Route planning method and device, equipment, storage medium and computer program product

Abstract

The invention discloses a route planning method, device and equipment, a storage medium and a computer program product, which are used for solving the problems of low efficiency and poor accuracy of existing land law enforcement patrol. The method comprises the following steps: performing feature extraction on acquired monitoring data of a territorial land parcel to obtain first feature information corresponding to the territorial land parcel; performing risk prediction on the territorial land parcel according to the first feature information based on a pre-constructed time sequence diagram network model to obtain a risk prediction result corresponding to the territorial land parcel; and determining an inspection route for the territorial land parcel according to the risk prediction result.

Description

Technical Field

[0001] The present application relates to the technical field of automated route planning, and in particular to a route planning method, apparatus, device, storage medium, and computer program product. Background Art

[0002] Land and resources management is an important task to maintain the order of land and resources and promote the sustainable use of land and resources. Land law enforcement inspection is a key link in land and resources management. At present, land inspection is mainly achieved through manual inspection and intelligent assistance based on Geographic Information System (GIS).

[0003] Traditional manual inspections require a lot of manpower, have high operating costs, and are easily affected by subjective factors, resulting in inconsistent inspection results. In addition, due to limitations in manpower, time, and location, it is difficult to conduct comprehensive and systematic inspections and monitoring of large areas.

[0004] In contrast, the GIS-based intelligent decision-making support system can integrate various geospatial data, provide comprehensive land information and multi-dimensional analysis results through spatial analysis and model construction. The system realizes automated data processing and analysis, greatly improves inspection efficiency, and reduces human resource investment. At the same time, the system can also intuitively present inspection results in the form of maps, supporting real-time monitoring and early warning functions.

[0005] However, the existing GIS-based intelligent decision-making system mainly relies on low-altitude remote sensing technology (such as drones) and satellite remote sensing data. These data sources have problems such as limited coverage, low resolution, and being affected by weather conditions, which leads to untimely data updates and difficulty in timely detection of land violations, restricting the actual application effect of the system. Moreover, after the GIS system identifies the state-owned land plots that may be at risk, law enforcement personnel still need to conduct a lot of field inspections.

[0006] It can be seen that there is an urgent need for a land inspection route planning method that can not only improve the efficiency and accuracy of land law enforcement inspections, but also reduce the human resource investment in land inspections. Summary of the invention

[0007] The embodiment of the present application provides a route planning method to solve the problem of low efficiency and poor accuracy of existing land law enforcement patrols.

[0008] The embodiment of the present application also provides a route planning device to solve the problem of low efficiency and poor accuracy of existing land law enforcement patrols.

[0009] The embodiment of the present application also provides a route planning device to solve the problem of low efficiency and poor accuracy of existing land law enforcement patrols.

[0010] The embodiment of the present application also provides a computer-readable storage medium to solve the problem of low efficiency and poor accuracy of existing land law enforcement patrols.

[0011] The embodiment of the present application also provides a computer program product to solve the problem of low efficiency and poor accuracy of existing land law enforcement patrols.

[0012] The present application embodiment adopts the following technical solutions:

[0013] A route planning method comprises: extracting features from monitoring data of a national land parcel to obtain first feature information corresponding to the national land parcel; based on a pre-constructed time series network model, performing risk prediction on the national land parcel according to the first feature information to obtain a risk prediction result corresponding to the national land parcel; and determining an inspection route for the national land parcel according to the risk prediction result.

[0014] A route planning device comprises: a feature extraction unit, used to extract features from monitoring data of a national land block to obtain first feature information corresponding to the national land block; a risk analysis unit, used to perform risk prediction on the national land block based on a pre-constructed time series network model and the first feature information to obtain a risk prediction result corresponding to the national land block; and a path planning unit, used to determine an inspection route for the national land block according to the risk prediction result.

[0015] A route planning device, comprising:

[0016] A processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor performs the following operations: extracting features from the monitoring data of the acquired national land block to obtain first feature information corresponding to the national land block; based on a pre-constructed time series network model, performing risk prediction on the national land block according to the first feature information to obtain a risk prediction result corresponding to the national land block; and determining an inspection route for the national land block according to the risk prediction result.

