Municipal environmental sanitation cleaning task scheduling method and system based on big data analysis

Abstract

The invention relates to the technical field of computers, in particular to a municipal environmental sanitation cleaning task scheduling method and system based on big data analysis, and the method comprises the steps: obtaining the historical garbage amount data of an urban traffic station, and carrying out the garbage amount partitioning through a K-means + + clustering algorithm according to the historical garbage amount data; acquiring urban meteorological data, and predicting an urban weather condition through a wavelet transform and LSTM neural network hybrid model according to the urban meteorological data; the state of the sanitation vehicle is monitored in real time through a vehicle-mounted sensor, and the position and the driving track of the sanitation vehicle are tracked in real time through the GPS positioning technology; according to the garbage quantity partition, the predicted urban weather condition, the environmental sanitation vehicle state and the position and the driving track of the environmental sanitation vehicle, a cleaning task is distributed to the environmental sanitation vehicle; and real-time traffic conditions are obtained, and the sanitation vehicles are scheduled to go to the garbage quantity subareas to execute cleaning tasks in combination with the real-time traffic conditions. The environmental sanitation management efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and system for scheduling municipal sanitation cleaning tasks based on big data analysis. Background Art

[0002] With the acceleration of urbanization, urban sanitation management has become a critical issue affecting residents' quality of life and environmental protection. Urban sanitation work requires real-time garbage detection, the generation of garbage removal tasks, and the rational allocation and scheduling of tasks to the appropriate sanitation vehicles and personnel.

[0003] The traditional sanitation model relies on manual scheduling and inspections, resulting in low mechanization, high labor costs, and a lack of scientific scheduling, which leads to wasted resources, insufficient manpower, and difficult supervision, making it difficult to achieve refined management. Furthermore, traditional scheduling methods cannot quickly respond to emergencies such as garbage accumulation and inclement weather, resulting in inefficient sanitation management. Summary of the Invention

[0004] Based on this, it is necessary to provide a municipal sanitation cleaning task scheduling method and system based on big data analysis that can improve the efficiency of sanitation management in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for scheduling municipal sanitation cleaning tasks based on big data analysis, the method comprising: Obtaining historical garbage volume data at urban transportation stations, and partitioning the garbage volume using a K-means++ clustering algorithm based on the historical garbage volume data; Obtain urban meteorological data, and predict urban weather conditions based on the urban meteorological data using a wavelet transform and LSTM neural network hybrid model; The status of sanitation vehicles can be monitored in real time through on-board sensors, and the location and driving trajectory of sanitation vehicles can be tracked in real time through GPS positioning technology; Assigning cleaning tasks to the sanitation vehicles based on the garbage volume zones, predicted urban weather conditions, the status of the sanitation vehicles, and the locations and driving trajectories of the sanitation vehicles; Acquire real-time traffic conditions, and in combination with the real-time traffic conditions, dispatch the sanitation vehicles to the garbage volume partition to perform the cleaning task.

[0006] In one embodiment, obtaining historical garbage volume data of a traffic station and partitioning the garbage volume using a K-means++ clustering algorithm based on the historical garbage volume data includes: Obtaining traffic card swiping data, and extracting the temporal distribution characteristics of passenger flow at the traffic station through passenger flow time series analysis based on the traffic card swiping data; Obtaining data on points of interest around transportation stations, and obtaining, based on the data on points of interest around transportation stations, a spatial relationship between land use and passenger flow around transportation stations using a geographically weighted regression model; Obtain historical garbage volume data of urban transportation stations, and perform garbage volume zoning using the K-means++ clustering algorithm based on the historical garbage volume data, the temporal distribution characteristics of passenger flow, and the spatial relationship between land use and passenger flow around the transportation stations.

[0007] In one embodiment, obtaining urban meteorological data and predicting urban weather conditions based on the urban meteorological data using a wavelet transform and LSTM neural network hybrid model includes: Acquire urban meteorological data, and perform noise reduction on the urban meteorological data using a wavelet transform threshold noise reduction method; Based on the denoised urban meteorological data, the urban weather conditions are predicted through the LSTM neural network.

[0008] In one embodiment, allocating cleaning tasks to the sanitation vehicles based on the garbage volume zones, predicted urban weather conditions, the status of the sanitation vehicles, and the locations and driving trajectories of the sanitation vehicles includes: Real-time monitoring of the overflow level of the garbage bins in the garbage volume partition; Prioritizing the waste volume zones based on the overflow level of the waste bins and predicted city weather conditions; Cleaning tasks are assigned to the sanitation vehicles according to the set priority, the status of the sanitation vehicles, and the location and driving trajectory of the sanitation vehicles.

