speed change and the preliminary predicted vehicle speed sequence, a revised predicted vehicle speed sequence for each vehicle within each monitoring area is obtained during the prediction period. This is based on the vehicle's current position at the monitoring point. target vehicle For example, the target vehicle Monitoring points arriving within the monitoring area during the predicted time period The corrected predicted vehicle speed at that time is: ;

[0098] In the formula, Indicates the current location at the monitoring point. target vehicle the monitoring point in the prediction period the correction predicted vehicle speed at the time of reaching the monitoring point in the prediction period; the preliminary predicted vehicle speed of the target vehicle currently at the monitoring point ; the preliminary predicted vehicle speed of the target vehicle currently at the monitoring point ; the predicted vehicle speed change amount at the corresponding position at each time point in the period from the monitoring point to the monitoring point ; the predicted vehicle speed change amount at the corresponding position at each time point in the period from the monitoring point to the monitoring point ; the average value of the time required for the corresponding multiple reference vehicles to reach the monitoring point from the monitoring point ; the average value of the time required for the corresponding multiple reference vehicles to reach the monitoring point from the monitoring point ;

[0099] Similarly, the correction predicted vehicle speed of the target vehicle at each monitoring point in the prediction period can be obtained, and then the correction predicted vehicle speed sequence of each vehicle in each monitoring region in the prediction period can be obtained.

[0100] Then, based on the correction predicted vehicle speed sequence, Lagrange interpolation processing is performed to obtain the fitted vehicle speed of the vehicle at each position in the monitoring region on the expressway. Specifically, after obtaining the correction predicted vehicle speed of the target vehicle at each monitoring point in each monitoring region in the prediction period, it is considered that the driving of a normal vehicle on the expressway is a relatively stable acceleration and deceleration process, and sudden speed reduction and speed increase do not occur. Therefore, after obtaining the speed of the vehicle passing through each monitoring point, the fitted vehicle speed of the vehicle at each position on the expressway can be directly obtained according to Lagrange interpolation or simple linear interpolation.

[0101] Finally, the fitted vehicle speed is arranged according to the position sequence of the vehicle on the expressway to obtain the predicted vehicle speed sequence of each vehicle in each monitoring region in the prediction period. Specifically, taking the target vehicle as an example, the corresponding fitted vehicle speed of the target vehicle is arranged according to the position sequence of the target vehicle on the expressway to form the predicted vehicle speed sequence of the target vehicle in the target monitoring region. Similarly, the predicted vehicle speed sequence of each vehicle in each monitoring region in the prediction period can be obtained.

[0102] S207: Based on the predicted vehicle speed sequence, a predicted vehicle flow sequence in each monitoring region in the prediction period is obtained.

[0103] ​​​​​Based on the predicted vehicle speed sequence, the predicted traffic flow sequence for each monitoring area within the predicted time period is obtained. Specifically, after obtaining the vehicle speeds at each location within the predicted time period, the traffic flow can be calculated using the integral relationship of speed-distance-time. ,in, Let v represent distance, t represent speed, and t represent time. The target vehicle... In the target monitoring area The predicted vehicle speed sequence within the time period is transformed into the target vehicle within the prediction time period. In the target monitoring area Similarly, the predicted vehicle speed time series within the predicted time period is obtained for each vehicle in each monitoring area. Then, based on the predicted vehicle speed time series, the predicted traffic flow sequence for each monitoring area within the predicted time period is obtained.

[0104] S300: Based on traffic flow, vehicle speed, and greenhouse gas content during historical periods, a neural network model is trained to predict the greenhouse gas content in each monitoring area.

[0105] Based on historical traffic flow, vehicle speed, and greenhouse gas concentrations, a neural network model is trained to predict greenhouse gas concentrations in each monitoring area. Specifically, the vehicle speed time series and traffic flow time series of various vehicle types within the target monitoring area during past reference periods are used as inputs to the neural network model. The greenhouse gas concentration time series detected by monitoring drones within the target monitoring area are used as the outputs of the neural network model. This training process yields a neural network model that predicts changes in greenhouse gas concentrations within the target monitoring area using vehicle speed and traffic flow sequences. Similarly, neural network models are trained to predict greenhouse gas concentrations in each monitoring area.

