Intelligent forest fire prevention monitoring and scheduling system based on unmanned aerial vehicle

Abstract

The invention provides an intelligent forest fire prevention monitoring and scheduling system based on an unmanned aerial vehicle, and the system comprises a forest region division module which is used for dividing a to-be-monitored forest region into a plurality of monitoring regions; the monitoring task allocation module is used for allocating corresponding forest monitoring tasks to the unmanned aerial vehicles according to the divided monitoring areas; the monitoring task scheduling module is used for calling a task optimization model and scheduling the forest monitoring task, and the optimization target of the task optimization model is to minimize the completion time of the forest monitoring task; and the data processing and monitoring module is used for processing forest monitoring data acquired when the unmanned aerial vehicle executes the forest monitoring task in real time and feeding back the execution state of the forest monitoring task. According to the method, the defects of traditional forest fire monitoring methods such as manual patrol, watchtower observation and satellite remote sensing are overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of forest prevention and control, and particularly to an intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles (UAVs). Background Art

[0002] Fires are regarded as one of the most severe crises faced by human civilization. In recent years, whether in urban, rural areas or uninhabited regions (such as forests), high fire incidences have been experienced globally. When the causes of fires are complex or the spread directions of fire are difficult to predict, it becomes particularly difficult to control fires, especially in forests or residential areas. Therefore, local governments strive to seek the most effective fire-fighting means to prevent the occurrence of fires, control the fire and stop its spread.

[0003] Traditional forest fire monitoring methods mainly include manual patrol, observation from watchtowers and satellite remote sensing. In the early stage of fire prevention and control, these technologies played important roles. Manual patrol can quickly respond to the fire situation on-site and is suitable for dealing with small-scale problems. However, its coverage is limited, and it consumes a large amount of manpower and material resources, making it difficult to meet the monitoring requirements for large areas and complex terrains. Although observation from watchtowers has low costs and can operate all-weather, its field of vision is greatly affected by terrain complexity and weather conditions, and fires in remote or hidden areas are often difficult to detect in a timely manner.

[0004] The emergence of satellite remote sensing technology provides the ability to cover large areas for fire monitoring and can be used to analyze historical data to help evaluate the trends and patterns of fire occurrences. However, its resolution is low and the data update speed is slow, unable to meet the requirements of real-time monitoring and rapid response, especially insufficient in the identification and control in the initial stage of fires. Summary of the Invention

[0005] The present invention provides an intelligent forest fire prevention monitoring and dispatching system based on UAVs to overcome the defects existing in traditional forest fire monitoring methods such as manual patrol, observation from watchtowers and satellite remote sensing.

[0006] The present invention provides an intelligent forest fire prevention monitoring and dispatching system based on UAVs, including: a forest area division module, a monitoring task assignment module, a monitoring task scheduling module, and a data processing and monitoring module. It is characterized in that the forest area division module is used to divide the forest area to be monitored into multiple monitoring areas; The monitoring task assignment module is used to assign corresponding forest monitoring tasks to UAVs according to the divided monitoring areas; The monitoring task scheduling module is used to call a task optimization model to schedule the forest monitoring tasks, wherein the optimization objective of the task optimization model is to minimize the completion time of the forest monitoring tasks; The data processing and monitoring module is used to process the forest monitoring data collected when the drone performs the forest monitoring task in real time, and feedback the execution status of the forest monitoring task.

[0007] In some embodiments, the process of the forest area division module dividing the forest area to be monitored into multiple monitoring areas includes: Determine the overall width range of the forest area to be monitored; Divide the overall width range into multiple equal or different sub - area ranges; Use the forest area formed by the sub - area ranges as the monitoring area of the drone.

[0008] In some embodiments, the monitoring task scheduling module calls the task optimization model to schedule the forest monitoring task, including: When setting a single drone to perform the forest monitoring task in multiple monitoring areas, calculate the first completion time for the single drone to complete the forest monitoring task according to the task optimization model, and schedule the forest monitoring tasks in multiple monitoring areas when the first completion time reaches the minimum value; When setting multiple drones to perform the forest monitoring task for each monitoring area respectively, calculate the second completion time for the multiple drones to complete the forest monitoring task according to the task optimization model, and schedule the forest monitoring tasks of each drone in the monitoring area when the second completion time reaches the minimum value.