[0009] In one embodiment, acquiring real-time traffic conditions and dispatching the sanitation vehicles to the garbage volume partition to perform the cleaning task based on the real-time traffic conditions includes: Acquiring real-time traffic conditions and obtaining current road congestion data based on the real-time traffic conditions; A dynamic environment model is established using the congestion-aware A* algorithm to divide the road into multiple grid nodes, and the real-time congestion level of each grid node is obtained based on the current road congestion data; Introducing a congestion awareness mechanism to calculate a path cost based on the real-time congestion level; An optimal path is obtained according to the path cost value, and the sanitation vehicle is dispatched according to the optimal path to the garbage volume partition to perform the cleaning task.

[0010] In one embodiment, introducing a congestion awareness mechanism and calculating a path cost according to the real-time congestion level includes: Calculating an actual cost from the grid node to a starting point and an estimated cost from the grid node to an end point; Introducing a congestion awareness mechanism to assign weight coefficients to the actual cost, the estimated cost, and the real-time congestion level, so as to dynamically adjust the priority of each cost; A path cost value is calculated according to the actual cost value, the estimated cost value, the real-time congestion level, and the weight coefficient.

[0011] In one embodiment, the acquiring of real-time traffic conditions and, in combination with the real-time traffic conditions, dispatching the sanitation vehicles to the garbage volume partition to perform the cleaning task further includes: If multiple sanitation vehicles need to be dispatched to work together in a large-scale operation scenario, the distributed vehicle group collaborative decision-making algorithm, combined with the real-time traffic conditions, will be used to dispatch the multiple sanitation vehicles to the garbage volume partition to collaboratively perform the cleaning task.

[0012] In a second aspect, the present application also provides a municipal sanitation cleaning task scheduling device based on big data analysis. The device includes: The garbage volume partitioning module is used to obtain historical garbage volume data of urban transportation stations and perform garbage volume partitioning based on the historical garbage volume data using the K-means++ clustering algorithm; A weather forecasting module is used to obtain urban meteorological data and, based on the urban meteorological data, predict urban weather conditions through a hybrid model of wavelet transform and LSTM neural network; The vehicle data acquisition module is used to monitor the status of sanitation vehicles in real time through on-board sensors, and to track the location and driving trajectory of sanitation vehicles in real time through GPS positioning technology; A cleaning task allocation module is used to allocate cleaning tasks to the sanitation vehicles based on the garbage volume partition, the predicted urban weather conditions, the status of the sanitation vehicles, and the location and driving trajectory of the sanitation vehicles; The sanitation vehicle scheduling module is used to obtain real-time traffic conditions and, based on the real-time traffic conditions, schedule the sanitation vehicles to go to the garbage volume partition to perform the cleaning task.

[0013] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed: Obtain historical garbage amount data of urban traffic sites, and perform garbage amount zoning through a K-means++ clustering algorithm according to the historical garbage amount data; Obtain urban meteorological data, and predict urban weather conditions through a wavelet transform and LSTM neural network hybrid model according to the urban meteorological data; Real-time monitor sanitation vehicle states through vehicle-mounted sensors, and real-time track positions and driving trajectories of the sanitation vehicles through GPS positioning technology; Assign cleaning tasks to the sanitation vehicles according to the garbage amount zoning, predicted urban weather conditions, sanitation vehicle states, and positions and driving trajectories of the sanitation vehicles; Obtain real-time traffic conditions, and dispatch the sanitation vehicles to the garbage amount zoning to perform the cleaning tasks in combination with the real-time traffic conditions.

[0014] In a fourth aspect, the present application further provides a computer readable storage medium. The computer readable storage medium has a computer program stored thereon, and the computer program is executed by a processor to implement the following steps: Obtain historical garbage amount data of urban traffic sites, and perform garbage amount zoning through a K-means++ clustering algorithm according to the historical garbage amount data; Obtain urban meteorological data, and predict urban weather conditions through a wavelet transform and LSTM neural network hybrid model according to the urban meteorological data; Real-time monitor sanitation vehicle states through vehicle-mounted sensors, and real-time track positions and driving trajectories of the sanitation vehicles through GPS positioning technology; Assign cleaning tasks to the sanitation vehicles according to the garbage amount zoning, predicted urban weather conditions, sanitation vehicle states, and positions and driving trajectories of the sanitation vehicles; Obtain real-time traffic conditions, and dispatch the sanitation vehicles to the garbage amount zoning to perform the cleaning tasks in combination with the real-time traffic conditions.