[0106] S400: Based on a neural network model, combining predicted vehicle speed sequences and predicted traffic flow sequences, the predicted greenhouse gas content sequences for each monitoring area within the predicted time period are obtained.

[0107] Based on a neural network model, combining predicted vehicle speed sequences and predicted traffic flow sequences, the predicted greenhouse gas content sequences for each monitoring area within the prediction period are obtained. Specifically, the predicted vehicle speed sequences and predicted traffic flow sequences for each monitoring area are used as inputs to a trained neural network model; the output is the predicted greenhouse gas content sequences for each monitoring area within the prediction period.

[0108] S500: Based on the historical greenhouse gas content sequence and the predicted greenhouse gas content sequence, analyze the consistency of the predicted greenhouse gas content in each monitoring area to obtain the degree of importance attached to each monitoring area.

[0109] Based on the historical and predicted greenhouse gas concentrations, the consistency of greenhouse gas concentration predictions across monitoring areas is analyzed to determine the level of importance placed on each monitoring area. Further analysis includes:

[0110] First, the DTW similarity between historical and predicted greenhouse gas concentrations is calculated to obtain the predicted past consistency of greenhouse gas concentrations in each monitoring area. Specifically, when the changes in greenhouse gas concentrations in a monitoring area differ significantly from the past, it requires focused monitoring to strengthen surveillance. Therefore, the DTW similarity between historical and predicted greenhouse gas concentrations in each monitoring area is calculated. This DTW similarity is the negative correlation mapping value between the historical and predicted greenhouse gas concentrations in each monitoring area. For example, the reciprocal of the DTW distance can be used as the negative correlation mapping value. This DTW similarity is recorded as the predicted past consistency, thus obtaining the predicted past consistency of greenhouse gas concentrations in each monitoring area.

[0111] Then, based on the consistency of predictions over the past period, the level of importance for each monitoring area is determined. Specifically, the smaller the DTW similarity between the historical and predicted greenhouse gas concentrations in each monitoring area, the greater the environmental (greenhouse gas) changes within that monitoring area during the prediction period, and the more attention needs to be paid to these changes. Based on this, the level of importance for that monitoring area is calculated. The degree of importance attached to The formula for calculation is:

[0112] In the formula, Indicates the monitoring area The degree of importance attached to it; Indicates the monitoring area Historical consistency in predicting greenhouse gas concentrations; This represents a non-linear activation function.

[0113] Similarly, this reveals the level of importance attached to each monitoring area.

[0114] S600: Adjust the density of drones in each monitoring area according to the level of importance attached to it.

[0115] The importance degree of each monitoring area in the prediction period is obtained through the foregoing steps, and when the importance degree of a monitoring area is greater than or equal to a preset threshold (0.3), it is considered that the change of the greenhouse gas in the monitoring area in the prediction period is inconsistent with the past, which indicates that the monitoring area needs to be paid attention to, and therefore the unmanned aerial vehicle density in the monitoring area needs to be increased.

[0116] Therefore, the unmanned aerial vehicle density in each monitoring area is adjusted according to the importance degree. Specifically, the following steps are included.

[0117] A preset importance degree threshold is provided, and the importance degree threshold includes a large threshold (0.85), a medium threshold (0.5) and a lower threshold (0.3);

[0118] If the importance degree of a monitoring area is greater than or equal to the large threshold, that is, the unmanned aerial vehicle density in the monitoring area is doubled.

[0119] If the importance degree of a monitoring area is greater than or equal to the medium threshold and less than the large threshold, that is, the unmanned aerial vehicle density in the monitoring area is increased by 0.5 times.

[0120] If the importance degree of a monitoring area is greater than or equal to the lower threshold and less than the medium threshold, that is, the unmanned aerial vehicle density in the monitoring area is increased by 0.25 times.

[0121] If the importance degree of a monitoring area is less than the lower threshold, that is, the unmanned aerial vehicle density in the monitoring area is not adjusted.