[0017] A computer-readable storage medium stores one or more programs, which, when executed by an electronic device including multiple application programs, enable the electronic device to perform the following operations: extract features from monitoring data of a national land parcel to obtain first feature information corresponding to the national land parcel; perform risk prediction on the national land parcel based on the first feature information based on a pre-constructed time series network model to obtain a risk prediction result corresponding to the national land parcel; and determine an inspection route for the national land parcel based on the risk prediction result.

[0018] A computer program product includes a computer program, which, when executed by a processor, implements: extracting features from monitoring data of a national land parcel to obtain first feature information corresponding to the national land parcel; based on a pre-constructed time-series graph network model, performing risk prediction on the national land parcel according to the first feature information to obtain a risk prediction result corresponding to the national land parcel; and determining an inspection route for the national land parcel according to the risk prediction result.

[0019] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0020] Using the route planning method provided in the embodiment of the present application, for the state-owned land plot to be inspected, firstly, the monitoring data of the state-owned land plot can be feature extracted to obtain the first feature information corresponding to the state-owned land plot; based on the pre-constructed time-series graph network model, the risk of the state-owned land plot is predicted according to the first feature information to obtain the risk prediction result corresponding to the state-owned land plot, and the inspection route for the state-owned land plot is determined according to the risk prediction result. Using the method provided in the embodiment of the present application, on the one hand, through the time-series deep graph network modeling method, it is possible to effectively capture the complex spatiotemporal dependencies between plots and extract plot feature representations that are more rich in time information, and by integrating these plot features with historical inspection records, weather and other auxiliary information to predict the probability of illegal risks in the plot, it also greatly improves the prediction accuracy; on the other hand, planning the inspection route based on the probability of illegal risks in the plot can guide law enforcement personnel to carry out inspections in a targeted manner, greatly improving the inspection efficiency of state-owned land plots. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 A schematic diagram of the specific structure of a national land parcel inspection system provided in an embodiment of the present application;

[0023] Figure 2 A schematic diagram of a specific process of a route planning method provided in an embodiment of the present application;

[0024] Figure 3 A schematic diagram of the specific structure of a timing diagram network model provided in an embodiment of the present application;

[0025] Figure 4 A schematic diagram of the specific structure of a route planning device provided in an embodiment of the present application;

[0026] Figure 5 A schematic diagram of the specific structure of a route planning device constructed for an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0028] The executor of the route planning method provided in the embodiment of the present application may be, but is not limited to, at least one of a land inspection server, a land resources monitoring server, or a route planning server; in addition, the executor of the method may also be an application (Application, APP) running on these servers or the system itself.

[0029] For ease of description, the following takes the implementation of the method as an example in which the execution subject of the method is a national land inspection system deployed on a national land inspection server to introduce the implementation method of the method. It can be understood that the execution subject of the method as a national land inspection system is only an exemplary description and should not be understood as a limitation of the method.

[0030] In one embodiment, the structure of the land parcel inspection system in the country is as follows Figure 1 As shown, it mainly includes the following modules:

[0031] 1. Data collection module, which is used to conduct video surveillance of land plots using video surveillance equipment installed nationwide, and upload the collected monitoring data to the national land plot inspection system.

[0032] 2. The time series analysis module is used to extract features based on the monitoring data collected by the data acquisition module (such as the time, alarm type, duration, etc. of the video detection abnormal alarm) to obtain the first feature information corresponding to the national land block.

[0033] 3. The time series graph network model is used to build a model based on the national land plot information collected by the data acquisition module and the first feature information extracted by the time series analysis module to obtain a dynamic graph structure, and use the RWKV (Receptance Weighted Key Value) model to extract features from the dynamic graph structure to obtain plot features that record the spatiotemporal relationship of the plots, and the plot features are sent to the prediction module.

[0034] 4. The prediction module predicts the risk probability of illegal behavior in the plot within a certain period of time in the future based on the plot characteristics uploaded by the time-series graph network model, the properties of the plot itself, the historical inspection records corresponding to the plot, and the time-series violation information (including the season and weather on the day of the violation).