[0015] In summary, the present application has the following beneficial technical effects: According to historical garbage amount data, garbage amount partitioning is performed through a K-means++ clustering algorithm, so that a more scientific cleaning planning of sanitation vehicles is realized, and sanitation vehicle scheduling paths are also optimized; through a wavelet transform and an LSTM neural network hybrid model, city weather conditions are predicted, the influence of weather changes on sanitation operations can be predicted in advance, so that operation time is reasonably arranged, and smooth sanitation work is ensured; through vehicle-mounted sensors and GPS positioning technology, the positions, driving trajectories, vehicle states and other key information of sanitation vehicles are obtained in real time, providing accurate basis for subsequent scheduling; combined with real-time traffic conditions, sanitation vehicles are dispatched to perform cleaning tasks, which can optimize the driving paths of sanitation vehicles, reduce empty running rate and repeated operations, and thus improve sanitation management efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of a municipal sanitation cleaning task scheduling method based on big data analysis in an embodiment is shown in the figure. Figure 2 A flowchart of a municipal sanitation cleaning task scheduling method based on big data analysis in another embodiment is shown in the figure. Figure 3 A block diagram of a municipal sanitation cleaning task scheduling device based on big data analysis in an embodiment is shown in the figure. Figure 4 An internal structure diagram of a computer device in an embodiment is shown in the figure. DETAILED DESCRIPTION

[0017] The embodiments of the present application provide a municipal sanitation cleaning task scheduling method and system based on big data analysis.

[0018] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms, and should not be interpreted as being limited to the embodiments described herein, but rather, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes, and are not intended to limit the scope of protection of the present application.

[0019] In the description of the embodiments of the present application, the term “comprising” and similar terms should be understood as open-ended inclusion, i.e. “including but not limited to”. The term “based on” should be understood as “at least partially based on”. The term “one embodiment” or “the embodiment” should be understood as “at least one embodiment”. The terms “first”, “second”, etc. can refer to different or same objects. Other explicit and implicit definitions can also be included below.

[0020] For ease of understanding, the specific flow of the embodiments of the present application is described below. Please refer to Figure 1 In one embodiment of the present invention, a method for scheduling municipal sanitation cleaning tasks based on big data analysis includes: S100, obtaining historical garbage volume data of urban transportation stations, and partitioning the garbage volume using a K-means++ clustering algorithm based on the historical garbage volume data.

[0021] Specifically, based on historical garbage volume data, we used the K-means++ algorithm to perform cluster analysis on transportation stations (such as subway stations) to obtain multiple garbage volume partitions. The algorithm steps are as follows: Step 1: Randomly select a sample point from the data set X as the first initial cluster center ; Step 2: Traverse each sample point in X and calculate its distance from the current cluster center The Euclidean distance is used to filter out the shortest distance Then according to Calculate the probability of each sample being selected as the next cluster center and determine the new cluster center according to the roulette method; Step 3: Repeat step 2 until you select Cluster centers , and then the clusters are obtained ; Step 4: Calculate all sample points in X and The distance between each cluster center in , and assign it to the class corresponding to the cluster center with the shortest distance; Step 5: For , calculate and update the cluster center to which the sample point belongs ; Step 6: Repeat steps 4 and 5 until the cluster center is stable or the algorithm reaches the maximum number of iterations.

[0022] Furthermore, the K-means++ algorithm uses unsupervised learning, making traditional evaluation metrics inappropriate for describing its clustering effectiveness. Therefore, the Silhouette Coefficient, Davies-Bouldi Index, and Calinski-Harabaz Index were used to evaluate clustering effectiveness. The Silhouette Coefficient is an internal evaluation metric based on the distance between data points, measuring the compactness of a data point within its cluster and its separation from other clusters. The Silhouette Coefficient ranges from -1 to 1, with larger values ​​indicating better clustering effectiveness. The Davies-Bouldi Index and Calinski-Harabaz Index are internal evaluation metrics based on intra- and inter-cluster distances, respectively, measuring the compactness and separation of clusters.

[0023] In this example, the K-means++ algorithm achieves higher clustering effectiveness and stability than the traditional K-means algorithm, resulting in more reasonable and accurate clustering results. For garbage volume zoning at urban transportation stations, the K-means++ algorithm can more effectively identify areas with high garbage production, thereby providing better routing for sanitation vehicle dispatch.

[0024] S200 obtains urban meteorological data and predicts urban weather conditions based on the urban meteorological data through a hybrid model of wavelet transform and LSTM neural network.