[0122] Please refer to Figure 3 which shows the basic components of an ecological environment monitoring system along the highway area according to an embodiment of the present application.

[0123] As shown in Figure 3 , an ecological environment monitoring system along the highway area includes a memory 10 and a processor 20, wherein:

[0124] The memory 10 is used to store program codes.

[0125] The processor 20 is configured to read program codes stored in the memory 10, and perform the following steps: dividing a highway area into a plurality of monitoring areas, collecting greenhouse gas content, traffic volume, vehicle speed, vehicle type and environmental parameters in each monitoring area; analyzing environmental similarity at different times and driving similarity of different vehicles of the same vehicle type according to the vehicle speed, the vehicle type and the environmental parameters, obtaining a predicted vehicle speed sequence and a predicted traffic volume sequence of each vehicle in each monitoring area in a prediction period; training a neural network model for predicting greenhouse gas content in each monitoring area according to traffic volume, vehicle speed and greenhouse gas content in a historical period; obtaining a predicted greenhouse gas content sequence in each monitoring area in the prediction period based on the neural network model, in combination with the predicted vehicle speed sequence and the predicted traffic volume sequence; analyzing prediction consistency of greenhouse gas content in each monitoring area according to the greenhouse gas content sequence in the historical period and the predicted greenhouse gas content sequence, obtaining a degree of attention of each monitoring area; and adjusting the density of unmanned aerial vehicles in each monitoring area according to the degree of attention.

[0126] Further, the processor 20 includes a data collection module 21, a road condition prediction module 22, a greenhouse gas content prediction module 23 and an unmanned aerial vehicle density adjustment module 24. Wherein:

[0127] The data collection module 21 is configured to divide a highway area into a plurality of monitoring areas, and collect greenhouse gas content, traffic volume, vehicle speed, vehicle type and environmental parameters in each monitoring area.

[0128] The road condition prediction module 22 is configured to analyze environmental similarity at different times and driving similarity of different vehicles of the same vehicle type according to the vehicle speed, the vehicle type and the environmental parameters, and obtain a predicted vehicle speed and a predicted traffic volume of each vehicle in each monitoring area in a prediction period.

[0129] The greenhouse gas content prediction module 23 is configured to train a neural network model for predicting greenhouse gas content in each monitoring area according to traffic volume, vehicle speed and greenhouse gas content in a historical period, and obtain a predicted greenhouse gas content in each monitoring area in a prediction period based on the neural network model, in combination with the predicted vehicle speed and the predicted traffic volume.

[0130] The unmanned aerial vehicle density adjustment module 24 is configured to analyze prediction consistency of greenhouse gas content in each monitoring area according to the greenhouse gas content in the historical period and the predicted greenhouse gas content, obtain a degree of attention of each monitoring area, and adjust the density of unmanned aerial vehicles in each monitoring area according to the degree of attention.

[0131] It is to be noted that the sequential order of the above-described embodiments of the present application only for the purpose of description, but not the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0132] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments.