[0035] 5. Route planning module, based on the DQN path planning algorithm, automatically generates the optimal patrol route according to the risk probability, road network information and plot location information output by the prediction module, driven by the reward module. Finally, the visual decision interface will show the optimal patrol path to law enforcement officers, and the real-time information during the patrol process will be fed back to the DQN path planning algorithm, so as to dynamically adjust the patrol path, and law enforcement officers can obtain and choose to use the latest patrol path at any time.

[0036] Based on the above-mentioned national land inspection system, the specific implementation flow diagram of the route planning method provided by this application is as follows: Figure 2 As shown, it mainly includes the following steps:

[0037] Step 11, extracting features from the acquired monitoring data of the national land block to obtain first feature information corresponding to the national land block;

[0038] In an embodiment of the present application, the national land plot inspection system can utilize video surveillance equipment installed throughout the country to conduct video surveillance of land plots, and thereby obtain monitoring data of the national land plots.

[0039] The acquired monitoring data includes the following two types:

[0040] 1. Plot time series data, which may specifically include: abnormal alarms generated during the monitoring process, alarm time, alarm type, alarm duration and other data.

[0041] 2. Static data of the land parcel, which may include: longitude and latitude coordinates of the land parcel, land type, land use status, land use rights information, land use change data, etc.

[0042] It should be noted here that in order to improve the subsequent processing effect, before feature extraction, the national land inspection system can first clean the monitoring data, process missing values, outliers and duplicate data, and normalize or standardize the cleaned data to obtain structured data corresponding to the monitoring data.

[0043] Since the monitoring data includes two types: time series data of plots and static data of plots, the national land plot inspection system can extract features of the above two types of data through the following two methods, including:

[0044] Solution a, through the time series feature extraction module, extract the features of the time series data in the monitoring data to obtain the time features corresponding to the national land block;

[0045] Specifically, the national land inspection system can input structured data into the time series feature extraction module to extract representative time series features, such as alarm type, alarm frequency and duration, etc.

[0046] Solution b, obtaining the static data of the land parcel in the monitoring data, performing feature extraction on the static data of the land parcel, and obtaining the static features corresponding to the national land parcel;

[0047] In an embodiment of the present application, the static features extracted by the national land parcel inspection system may include: longitude and latitude coordinates of the parcel, land type, land use status, land use rights information, land use change data, and the last inspection time of the parcel, etc.

[0048] The national land block inspection system uses the static features and the time features obtained by the above method as the first feature information corresponding to the national land block.

[0049] Step 12, based on the pre-constructed time-series graph network model, and according to the first characteristic information obtained by executing step 11, risk prediction is performed on the state-owned land parcel to obtain a risk prediction result corresponding to the state-owned land parcel;

[0050] It should be noted here that in an embodiment of the present application, the timing graph network model can be built based on the RWKV (Read-Write Knowledge Vector) model, which is a new type of deep learning model that aims to combine the advantages of RNN and Transformer while avoiding the shortcomings of both. The architectural features of the RWKV model include time mixing and channel mixing blocks, in which four main elements are used: Receptance vector (R), weight vector (W), key vector (K), and value vector (V). These elements interact with each other through a specific calculation method to achieve the implementation of linear attention and avoid the limitations of traditional dot product token interactions.

[0051] In the embodiment of the present application, the overall structure of the timing diagram network model is as follows: Figure 3 In one embodiment, the time-series graph network model provided in the embodiment of the present application uses a time-mixing block and a channel-mixing block as time-series data feature extraction modules, the time-series graph network model has 6 network layers, 3 graph convolutional network layers, and an activation function of Leaky ReLU. The Huber loss function is used as the loss function for model training, and the Adam (Adaptive Moment Estimation) optimizer is used to update the weights of the model.

[0052] In an embodiment of the present application, the national land inspection system can construct a training data set based on the monitoring data obtained by executing step 11, and use the training set data to train the above-mentioned time series graph network model constructed based on RWKV, adjust the network weights through multiple rounds of iterations, monitor the training process, and use the validation set to evaluate the model performance to prevent overfitting, and further adjust the model parameters, such as learning rate, batch size, etc., to optimize the model performance, thereby obtaining the time series graph network model.

[0053] In an embodiment of the present application, based on the trained time-series graph network model, the national land inspection system can use the following method to predict the risk of the national land plot, including: constructing a graph structure based on the first feature information; determining the risk prediction result corresponding to the national land plot based on the time-series graph network model and the graph structure.