[0025] Specifically, wavelet transform is a tool that uses the translation and scaling of wavelet functions to form time domain and frequency windows of different resolutions, thereby completing the high and low frequency multi-scale decomposition of the signal and realizing the localized analysis of the signal time and frequency. A wavelet function is a type of oscillatory wave with a finite length that decays rapidly to zero. Its amplitude exhibits alternating positive and negative amplitudes, with an average value of zero. Wavelet functions can have asymmetric and irregular shapes, and the waveform can be scaled and translated to match different input signals. Because different wavelet functions have their own strengths and weaknesses when processing time series signals, different wavelet functions are suitable for denoising different types of time series data. No single wavelet function is universally applicable to all types of time series signals. Meteorological data is a type of time series data, and using the sym8 wavelet function for decomposition and denoising is most effective. Furthermore, LSTM neural networks are used for weather forecasting because they meet the needs of meteorological time series data forecasting and demonstrate excellent performance in this area. Specifically, firstly, urban meteorological data is acquired and features are extracted to obtain meteorological signals. The sym8 wavelet function is selected to perform wavelet decomposition on the meteorological signals, and the wavelet coefficients obtained after decomposition are denoised by the wavelet threshold denoising method to solve the problem of noise affecting the accuracy of the prediction results. The denoised meteorological data is then input into the LSTM neural network for prediction to obtain the predicted urban weather conditions.

[0026] In this example, combined with the noise-reduced data after wavelet transform, the LSTM model significantly outperformed single LSTM models or traditional statistical methods in predicting indicators such as average wind speed, surface temperature, and precipitation. This hybrid wavelet transform and LSTM neural network model achieved lower prediction errors and greater stability across multiple meteorological indicators, providing a scientific basis for meteorological warnings.

[0027] S300 monitors the status of sanitation vehicles in real time through on-board sensors, and tracks the location and driving trajectory of sanitation vehicles in real time through GPS positioning technology.

[0028] Specifically, on-board sensors and GPS positioning technology are the core means for real-time monitoring and tracking of sanitation vehicles. On-board sensors collect a variety of operational data from the vehicle, while GPS provides real-time information on the vehicle's location and trajectory. This combination enables managers to fully understand the operational status of sanitation vehicles, enabling intelligent management. On-board sensors enable real-time monitoring of sanitation vehicle status, including key data such as speed, fuel consumption, and operating conditions. This data not only helps managers understand the real-time status of sanitation vehicles but also, through data analysis and intelligent algorithms, optimizes sanitation vehicle scheduling and routes, reducing inefficient operations and idle driving. GPS positioning technology tracks the location and trajectory of sanitation vehicles in real time, allowing managers to visually visualize vehicle operations on a map. This real-time tracking not only helps optimize scheduling but also provides timely alerts for vehicle anomalies, such as speeding, excessive parking, or exceeding operating ranges.

[0029] S400 allocates cleaning tasks to sanitation vehicles based on garbage volume zoning, predicted urban weather conditions, sanitation vehicle status, and the location and driving trajectory of the sanitation vehicles.

[0030] Specifically, based on historical garbage volume data, multiple traffic stops are clustered and divided into garbage volume zones. Cleaning tasks in different areas are then assigned to corresponding vehicles, ensuring timely garbage removal in every area. Based on the city's predicted weather conditions, different operation methods can be selected. For example, during rainy days, water suction trucks are dispatched to low-lying road sections; during hot weather, circulatory spraying is performed in hot spots at traffic stops. Sanitation vehicles are dispatched based on their status, location, and driving trajectory, prioritizing those near garbage volume zones. Vehicles running low on fuel are assigned short-distance missions or relocated to areas near gas stations.

[0031] In this embodiment, cleaning tasks are assigned to sanitation vehicles based on the garbage volume zoning, predicted urban weather conditions, sanitation vehicle status, and the location and driving trajectory of the sanitation vehicles. This can achieve intelligent task allocation, thereby improving the intelligence level of urban sanitation work and improving work efficiency and quality.

[0032] S500 obtains real-time traffic conditions and, based on the real-time traffic conditions, dispatches sanitation vehicles to garbage volume zones to perform cleaning tasks.

[0033] Specifically, based on the real-time traffic conditions, the degree of traffic congestion is judged, and based on the degree of traffic congestion, the optimal route is planned for the sanitation vehicles, and the sanitation vehicles are dispatched according to the optimal route to the garbage volume area to perform cleaning tasks.

[0034] In one embodiment, Figure 2 As shown, S100 includes: S120, obtaining transportation card swiping data, and extracting the time distribution characteristics of passenger flow at the transportation station through passenger flow time series analysis based on the transportation card swiping data; S140, obtaining data of points of interest around the transportation station, and obtaining a spatial relationship between land use and passenger flow around the transportation station through a geographically weighted regression model based on the data of points of interest around the transportation station; S160, historical garbage volume data of urban transportation stations is obtained, and garbage volume partitioning is performed using a K-means++ clustering algorithm based on the historical garbage volume data, temporal distribution characteristics of passenger flow, and the spatial relationship between land use and passenger flow around the transportation stations.