Claims

1. A method for monitoring the ecological environment along a highway, characterized in that, The method includes: The highway area is divided into several monitoring zones, and greenhouse gas content, traffic flow, vehicle speed, vehicle type and environmental parameters are collected in each monitoring zone. Based on the vehicle speed, combined with the vehicle type and environmental parameters, the environmental similarity at different times and the driving similarity of different vehicles of the same vehicle type are analyzed to obtain the predicted vehicle speed sequence and predicted traffic flow sequence for each vehicle in each monitoring area within the prediction period, including: Based on the environmental parameters, the environmental similarity of each monitoring point in each monitoring area at different times is analyzed to obtain the reference time period at the current time of each monitoring point in the monitoring area and the environmental reference degree of the reference time period; Based on the vehicle speed and environmental parameters of all vehicles of the same model at each monitoring point at the current time and within the reference time period, and combined with the environmental referenceability, a preliminary predicted vehicle speed sequence for each vehicle of each model in each monitoring area is obtained within the prediction time period. Based on the vehicle speed, a known speed change sequence for each vehicle type in each monitoring area during the historical period is obtained; By analyzing the differences in vehicle speeds among different vehicles of the same model and combining this with the environmental similarity, the driving similarity of different vehicles of the same model within each monitoring area is obtained. By combining the known vehicle speed change sequence with the driving similarity, the predicted vehicle speed change of each vehicle at each monitoring point in each monitoring area is obtained; Based on the predicted vehicle speed change and combined with the preliminary predicted vehicle speed sequence, the predicted vehicle speed sequence of each vehicle in each monitoring area within the prediction period is obtained. Based on the predicted vehicle speed sequence, the predicted traffic flow sequence for each monitoring area within the predicted time period is obtained; Based on traffic flow, vehicle speed, and greenhouse gas content during historical periods, a neural network model is trained to predict the greenhouse gas content in each of the monitoring areas. Based on the neural network model, and combined with the predicted vehicle speed sequence and the predicted traffic flow sequence, the predicted greenhouse gas content sequence for each monitoring area within the predicted time period is obtained. Based on the historical greenhouse gas content sequence and the predicted greenhouse gas content sequence, the consistency of the predicted greenhouse gas content in each monitoring area is analyzed to obtain the degree of importance attached to each monitoring area. The density of drones in each of the aforementioned monitoring areas will be adjusted according to the level of importance attached to them.

2. The method for monitoring the ecological environment along a highway according to claim 1, characterized in that, Based on the environmental parameters, the environmental similarity of each monitoring point within each monitoring area at different times is analyzed to obtain the reference time period at the current time of each monitoring point within the monitoring area and the environmental referenceability of the reference time period, including: Based on the environmental parameters, the differences of the same type of environmental parameters corresponding to each monitoring point at different times in each monitoring area are analyzed. All types of environmental parameters are traversed to obtain the environmental similarity of each monitoring point at different times in each monitoring area. A preset environmental similarity threshold is set. If the environmental similarity between all times in the past period and the current time is greater than or equal to the environmental similarity threshold, then the past period is recorded as the reference time period for the current time, thus obtaining multiple reference time periods for the current time at each monitoring point in the monitoring area. Calculate the mean environmental similarity between each time point within the reference time period and the current time point to obtain the environmental referenceability of the reference time period.

3. The method for monitoring the ecological environment along a highway according to claim 1, characterized in that, Based on the vehicle speed and environmental parameters of all vehicles of the same model at each monitoring point at the current time and within its reference time period, and combined with the environmental referenceability, a preliminary predicted vehicle speed sequence for each vehicle of each model within each monitoring area is obtained within the prediction time period, including: based on The prediction model takes as input the vehicle speed and environmental parameters of all vehicles of the same model at each monitoring point at the current time and within multiple reference time periods. The environmental referenceability is used as the reference weight for the vehicle speed and environmental parameters in each reference time period. The model outputs a preliminary predicted vehicle speed sequence for each vehicle of each model in each monitoring area within the prediction time period.

4. The method for monitoring the ecological environment along a highway as described in claim 1, characterized in that, Analyzing the differences in vehicle speeds among different vehicles of the same model, and combining this with environmental similarity, yields the driving similarity of different vehicles of the same model within each monitoring area, including: Calculate the absolute value of the speed difference between different vehicles of the same model passing through the same monitoring point, combine it with the environmental similarity at different times corresponding to different vehicles of the same model passing through the same monitoring point, and traverse all monitoring points that different vehicles of the same model have passed through together to obtain the driving similarity of different vehicles of the same model in each monitoring area.

5. The method for monitoring the ecological environment along a highway according to claim 1, characterized in that, Based on the predicted vehicle speed change and combined with the preliminary predicted vehicle speed sequence, the predicted vehicle speed sequence for each vehicle within each monitoring area during the prediction period is obtained, including: Based on the predicted vehicle speed change and combined with the preliminary predicted vehicle speed sequence, the corrected predicted vehicle speed sequence for each vehicle in each monitoring area during the prediction period is obtained. Based on the corrected predicted vehicle speed sequence, Lagrange interpolation is performed to obtain the fitted vehicle speed at each location within the monitoring area on the highway. The fitted vehicle speeds are arranged according to the order of the vehicles' positions on the highway to obtain the predicted vehicle speed sequence of each vehicle in each monitoring area within the predicted time period.