[0009] Further, the task optimization model is updated in real time by obtaining the scheduling data of the forest monitoring task, and the algorithms adopted by the task optimization model include any one of the following: Decreasing total flight time algorithm; Random iteration algorithm.

[0010] In some embodiments, the system further includes a drone task execution module, and the drone task execution module is used to convert the task assignment scheme generated by the task optimization model in the monitoring task scheduling module into specific action operations of the drone.

[0011] Further, the specific action operations of the drone include: path planning operation, instruction execution operation, and task feedback operation; The path planning operation includes planning the best flight path of the drone in the monitoring area according to the task assignment scheme; The instruction execution operation includes converting the instructions of the best flight path into flight control instructions of the drone, and the flight control instructions are used to control the drone to fly according to the best flight path; The task feedback operation includes real-time monitoring of the execution status of the UAV performing the forest monitoring task, and feedback on the completion progress of the forest monitoring task and flight data.

[0012] In some embodiments, the process of the data processing and monitoring module for real-time processing of the forest monitoring data collected when the UAV performs the forest monitoring task includes: Preprocessing the forest monitoring data collected when the UAV performs the forest monitoring task; Invoking a deep learning algorithm to identify fire targets from the preprocessed forest monitoring data; Constructing a dynamic environment model of the forest area based on the preprocessed forest monitoring data.

[0013] Further, when the deep learning algorithm identifies a fire target in the monitoring area, the following further processing is performed: Predicting the location of the fire target; Predicting the scale of the fire target; Predicting the future development trend of the fire target.

[0014] In some embodiments, the system further includes a security mechanism module, and the operations performed by the security mechanism module include: fault detection operation, emergency scheduling operation, and data redundancy operation.

[0015] Further, the fault detection operation includes: real-time monitoring of the running status of the system and performing fault detection on the system; The emergency scheduling operation includes: when it is detected that the UAV is in a fire emergency, setting the system to enter the emergency scheduling mode, and adjusting the forest monitoring task allocation plan generated by the task optimization model in the emergency scheduling mode; The data redundancy operation includes: when allocating forest monitoring tasks and feedback data, backing up the forest monitoring tasks and the data to be feedback according to a preset redundancy mechanism.

[0016] The intelligent forest fire prevention monitoring and scheduling system based on UAV provided by the present invention has the following beneficial effects: (1) Through the forest area division module and the monitoring task allocation module, the forest area is divided, and UAVs are set for the divided monitoring areas to perform corresponding forest monitoring tasks. In this way, the problems that the coverage of manual patrol is limited, and it consumes a large amount of manpower and material resources and is difficult to meet the monitoring requirements of large areas and complex terrains are solved.

[0017] (2) Through the monitoring task scheduling module, the task optimization model is called to schedule the UAV forest monitoring tasks, so as to minimize the completion time of the forest monitoring tasks, improve the monitoring efficiency, and timely detect the fire conditions in the forest area. It overcomes the defects that the field of vision of the observation tower is greatly affected by the terrain complexity and weather conditions, and it is often difficult to detect the fires in remote or hidden areas in a timely manner.

[0018] (3) Through the data processing and monitoring module, the forest monitoring data collected by the UAV during the forest monitoring tasks is processed in real time, and the execution status of the forest monitoring tasks is fed back. In this way, timely prediction and feedback can be achieved for forest fires, so as to timely give early warnings for forest fire prevention work and provide decision-making support. It overcomes the defects that satellite remote sensing technology cannot meet the requirements of real-time monitoring and rapid response, and there are deficiencies in the identification and control in the initial stage of the fire. Brief Description of the Drawings

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the description of the embodiments or the prior art will be briefly introduced one by one below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 It is one of the structural schematic diagrams of the intelligent forest fire prevention monitoring and scheduling system based on UAVs provided by the present invention.

[0021] Figure 2 It is the schematic diagram of the division of the forest monitoring area provided by the present invention.

[0022] Figure 3 It is the second structural schematic diagram of the intelligent forest fire prevention monitoring and scheduling system based on UAVs provided by the present invention.