[0035] Specifically, based on subway card swipe data and data on points of interest (POIs) near stations, temporal and spatial variables were extracted through passenger flow time series analysis and a geographically weighted regression model. Combined with historical garbage volume data, the K-means++ clustering algorithm was applied to zoning garbage volume, classifying transportation stations into different levels, with higher levels indicating greater garbage production. Specifically, in the temporal dimension, analysis of station card swipe data revealed the temporal distribution characteristics of passenger flow at each station. For example, significant differences were found in passenger flow distribution between weekdays and weekends, and different types of stations exhibited distinct passenger flow intensities and distribution patterns during peak hours in the morning and evening. A two-dimensional matrix was constructed, using the station's inbound and outbound passenger flow time series data as clustering variables, further providing a foundation for cluster analysis. In the spatial dimension, a geographically weighted regression (GWR) model was used to analyze the impact of POIs near stations on passenger flow. A regression model was constructed by selecting data from 14 types of points of interest (POIs), including catering services, shopping services, and healthcare, as independent variables. This model, combined with weekday morning peak and evening peak inbound and outbound passenger flow, as well as weekend passenger flow throughout the day, served as dependent variables. This model reveals the spatial relationship between land use and passenger flow around stations, providing richer spatial information for cluster analysis. After determining the clustering variables in the spatiotemporal dimensions, the K-means++ clustering algorithm was used to cluster transportation stations. Based on the clustering results, garbage volume was zoned.

[0036] In this embodiment, the temporal distribution characteristics of passenger flow, the spatial relationship of land use, and garbage volume data are combined to achieve refined zoning of garbage volume through the K-means++ clustering algorithm. This not only helps city managers identify garbage hotspots, but also provides a scientific basis for the formulation of urban garbage management strategies.

[0037] In one embodiment, obtaining urban meteorological data and predicting urban weather conditions based on the urban meteorological data using a wavelet transform and LSTM neural network hybrid model includes: Obtain urban meteorological data and perform noise reduction on the data using the wavelet transform threshold noise reduction method; then use the LSTM neural network to predict urban weather conditions based on the noise-reduced urban meteorological data.

[0038] Specifically, features are extracted from urban meteorological data to obtain meteorological signals. A wavelet function is then selected and decomposed using a specified number of layers. A threshold estimation criterion is used to select an appropriate threshold. The corresponding threshold function is then used to decompose the wavelet transform at each scale, resulting in wavelet coefficients that are then processed. Specifically, after multi-scale analysis of the noisy signal, wavelet coefficients are obtained. A threshold estimation is then performed to determine if wavelet coefficients below the threshold are noise-generated and should be ignored to achieve noise reduction. Wavelet coefficients above the selected threshold are considered valid signals and are then processed using a threshold function for signal reconstruction to obtain de-noised urban meteorological data. Finally, the de-noised urban meteorological data is fed into an LSTM neural network to predict urban weather conditions.

[0039] In one embodiment, assigning cleaning tasks to sanitation vehicles based on garbage volume zones, predicted city weather conditions, sanitation vehicle status, and the location and driving trajectory of the sanitation vehicles includes: Monitor the overflow level of garbage bins in each garbage volume zone in real time; set priorities for garbage volume zones based on the overflow level of garbage bins and predicted urban weather conditions; and assign cleaning tasks to sanitation vehicles based on the set priorities, the status of sanitation vehicles, and their locations and driving trajectories.

[0040] Specifically, priorities are set for garbage volume zones based on the overflow level of the garbage bins and the city's weather conditions. Sanitation vehicles are then dispatched to the designated areas for cleaning operations based on these priorities. Smart sensors deployed on the garbage bins (such as ultrasonic sensors and LoRaWAN communication modules) monitor the overflow status of the bins in real time. These sensors detect the level of garbage in the bins. When the garbage level reaches a set value, garbage collection vehicles are prioritized to the garbage area to prevent accumulation. Weather conditions have a significant impact on sanitation operations, and operational plans can be dynamically adjusted based on weather conditions. For example, in inclement weather, sanitation vehicles can be prioritized for cleaning operations in key areas. Using GPS positioning and on-board sensors, sanitation vehicles' location, route, operating status, and fuel consumption are monitored in real time, and cleaning tasks are intelligently assigned based on the set priorities.