[0054] Among them, the graph structure can be obtained based on the fusion of static features and time features. Specifically, in the embodiment of the present application, the national land inspection system can vectorize the static features to obtain static feature vectors, and then obtain the graph structure by fusing the static feature vectors with the time features.

[0055] After constructing the graph structure, the national land inspection system can perform time-series encoding on the graph structure to obtain the second characteristic information corresponding to the national land blocks, and then perform risk prediction on the national land blocks based on the time-series graph network model and the second characteristic information, determine the risk prediction results corresponding to each national land block, and output the probability risk of illegal behavior in each block within a certain period of time in the future.

[0056] Step 13, based on the risk prediction results obtained by executing step 12, determine the inspection route for the state-owned land parcel.

[0057] In an embodiment of the present application, the national land inspection system can construct a directed graph based on all national land plots to be inspected, where each plot corresponds to a node, and there are directed edges between the nodes corresponding to adjacent plots. Then, a pre-trained path planning model is used to generate an inspection route based on the risk prediction results and the directed graph.

[0058] Specifically, the path planning model can be trained according to the DQN (Deep Q-Network) path planning algorithm. The path planning model can automatically find the optimal path from the starting point to the end point through training based on the current state (inspected plots and their adjacent uninspected plots) and environmental rewards (risk probabilities corresponding to each plot).

[0059] In one implementation, the path planning model may be trained and generated specifically according to the following steps, including:

[0060] Sub-step 1301, determining a reward function corresponding to the initial model according to the risk prediction result;

[0061] In an embodiment of the present application, the reward function of the initial model can be set according to the risk prediction results corresponding to each land plot, and appropriate weights can be set, so that the path planning model can learn to pay more attention to the inspection path of high-risk plots and avoid low-risk areas, thereby improving law enforcement efficiency.

[0062] In one implementation, the reward function of the path planning model may be set according to the following formula [1]:

[0063] R = a × R risk +(1-a)×R length [1]

[0064] Among them, R represents the total reward, R risk Represents the reward item related to the risk prediction result of the plot, R length represents the reward item related to the path length, a is a dynamic weight factor, which controls the relative importance of the two rewards. In one embodiment, the value of a can be 0.63.

[0065] Specifically, R risk It can be calculated according to the following formula [2]:

[0066] R risk =∑p i ×ε i [2]

[0067] Among them, p i represents the risk prediction result of the i-th uninspected plot, ε i represents the coefficient score of the most recent inspection interval of the plot. In the embodiment of the present application, εi It can be calculated according to the following formula [3]:

[0068]

[0069] Sub-step 1302, determining the state space and action space corresponding to the initial model according to the land parcel topology map;

[0070] Sub-step 1303, according to the reward function, the state space and the action space, the initial model is trained using a reinforcement learning algorithm to obtain a path planning model

[0071] The national land inspection system inputs the land topology map and risk prediction results into the path planning model, thereby determining the inspection route for the national land block, and displays the inspection route to law enforcement personnel through a visual decision-making interface.

[0072] In addition, it should be noted here that the path planning method provided in the embodiment of the present application does not generate a complete path from the starting point to the end point at one time, but adopts a rolling optimization strategy during the law enforcement process, that is, only the local path between the current position to the next optional plot is optimized each time, and this process is repeated as the law enforcement personnel advance.

[0073] Specifically, when law enforcement officers conduct actual inspections, they can use mobile terminals (such as mobile phones) to feed back the information collected during the inspection (such as the status of inspected plots of land, etc.) in real time to the national land plot inspection system, so that the national land plot inspection system can dynamically adjust the inspection path according to the path planning model.

[0074] Using the route planning method provided in the embodiment of the present application, for the state-owned land plot to be inspected, firstly, the monitoring data of the state-owned land plot can be feature extracted to obtain the first feature information corresponding to the state-owned land plot; based on the pre-constructed time-series graph network model, the risk of the state-owned land plot is predicted according to the first feature information to obtain the risk prediction result corresponding to the state-owned land plot, and the inspection route for the state-owned land plot is determined according to the risk prediction result. Using the method provided in the embodiment of the present application, on the one hand, through the time-series deep graph network modeling method, it is possible to effectively capture the complex spatiotemporal dependencies between plots and extract plot feature representations that are more rich in time information, and by integrating these plot features with historical inspection records, weather and other auxiliary information to predict the probability of illegal risks in the plot, it also greatly improves the prediction accuracy; on the other hand, planning the inspection route based on the probability of illegal risks in the plot can guide law enforcement personnel to carry out inspections in a targeted manner, greatly improving the inspection efficiency of state-owned land plots.