[0023] Figure 4 It is the third structural schematic diagram of the intelligent forest fire prevention monitoring and scheduling system based on UAVs provided by the present invention. Detailed Embodiments

[0024] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0025] The intelligent forest fire prevention monitoring and dispatching system based on drones provided by the present invention can be deployed in a drone station for monitoring forests, and can communicate, transmit data, and synchronize in real time with the drones controlled by the drone station. The following describes the drone river channel inspection system based on deep learning of the present invention with reference to the accompanying drawings.

[0026] Figure 1 It is one of the structural schematic diagrams of the intelligent forest fire prevention monitoring and dispatching system based on drones provided by the present invention. As Figure 1 shown, the intelligent forest fire prevention monitoring and dispatching system based on drones (hereinafter referred to as the system) includes: a forest area division module, a monitoring task assignment module, a monitoring task scheduling module, and a data processing and monitoring module, which will be described one by one below.

[0027] The forest area division module is used to divide the forest area to be monitored into multiple monitoring areas. Here, due to the complex terrain and geographical environment information of the forest area, and the limited flight ability and battery life of the drones, it is necessary to divide the forest area to be monitored into multiple monitoring areas to set the drones to perform forest monitoring tasks on these monitoring areas respectively. The division result of the forest area will be sent to the monitoring task assignment module to assign corresponding forest monitoring tasks to the drones. In this way, the flight performance of the drones can be effectively utilized and the energy consumption can be saved, ensuring the comprehensive coverage of the forest monitoring tasks in each monitoring area.

[0028] The monitoring task assignment module is used to assign corresponding forest monitoring tasks to the drones according to the divided monitoring areas. Here, the monitoring task assignment module will reasonably assign forest monitoring tasks according to the result of the monitoring area division, combined with the flight ability and battery life of the drones, to ensure that the drones can efficiently complete the monitoring tasks of the entire forest area.

[0029] In the embodiment of the present invention, through the forest area division module and the monitoring task assignment module, the forest area is divided, and drones are set for the divided monitoring areas to perform corresponding forest monitoring tasks. In this way, the problem that the manual patrol coverage is limited, and it consumes a large amount of manpower and material resources and is difficult to meet the monitoring requirements of large areas and complex terrains is solved.

[0030] The monitoring task scheduling module is used to call the task optimization model to schedule forest monitoring tasks. The forest monitoring tasks are dynamically adjusted through the task optimization model. The forest monitoring tasks of the drones are scheduled and adjusted in real time, and the allocation plan of the forest monitoring tasks is sent to the monitoring task allocation module for implementation. The task optimization model therein can be trained according to the actual monitoring task scheduling data, so as to achieve performance optimization and improve the scheduling efficiency. Moreover, the optimization goal of the task optimization model is to minimize the completion time of forest monitoring tasks, that is, each scheduling of monitoring tasks is targeted at minimizing the completion time of forest monitoring tasks. This not only improves the utilization efficiency of drone resources but also ensures the rapid completion of forest monitoring tasks.

[0031] The data processing and monitoring module is used to process the forest monitoring data collected by the drones during the execution of forest monitoring tasks in real time and feedback the execution status of the forest monitoring tasks. A variety of sensors are carried on the drones, including high-definition cameras, infrared sensors, lidar, temperature sensors, humidity sensors, etc. This enables the drones to collect forest monitoring data of the monitoring area during forest monitoring tasks, including forest images, videos, temperature, and humidity information. Then the drones will feedback these forest monitoring data to the data processing and monitoring module, and the data processing and monitoring module calls data processing algorithms to process the forest monitoring data in real time, including the identification and prediction of forest fires. And it feedbacks the execution status of the forest monitoring tasks, for example, to the drone station. The execution status can be that the forest monitoring task has not started on the outbound journey, or is in the process of executing the forest monitoring task, or has completed the forest monitoring task on the return journey.

[0032] In the embodiment of the present invention, through the data processing and monitoring module, the forest monitoring data collected by the drones during the execution of forest monitoring tasks is processed in real time, and the execution status of the forest monitoring tasks is feedback. It can provide early warnings and decision-making support for forest fire prevention work according to the processing results and execution status of the forest monitoring data, overcoming the defects that the field of vision of the watchtower observation is greatly affected by the terrain complexity and weather conditions, and it is often difficult to detect fires in remote or hidden areas in a timely manner. It also overcomes the defects that satellite remote sensing technology cannot meet the requirements of real-time monitoring and rapid response and has deficiencies in the identification and control in the initial stage of fires.