6. The method for monitoring the ecological environment along a highway as described in claim 1, characterized in that, Based on the historical and predicted greenhouse gas concentrations, the consistency of predicted greenhouse gas concentrations within each monitoring area is analyzed to determine the level of importance attached to each monitoring area, including: Calculate the levels of greenhouse gases in historical periods and the predicted levels of greenhouse gases. Similarity is used to obtain the past consistency of predicted greenhouse gas content in each monitoring area; Based on the predicted consistency over the past, the degree of importance attached to each of the monitored areas is obtained.

7. The method for monitoring the ecological environment along a highway according to claim 1, characterized in that, Based on the stated level of importance, the density of drones within each monitoring area is adjusted, including: A preset importance threshold is defined, which includes a high threshold, a medium threshold, and a lower threshold. If the level of importance attached to the monitored area is greater than or equal to the large threshold, the density of drones in the monitored area will be doubled. If the level of importance of the monitored area is greater than or equal to the medium threshold and less than the large threshold, then the density of drones in the monitored area will be increased by 0.5 times. If the level of importance of the monitored area is greater than or equal to the lower threshold and less than the middle threshold, then the density of drones in the monitored area will be increased by 0.25 times.

8. An ecological environment monitoring system along a highway, characterized in that, The system includes: a memory and a processor, wherein: The memory is used to store program code; The processor is configured to read program code stored in the memory and execute the method as described in any one of claims 1 to 7.

9. The ecological environment monitoring system along the highway according to claim 8, characterized in that, The processor includes: The data acquisition module is used to divide the highway area into several monitoring zones and collect greenhouse gas content, traffic flow, vehicle speed, vehicle type and environmental parameters in each monitoring zone; The road condition prediction module is used to analyze the environmental similarity at different times and the driving similarity of different vehicles of the same vehicle type based on the vehicle speed, combined with the vehicle type and the environmental parameters, to obtain the predicted vehicle speed and predicted traffic flow of each vehicle in each monitoring area during the prediction period. The road condition prediction module is specifically used to analyze the environmental similarity of each monitoring point in each monitoring area at different times based on the environmental parameters, and to obtain the reference time period at the current time of each monitoring point in the monitoring area and the environmental reference degree of the reference time period. Based on the vehicle speed and environmental parameters of all vehicles of the same model at each monitoring point at the current time and within the reference time period, and combined with the environmental referenceability, a preliminary predicted vehicle speed sequence for each vehicle of each model in each monitoring area is obtained within the prediction time period. Based on the vehicle speed, a known speed change sequence for each vehicle type in each monitoring area during the historical period is obtained; By analyzing the differences in vehicle speeds among different vehicles of the same model and combining this with the environmental similarity, the driving similarity of different vehicles of the same model within each monitoring area is obtained. By combining the known vehicle speed change sequence with the driving similarity, the predicted vehicle speed change of each vehicle at each monitoring point in each monitoring area is obtained; Based on the predicted vehicle speed change and combined with the preliminary predicted vehicle speed sequence, the predicted vehicle speed sequence of each vehicle in each monitoring area within the prediction period is obtained. Based on the predicted vehicle speed sequence, the predicted traffic flow sequence for each monitoring area within the predicted time period is obtained; The greenhouse gas content prediction module is used to train a neural network model to predict the greenhouse gas content in each of the monitoring areas based on traffic flow, vehicle speed and greenhouse gas content in historical time periods; and to obtain the predicted greenhouse gas content in each of the monitoring areas in the prediction time period based on the neural network model and the predicted vehicle speed and the predicted traffic flow. The drone density adjustment module is used to analyze the historical consistency of greenhouse gas content with predicted greenhouse gas content in each monitoring area based on historical data and predicted greenhouse gas content, thereby obtaining the level of importance attached to each monitoring area; and to adjust the drone density in each monitoring area based on the level of importance attached.

Images