[0075] In one embodiment, the present application also provides a route planning device to solve the problem of low efficiency and poor accuracy of existing land law enforcement inspections. Figure 4 As shown, it includes: a feature extraction unit 41, a risk analysis unit 42 and a path planning unit 43.

[0076] The feature extraction unit 41 is used to extract features from the monitoring data of the acquired national land block to obtain first feature information corresponding to the national land block;

[0077] The risk analysis unit 42 is used to perform risk prediction on the national land block based on the pre-built time-series graph network model and the first characteristic information to obtain a risk prediction result corresponding to the national land block;

[0078] The path planning unit 43 is used to determine the inspection route for the national land block according to the risk prediction result.

[0079] In one embodiment, the feature extraction unit 41 is specifically used to: obtain time series data in the monitoring data, perform feature extraction on the time series data, and obtain time features corresponding to the national land plot; obtain land plot static data in the monitoring data, perform feature extraction on the land plot static data, and obtain static features corresponding to the national land plot; use the static features and the time features as the first feature information corresponding to the national land plot.

[0080] In one implementation, the risk analysis unit 42 is specifically used to: construct a graph structure according to the first feature information; and determine the risk prediction result corresponding to the national land parcel according to the time series graph network model and the graph structure.

[0081] In one implementation, the risk analysis unit 42 is specifically configured to: perform vectorization processing on the static feature to obtain a static feature vector; and fuse the static feature vector with the time feature to obtain the graph structure.

[0082] In one embodiment, the risk analysis unit 42 is specifically used to: perform time-series encoding on the graph structure according to the time-series graph network model to obtain second characteristic information corresponding to the national land plot; and determine the risk prediction result corresponding to the national land plot according to the time-series graph network model and the second characteristic information.

[0083] In one embodiment, the path planning unit 43 is specifically used to: generate a land parcel topology map corresponding to the national land parcel; and determine an inspection route for the national land parcel based on the land parcel topology map and the risk prediction result according to a pre-trained path planning model.

[0084] In one embodiment, it also includes a model training unit, which is specifically used to: determine the reward function corresponding to the initial model based on the risk prediction result; determine the state space and action space corresponding to the initial model based on the plot topology map; and train the initial model using a reinforcement learning algorithm based on the reward function, the state space and the action space to obtain the path planning model.

[0085] By using the route planning device provided in the embodiment of the present application, for the state-owned land plot to be inspected, the monitoring data of the state-owned land plot can first be feature extracted to obtain the first feature information corresponding to the state-owned land plot; based on the pre-constructed time-series graph network model, the risk of the state-owned land plot is predicted according to the first feature information to obtain the risk prediction result corresponding to the state-owned land plot, and the inspection route for the state-owned land plot is determined according to the risk prediction result. By using the method provided in the embodiment of the present application, on the one hand, the complex spatiotemporal dependency relationship between plots can be effectively captured through the time-series deep graph network modeling method, and the plot feature representation richer in time information can be extracted. By integrating these plot features with historical inspection records, weather and other auxiliary information, the probability of illegal risks in the plot can be predicted, which greatly improves the prediction accuracy; on the other hand, the inspection route is planned based on the probability of illegal risks in the plot, which can guide law enforcement personnel to carry out inspections in a targeted manner, greatly improving the inspection efficiency of state-owned land plots.

[0086] Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 5 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0087] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0088] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0089] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a route planning device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations: extracting features from the monitoring data of the acquired national land block to obtain the first feature information corresponding to the national land block; based on the pre-constructed time-series graph network model, according to the first feature information, predicting the risk of the national land block to obtain the risk prediction result corresponding to the national land block; and determining the inspection route for the national land block according to the risk prediction result.