[0033] In some embodiments, the system can be designed with a flexible system architecture to ensure seamless collaboration among various modules, real-time data transmission and communication, and achieve a smooth connection between the unmanned aerial vehicle (UAV) and the UAV station. In response to the real-time and precision requirements of tasks, in the monitoring task scheduling module, the performance of the task optimization model is optimized to improve the computing efficiency, shorten the task completion time, and enhance the scheduling response speed. For forest monitoring tasks, high-performance UAVs and sensor devices suitable for forest monitoring tasks are selected to ensure effective data collection and accurate execution during the data processing process. In the monitoring task scheduling module and the data processing and monitoring module, parallel computing and distributed processing technologies are used to run the task optimization model and data processing algorithms, accelerating the task scheduling and data processing speeds, and ensuring the real-time performance and efficiency of the system during large-scale forest monitoring.

[0034] In some embodiments, due to the diverse terrains and complex geographical environment information of the forest areas to be monitored, the forest area division module of the system divides the forest areas to be monitored into multiple monitoring areas, which need to be divided in combination with the specific actual situation of the forest areas. First, the forest to be monitored needs to be within the coverage range of the UAV station. Thus, the overall width range of the forest area to be monitored can be determined according to the coverage range of the UAV station. As Figure 2 shown, the overall width range of the monitored forest needs to meet the coverage range of the UAV station. Then, the overall width range is divided into multiple equal or different sub-region ranges.

[0035] Here, due to the diverse terrains and complex geographical environment information of the forest areas, corresponding division principles need to be formulated. If the terrain and environmental information distribution of the entire forest area is relatively balanced, the overall width range can be appropriately divided into multiple equal sub-region ranges. If the terrain and environmental information distribution of the entire forest area varies greatly, for example, the altitude is relatively high in some sub-region ranges and there are steep mountains, then the flight altitude of the UAV is relatively high, and the corresponding sub-region range should be divided smaller. In some sub-region ranges where the altitude is relatively low and the terrain is flat, the flight altitude of the UAV is relatively low, and the corresponding sub-region range should be divided larger. In this way, the overall width range will be divided into multiple different sub-region ranges.

[0036] And, as Figure 2 shown, a certain UAV departs from the UAV station, flies uniformly to monitoring area 1 to complete the forest monitoring task (the required time is t t1 ) and then needs to return to the UAV station (the round-trip time is d t1), and depending on the size of the sub-region range division, the flight time of the drone (including the time for performing forest monitoring tasks and the round-trip time) will also vary. Therefore, in actual scenarios, it is also necessary to reasonably adjust and divide the sub-region range according to the flight time of the drone.

[0037] After the sub-region division is completed, the forest area formed by the sub-region range is used as the monitoring area of the drone. For example Figure 2 As shown, the monitored forest area is divided into a total of j monitoring areas, namely Area 1, Area 2, Area 3, …, Area j, in order to allocate the corresponding drones to perform forest monitoring tasks.

[0038] In the embodiment of the present invention, each monitoring area is reasonably divided according to the terrain and geographical environment information of the forest area and the flight time of the drone, which can effectively utilize the flight performance of the drone and save energy consumption, ensuring the comprehensive coverage of the forest monitoring tasks in each monitoring area.

[0039] In some embodiments, the monitoring task scheduling module is used to call the task optimization model to schedule the forest monitoring tasks. The purpose of the scheduling is to ensure the rapid completion of the forest monitoring tasks of the drone, and the optimization goal of the task optimization model is to minimize the completion time of the forest monitoring tasks. Therefore, it is necessary to reasonably schedule the forest monitoring tasks to minimize this completion time.