[0041] In one embodiment, obtaining real-time traffic conditions and dispatching sanitation vehicles to garbage volume zones to perform cleaning tasks based on the real-time traffic conditions includes: Obtain real-time traffic conditions and obtain congestion data of the current road according to the real-time traffic conditions; a dynamic environment model is established through the congestion-aware A* algorithm, the road is divided into multiple grid nodes, and the real-time congestion degree of each grid node is obtained according to the congestion data of the current road; a congestion-aware mechanism is introduced, the path cost value is calculated according to the real-time congestion degree; the optimal path is obtained according to the path cost value, and the sanitation vehicle is dispatched to go to the garbage quantity partition according to the optimal path to perform the cleaning task.

[0042] Specifically, in the congestion-aware A* algorithm, a dynamic environment model needs to be established first, the road or path is divided into multiple grid nodes, and real-time traffic conditions are obtained to obtain congestion data of the current road; the congestion data is input into each grid node to obtain the real-time congestion degree of each grid node. On the basis of the traditional A* algorithm, a congestion-aware mechanism is introduced, and the real-time congestion degree is introduced when calculating the path cost value to avoid congestion paths; according to the calculated path cost value, an optimal path is obtained, and the sanitation vehicle is dispatched to go to the garbage quantity partition according to the optimal path to perform the cleaning task. During the path execution process, the system will continuously monitor the traffic conditions and adjust the path according to the real-time feedback to cope with sudden congestion.

[0043] In this embodiment, the congestion-aware A* algorithm is an improved algorithm based on the traditional A* algorithm, which integrates real-time traffic congestion information for dynamic path optimization, can predict and avoid potential congestion nodes, and thus select a smoother and congestion-avoiding path, thereby improving the efficiency and safety of the overall path planning.

[0044] In one embodiment, the congestion-aware mechanism is introduced to calculate the path cost value according to the real-time congestion degree, which includes: Calculate the actual cost value of the grid node to the starting point and the estimated cost value of the grid node to the terminal point; introduce the congestion-aware mechanism to assign weight coefficients to the actual cost value, the estimated cost value and the real-time congestion degree for dynamically adjusting the priority of each cost; calculate the path cost value according to the actual cost value, the estimated cost value, the real-time congestion degree and the weight coefficients.

[0045] Specifically, in the path cost value a congestion cost factor is introduced to form a new path cost value: wherein, is the congestion degree calculated based on real-time traffic data (such as vehicle density, average speed), is the actual cost value from the starting point to the current node, is the estimated cost value from the current node to the target node, , , is the weight coefficient.

[0046] Add the starting point and the target point to the open list. The open list is used to store the nodes to be evaluated, and a closed list is created to store the nodes that have been explored. Add the starting point to the open list and set , , from the open list The node P with the smallest value, if P is the end point, backtrack to generate a path; otherwise, expand P's adjacent node N and determine whether the adjacent node N is in the closed list. If not, calculate: Where d is the geometric distance and C is the congestion cost; like <Existing , update N's parent node to P, and recalculate And update the list, if If the threshold is exceeded, it is considered a high-risk congestion node and can be skipped directly; repeat until a path is found or the open list is empty.

[0047] In one embodiment, obtaining real-time traffic conditions and dispatching sanitation vehicles to garbage volume zones to perform cleaning tasks based on the real-time traffic conditions further includes: If multiple sanitation vehicles need to be dispatched to work together in a large-scale operation scenario, the distributed vehicle group collaborative decision-making algorithm, combined with real-time traffic conditions, will be used to dispatch multiple sanitation vehicles to the garbage volume area to collaboratively perform cleaning tasks.

[0048] Specifically, if multiple sanitation vehicles need to be coordinated in a large-scale operation, they are dispatched to perform the cleaning task. Using a distributed vehicle swarm collaborative decision-making algorithm, sanitation vehicle queues on different lanes are treated as equivalent vehicle queues. The headway times of sanitation vehicles in the equivalent vehicle queues are calculated based on real-time traffic conditions. Based on these headway times, the speeds of the sanitation vehicles are controlled to dispatch multiple sanitation vehicles to coordinate cleaning tasks in the waste volume zones.

[0049] In one embodiment, Figure 3 As shown, a municipal sanitation cleaning task scheduling device based on big data analysis is provided, including: a garbage volume partitioning module 10, a weather condition prediction module 20, a vehicle data acquisition module 30, a cleaning task allocation module 40 and a sanitation vehicle scheduling module 50, wherein: The garbage volume partitioning module 10 is used to obtain historical garbage volume data of urban transportation stations and perform garbage volume partitioning based on the historical garbage volume data using the K-means++ clustering algorithm; The weather condition prediction module 20 is used to obtain urban meteorological data and predict urban weather conditions based on the urban meteorological data through a hybrid model of wavelet transform and LSTM neural network; The vehicle data acquisition module 30 is used to monitor the status of sanitation vehicles in real time through on-board sensors and track the location and driving trajectory of sanitation vehicles in real time through GPS positioning technology; A cleaning task assignment module 40 is used to assign cleaning tasks to sanitation vehicles based on the garbage volume zones, predicted urban weather conditions, sanitation vehicle status, and the location and driving trajectory of the sanitation vehicles; The sanitation vehicle dispatching module 50 is used to obtain real-time traffic conditions and, based on the real-time traffic conditions, dispatch sanitation vehicles to garbage volume zones to perform cleaning tasks.