[0090] The above application Figure 5 The method performed by the route planning electronic device disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0091] Of course, in addition to software implementation, the electronic device of the present application does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0092] The embodiment of the present application also provides a computer-readable storage medium, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by a portable electronic device including a plurality of application programs, enable the portable electronic device to execute Figure 1 The method of the embodiment shown is specifically used to perform the following operations:

[0093] Perform feature extraction on the monitoring data of the national land block to obtain first feature information corresponding to the national land block; perform risk prediction on the national land block based on the first feature information based on a pre-constructed time series network model to obtain a risk prediction result corresponding to the national land block; determine an inspection route for the national land block based on the risk prediction result.

[0094] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0096] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0098] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0099] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0100] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0101] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0102] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0103] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A route planning method, characterized in that: include: Extracting features from the acquired monitoring data of the national land block to obtain first feature information corresponding to the national land block; Based on the pre-constructed time-series graph network model, and according to the first characteristic information, risk prediction is performed on the national land block to obtain a risk prediction result corresponding to the national land block; Based on the risk prediction results, an inspection route for the national land parcel is determined.

2. The method according to claim 1, characterized in that The feature extraction of the acquired monitoring data of the national land block to obtain the first feature information corresponding to the national land block specifically includes: Acquire time series data from the monitoring data, perform feature extraction on the time series data, and obtain time features corresponding to the national land block; Obtaining static data of the land parcel in the monitoring data, performing feature extraction on the static data of the land parcel, and obtaining static features corresponding to the national land parcel; The static feature and the time feature are used as the first feature information corresponding to the national land block.

3. The method according to claim 2, characterized in that The pre-built time-series graph network model performs risk prediction on the national land parcel according to the first feature information to obtain a risk prediction result corresponding to the national land parcel, specifically including: Constructing a graph structure according to the first feature information; According to the time series graph network model and the graph structure, the risk prediction result corresponding to the national land parcel is determined.

4. The method according to claim 3, characterized in that The step of constructing a graph structure according to the first feature information specifically includes: Performing vectorization processing on the static features to obtain a static feature vector; The static feature vector is fused with the temporal feature to obtain the graph structure.

5. The method according to claim 3, characterized in that: Determining the risk prediction result corresponding to the national land parcel according to the time sequence graph network model and the graph structure specifically includes: Performing temporal coding on the graph structure according to the temporal graph network model to obtain second characteristic information corresponding to the national land block; The risk prediction result corresponding to the national land parcel is determined according to the time series graph network model and the second characteristic information.

6. The method according to claim 1, characterized in that Determining the inspection route for the national land block according to the risk prediction result specifically includes: Generate a land parcel topology map corresponding to the national land parcel; According to the pre-trained path planning model, based on the land parcel topology map and the risk prediction result, an inspection route for the national land parcel is determined.

7. The method according to claim 6, characterized in that Pre-training the path planning model specifically includes: Determining a reward function corresponding to the initial model according to the risk prediction result; Determine the state space and action space corresponding to the initial model according to the land parcel topology map; According to the reward function, the state space and the action space, the initial model is trained using a reinforcement learning algorithm to obtain the path planning model.

8. A route planning device, characterized in that: include: A feature extraction unit, used to extract features from the acquired monitoring data of the national land block to obtain first feature information corresponding to the national land block; A risk analysis unit, configured to perform risk prediction on the national land parcel based on a pre-built time-series graph network model and the first characteristic information, and obtain a risk prediction result corresponding to the national land parcel; A path planning unit is used to determine an inspection route for the national land block according to the risk prediction result.

9. A route planning device, comprising: processor; a memory arranged to store computer executable instructions which, when executed, cause the processor to: Extracting features from the acquired monitoring data of the national land block to obtain first feature information corresponding to the national land block; Based on the pre-constructed time-series graph network model, and according to the first characteristic information, risk prediction is performed on the national land block to obtain a risk prediction result corresponding to the national land block; Based on the risk prediction results, an inspection route for the national land parcel is determined.

10. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, enable the electronic device to execute the route planning method as described in any one of claims 1 to 7.

11. A computer program product, characterized in that It comprises a computer program, which, when executed by a processor, implements the route planning method as claimed in any one of claims 1 to 7.

Images