[0040] As Figure 2 shown, after the forest area is divided by the forest area division module of the system, it is necessary to set the corresponding drones to perform forest monitoring tasks on the divided monitoring areas. In some embodiments, it can be to set only a single drone to sequentially perform forest monitoring tasks on multiple monitoring areas according to the task scheduling, or to set multiple drones to perform forest monitoring tasks on each monitoring area, that is, each drone respectively performs the forest monitoring task of one monitoring area. Therefore, it is necessary to schedule the monitoring tasks according to the settings of the drones.

[0041] Specifically, when a single drone is set to perform forest monitoring tasks in multiple monitoring areas, then the drone needs to complete the forest monitoring tasks of each monitoring area one by one. Calculate the first completion time for the single drone to complete the forest monitoring tasks according to the task optimization model. This first completion time includes the time for the drone to perform the forest monitoring tasks and the flight time to and from the drone station. Then, based on the minimum value of the first completion time, the task optimization model schedules the forest monitoring tasks of multiple monitoring areas. For example, first perform the forest monitoring tasks of the 2nd monitoring area, and then perform the forest monitoring tasks of the 5th monitoring area.

[0042] When multiple drones are set to perform forest monitoring tasks for each monitoring area respectively, each drone performs the corresponding forest monitoring task in the corresponding monitoring area. First, the second completion time for multiple drones to complete the forest monitoring tasks is calculated according to the task optimization model. This second completion time still includes the completion time of performing the forest monitoring task and the flight time for the round trip to and from the drone station. Moreover, since multiple drones depart from the drone station simultaneously, and the flight performances of the drones are different and the monitoring areas vary, the completion times of each drone are also different. Therefore, generally, the completion time after the last drone returns is used as the second completion time for this execution of the forest monitoring task.

[0043] Next, when the second completion time reaches the minimum value, the task optimization model schedules the forest monitoring tasks of each drone in the monitoring area. For example, it selects a suitable monitoring area for each drone to perform the corresponding forest monitoring task. Drones with low flight performance are assigned to monitoring areas with flat terrain or small sub - area ranges, while drones with high flight performance are assigned to monitoring areas with steep terrain or large sub - area ranges, so as to minimize the second completion time for the drones to perform the forest monitoring tasks.

[0044] In the embodiments of the present invention, according to the specific situation of the drones set and the differences in the flight performances of the drones, a reasonable schedule for the executed forest monitoring tasks can be carried out, which can effectively reduce the completion time of the forest monitoring tasks, thus ensuring the efficient and rapid completion of the forest monitoring tasks.

[0045] In some embodiments, to ensure a reasonable schedule for the forest monitoring tasks, it is necessary to optimize the task optimization model called in the monitoring task scheduling module in real - time, and the task optimization model is updated in real - time by obtaining the scheduling data of the forest monitoring tasks. For example, it can be trained in real - time according to the actual scheduling data of the forest monitoring tasks, so as to achieve performance optimization and improve the scheduling efficiency.

[0046] The scheduling algorithms adopted by the task optimization model include any one of the following: the decreasing total flight time algorithm; the random iterative algorithm (RID). The decreasing total flight time algorithm is a directed transfer function (DTF) algorithm, while the random iterative algorithm can transform the scheduling problem into a non - linear optimization problem, and uses local convergence and randomness to accelerate the global search. Here, the convergence of the completion time of the forest monitoring tasks is taken as the goal, and the forest monitoring tasks are randomly scheduled to search for all possible task scheduling combinations, and the task scheduling combination that can make the completion time converge is selected to determine the corresponding monitoring task scheduling plan.

[0047] In the embodiments of the present invention, by adopting various scheduling algorithms to model the task optimization model, it is possible to combine specific monitoring task scheduling data, calculate the optimal scheduling strategy, and minimize the completion time of forest monitoring tasks, so as to achieve the efficient scheduling of forest monitoring tasks by the system, and the scheduling algorithm is universal and not restricted by specific forest area conditions.

[0048] In some embodiments, as Figure 3 shown, the intelligent forest fire prevention monitoring and scheduling system based on drones further includes a drone task execution module, and the drone task execution module is used to convert the task allocation plan generated by the task optimization model in the monitoring task scheduling module into specific action operations of the drone.