[0050] In one embodiment, the garbage volume partitioning module 10 is also used to obtain traffic card swiping data, and based on the traffic card swiping data, extract the passenger flow time distribution characteristics of the traffic station through passenger flow time series analysis; obtain the interest point data around the traffic station, and based on the interest point data around the traffic station, obtain the spatial relationship between land use and passenger flow around the traffic station through a geographically weighted regression model; obtain the historical garbage volume data of urban traffic stations, and based on the historical garbage volume data, passenger flow time distribution characteristics and the spatial relationship between land use and passenger flow around the traffic station, perform garbage volume partitioning through the K-means++ clustering algorithm.

[0051] In one embodiment, the weather condition prediction module 20 is also used to obtain urban meteorological data and perform noise reduction on the urban meteorological data through a wavelet transform threshold noise reduction method; and predict urban weather conditions based on the noise-reduced urban meteorological data through an LSTM neural network.

[0052] In one embodiment, the cleaning task allocation module 40 is also used to monitor the overflow level of garbage bins in garbage volume partitions in real time; set priorities for garbage volume partitions based on the overflow level of garbage bins and predicted urban weather conditions; and allocate cleaning tasks to sanitation vehicles based on the set priorities, the status of sanitation vehicles, and the location and driving trajectory of sanitation vehicles.

[0053] In one embodiment, the sanitation vehicle dispatching module 50 is also used to obtain real-time traffic conditions and obtain the congestion data of the current road based on the real-time traffic conditions; establish a dynamic environment model through the congestion-aware A* algorithm, divide the road into multiple grid nodes, and obtain the real-time congestion level of each grid node based on the congestion data of the current road; introduce a congestion-aware mechanism to calculate the path cost value based on the real-time congestion level; obtain the optimal path based on the path cost value, and dispatch sanitation vehicles to the garbage volume partition according to the optimal path to perform cleaning tasks.

[0054] In one embodiment, the sanitation vehicle scheduling module 50 is also used to calculate the actual cost value from the grid node to the starting point and the estimated cost value from the grid node to the end point; introduce a congestion perception mechanism to assign weight coefficients to the actual cost value, estimated cost value and real-time congestion level, so as to dynamically adjust the priority of each cost; calculate the path cost value based on the actual cost value, estimated cost value, real-time congestion level and weight coefficient.

[0055] In one embodiment, the municipal sanitation cleaning task scheduling device based on big data analysis also includes an operation collaboration module, which is used to dispatch multiple sanitation vehicles to the garbage volume area to collaborate in performing cleaning tasks through a distributed vehicle group collaborative decision-making algorithm combined with real-time traffic conditions if multiple sanitation vehicles need to be dispatched to collaborate in a large-scale operation scenario.

[0056] Each module in the aforementioned municipal sanitation cleaning task scheduling device based on big data analysis can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a computer device's memory in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0057] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store infrared image data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a municipal sanitation cleaning task scheduling method based on big data analysis is implemented.

[0058] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0059] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A municipal sanitation cleaning task scheduling method based on big data analysis, characterized in that: include: Obtaining historical garbage volume data at urban transportation stations, and partitioning the garbage volume using a K-means++ clustering algorithm based on the historical garbage volume data; Obtain urban meteorological data, and predict urban weather conditions based on the urban meteorological data using a wavelet transform and LSTM neural network hybrid model; The status of sanitation vehicles can be monitored in real time through on-board sensors, and the location and driving trajectory of sanitation vehicles can be tracked in real time through GPS positioning technology; Assigning cleaning tasks to the sanitation vehicles based on the garbage volume zones, predicted urban weather conditions, the status of the sanitation vehicles, and the locations and driving trajectories of the sanitation vehicles; Acquire real-time traffic conditions, and in combination with the real-time traffic conditions, dispatch the sanitation vehicles to the garbage volume partition to perform the cleaning task.