[0049] Specifically, the drone task execution module is used to guide the drone to perform flight operations according to the forest monitoring tasks assigned to the drone, so as to perform forest monitoring tasks in the monitoring area. After the corresponding forest monitoring tasks are assigned by the monitoring task allocation module and sent to the drone task execution module, the drone task execution module guides the drone to perform flight operations in the corresponding monitoring area according to the forest monitoring tasks assigned to the drone, so as to perform forest monitoring tasks, and collect forest monitoring data through the sensor equipment carried during the execution of the forest monitoring tasks and give feedback. Thus, the accuracy and safety of the execution of forest monitoring tasks are ensured.

[0050] Furthermore, the specific action operations of the drone include: path planning operation, instruction execution operation, and task feedback operation. Specifically, the path planning operation includes planning the best flight path of the drone in the monitoring area according to the task allocation plan. In this way, it is possible to avoid the drone from repeatedly monitoring or missing the monitoring area, and optimize the flight time and energy consumption.

[0051] The instruction execution operation includes converting the instructions of the best flight path into flight control instructions of the drone, and the flight control instructions are used to control the drone to fly according to the best flight path. In this way, it is possible to accurately control the flight speed and flight direction of the drone according to the flight control instructions, and ensure the stability and accuracy of the monitoring process.

[0052] The task feedback operation includes real-time monitoring of the execution status of the drone performing forest monitoring tasks, and feedback of the completion progress of the forest monitoring tasks and flight data, which is used to support subsequent timely optimization adjustment and scheduling of the monitoring tasks. The execution status can be that the forest monitoring task has not started on the outbound journey, or is in the process of executing the forest monitoring task, or is in the return journey and has completed the forest monitoring task. The completion progress of the forest monitoring task can be, for example, the proportion of the monitored area in the monitoring area that has been completed, and the flight data includes the flight speed and flight direction.

[0053] In some embodiments, the data processing and monitoring module is used to process in real time the forest monitoring data collected when the drone performs the forest monitoring task, and the real-time processing process generally calls data processing algorithms to implement.

[0054] Specifically, first is the data preprocessing process, which is used to preprocess the forest monitoring data collected when the drone performs the forest monitoring task. The forest monitoring data collected when the drone performs the forest monitoring task is subjected to data cleaning to obtain target data suitable for analysis by data processing algorithms. Data cleaning includes denoising and normalization processing, which can improve data quality.

[0055] After the data preprocessing process, the data analysis process can be executed to identify the fire situation in the forest area. Here, the target data is analyzed through data processing algorithms. Among them, the data processing algorithms are generally deep learning algorithms. The deep learning algorithm is called to perform fire target recognition on the preprocessed forest monitoring data. Since the forest monitoring data collected by the drone is generally image data or video data, the deep learning algorithm can be called for image detection for these data to identify possible fire situations in the forest area, providing decision-making support for the environmental control and treatment of forest fires.

[0056] Specifically, the target data involves images or videos and can be used for target detection of the specific environmental conditions of the forest. For image data, the deep learning algorithm is directly called for target detection. For videos, some key frames can be extracted for target detection. And target detection requires data processing for forest images or videos, and the deep learning algorithms include at least one of the following: image processing algorithms; target detection algorithms. Specifically, it can be a convolutional neural network model, a target detection model or other models for target detection of images.

[0057] After the data preprocessing process, the dynamic modeling process can also be executed. According to the preprocessed forest monitoring data, a dynamic environment model of the forest area is constructed. The real forest area environmental scene is mapped into a dynamic model in a virtual network, such as a twin model, etc., and the dynamic model is fed back to the drone station to achieve all-round monitoring of the forest area, providing a basis for scheduling optimization and fire decision-making.

[0058] In some embodiments, after invoking a deep learning algorithm to identify fire targets from the preprocessed forest monitoring data, when the deep learning algorithm identifies a fire target in the monitored area, some further processing procedures can be executed. One is to predict the location of the fire target, so as to locate the fire area in a timely and accurate manner, such as which monitored area the fire occurs in, and determine the position coordinates or reference objects. The second is to predict the scale of the fire target, such as predicting the area affected by the fire or the spreading range, so as to timely and accurately predict the specific situation of the fire development. The third is to predict the future development trend of the fire target, such as predicting whether the fire will continue to expand or spread continuously, so that the future development situation of the fire can be accurately grasped, in order to take fire-fighting measures in a timely manner.