2. A municipal sanitation cleaning task scheduling method based on big data analysis according to claim 1, characterized in that: The step of obtaining historical garbage volume data of a traffic station and partitioning the garbage volume using a K-means++ clustering algorithm based on the historical garbage volume data includes: Obtaining traffic card swiping data, and extracting the temporal distribution characteristics of passenger flow at the traffic station through passenger flow time series analysis based on the traffic card swiping data; Obtaining data on points of interest around transportation stations, and obtaining, based on the data on points of interest around transportation stations, a spatial relationship between land use and passenger flow around transportation stations using a geographically weighted regression model; Obtain historical garbage volume data of urban transportation stations, and perform garbage volume zoning using the K-means++ clustering algorithm based on the historical garbage volume data, the temporal distribution characteristics of passenger flow, and the spatial relationship between land use and passenger flow around the transportation stations.

3. The municipal sanitation cleaning task scheduling method based on big data analysis according to claim 1 is characterized in that: The obtaining of urban meteorological data and predicting urban weather conditions based on the urban meteorological data by using a wavelet transform and LSTM neural network hybrid model include: Acquire urban meteorological data, and perform noise reduction on the urban meteorological data using a wavelet transform threshold noise reduction method; Based on the denoised urban meteorological data, the urban weather conditions are predicted through the LSTM neural network.

4. The method for scheduling municipal sanitation cleaning tasks based on big data analysis according to claim 1 is characterized in that: The allocating cleaning tasks to the sanitation vehicles according to the garbage volume partition, the predicted urban weather conditions, the status of the sanitation vehicles, and the location and driving trajectory of the sanitation vehicles includes: Real-time monitoring of the overflow level of the garbage bins in the garbage volume partition; Prioritizing the waste volume zones based on the overflow level of the waste bins and predicted city weather conditions; Cleaning tasks are assigned to the sanitation vehicles according to the set priority, the status of the sanitation vehicles, and the location and driving trajectory of the sanitation vehicles.

5. The municipal sanitation cleaning task scheduling method based on big data analysis according to claim 1 is characterized in that: The acquiring of real-time traffic conditions and, in combination with the real-time traffic conditions, dispatching the sanitation vehicles to the garbage volume partitions to perform the cleaning tasks includes: Acquiring real-time traffic conditions and obtaining current road congestion data based on the real-time traffic conditions; A dynamic environment model is established using the congestion-aware A* algorithm to divide the road into multiple grid nodes, and the real-time congestion level of each grid node is obtained based on the current road congestion data; Introducing a congestion awareness mechanism to calculate a path cost based on the real-time congestion level; An optimal path is obtained according to the path cost value, and the sanitation vehicle is dispatched according to the optimal path to the garbage volume partition to perform the cleaning task.

6. A municipal sanitation cleaning task scheduling method based on big data analysis according to claim 5, characterized in that: The introduction of the congestion perception mechanism and calculation of the path cost according to the real-time congestion level include: Calculating an actual cost from the grid node to a starting point and an estimated cost from the grid node to an end point; Introducing a congestion awareness mechanism to assign weight coefficients to the actual cost, the estimated cost, and the real-time congestion level, so as to dynamically adjust the priority of each cost; A path cost value is calculated according to the actual cost value, the estimated cost value, the real-time congestion level, and the weight coefficient.

7. The method for scheduling municipal sanitation cleaning tasks based on big data analysis according to claim 1 is characterized in that: The acquiring of real-time traffic conditions and, in combination with the real-time traffic conditions, dispatching the sanitation vehicles to the garbage volume partition to perform the cleaning task further includes: If multiple sanitation vehicles need to be dispatched to work together in a large-scale operation scenario, the distributed vehicle group collaborative decision-making algorithm, combined with the real-time traffic conditions, will be used to dispatch the multiple sanitation vehicles to the garbage volume partition to collaboratively perform the cleaning task.

8. A municipal sanitation cleaning task scheduling device based on big data analysis, characterized in that: include: The garbage volume partitioning module is used to obtain historical garbage volume data of urban transportation stations and perform garbage volume partitioning based on the historical garbage volume data using the K-means++ clustering algorithm; A weather forecasting module is used to obtain urban meteorological data and, based on the urban meteorological data, predict urban weather conditions through a hybrid model of wavelet transform and LSTM neural network; The vehicle data acquisition module is used to monitor the status of sanitation vehicles in real time through on-board sensors, and to track the location and driving trajectory of sanitation vehicles in real time through GPS positioning technology; A cleaning task allocation module is used to allocate cleaning tasks to the sanitation vehicles based on the garbage volume partition, the predicted urban weather conditions, the status of the sanitation vehicles, and the location and driving trajectory of the sanitation vehicles; The sanitation vehicle scheduling module is used to obtain real-time traffic conditions and, based on the real-time traffic conditions, schedule the sanitation vehicles to go to the garbage volume partition to perform the cleaning task.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images