[0059] After performing these processes, the data such as the location prediction result, scale, and future development trend of the fire are also timely fed back to the drone station through the data processing and feedback module, for use as data support for fire-fighting decisions.

[0060] In the embodiments of the present invention, when a fire target is identified in the monitored area, the fast data processing ability is also utilized to further locate the fire, detect the fire scale, and predict the fire development trend, thereby achieving an all-round control of the forest fire situation, improving the monitoring accuracy and response speed. Finally, by feeding back the fire situation results, high-precision fire monitoring information is generated, providing data support for relevant departments to formulate scientific fire response strategies, and significantly improving the real-time performance and efficiency of forest fire management.

[0061] In some embodiments, as Figure 4 shown, the intelligent forest fire prevention monitoring and dispatching system based on drones further includes a security mechanism module. The security mechanism module is connected to other modules in the system and can conduct data communication, issue early warnings about security issues during the system operation, and timely make adjustments to forest monitoring tasks. The operations performed by the security mechanism module include: fault detection operations, emergency dispatching operations, and data redundancy operations. These operations are all emergency measures taken for the system when unexpected situations or failures occur in the system or the drone, affecting the execution of forest monitoring tasks, so as to ensure the safety and reliability of the system operation.

[0062] In some embodiments, the fault detection operation includes: real-time monitoring of the operating state of the system and performing fault detection on the system. When a fault is detected, automatically adjust the allocation scheme of the current forest monitoring tasks and re-allocate the forest monitoring tasks. By real-time monitoring the operating state of the system when each module of the system is running, quickly detect and handle possible faults. And through the self-recovery function, when a fault is detected, automatically adjust the current forest monitoring task allocation scheme in the monitoring task scheduling module and re-allocate the forest monitoring tasks. In this way, it can ensure the normal execution of the forest monitoring tasks and not be affected by system faults.

[0063] In some embodiments, the emergency scheduling operation includes: when it is detected that the drone is in a sudden fire situation, set the system to enter the emergency scheduling mode and adjust the forest monitoring task allocation scheme generated by the task optimization model in the emergency scheduling mode. Through the real-time monitoring of the forest area, when a sudden situation is detected, such as a sudden forest fire, heavy rain, or a drone failure, it will affect the forest monitoring tasks. At this time, turn on the emergency scheduling mode of the system and adjust the forest monitoring tasks allocated by the monitoring task allocation module to cope with the sudden situation. For example, change the monitored area divided by the forest area division module, change the path planning in the drone task execution module, etc. In this way, it can reduce or even eliminate the impact of sudden situations on the forest monitoring tasks and avoid task delays.

[0064] In some embodiments, the data redundancy operation includes: when allocating and feeding back data for forest monitoring tasks, according to a preset redundancy mechanism, back up the forest monitoring tasks and the data to be fed back. Because there is data communication among the drone, the system (each module), and the drone station. For example, the forest monitoring tasks allocated by the system to the drone, the task execution status and the collected forest monitoring data fed back by the drone to the system, and the processing results of the forest monitoring data fed back by the system to the drone station. These processes all involve the real-time transmission of data. When the system fails or a sudden situation occurs during the drone inspection, it may affect the real-time transmission of data, resulting in intermittent data transmission or data loss.

[0065] Based on this, in the embodiments of the present invention, a redundancy mechanism is added during the data transmission process through the security mechanism module. When allocating forest monitoring tasks and feeding back data, according to the preset redundancy mechanism, back up the forest monitoring tasks and the data to be fed back. Thus, when the data is intermittently transmitted or lost, use the backed-up data to continue communication, ensure the real-time transmission of data, and ensure the integrity and security of the data when the system fails or a sudden situation occurs during the drone monitoring.

[0066] For example, in the data processing and feedback module, due to a system failure, it is impossible to obtain the forest monitoring data collected by the drone, and it is also impossible to feedback the task execution status to the drone station. At this time, according to the preset redundancy mechanism, when the forest monitoring data is transmitted and the task execution status is feedback, the forest monitoring data and the task execution status data are immediately backed up. When the transmission and feedback process is interrupted due to a failure, it is still possible to use the backed-up forest monitoring data and task execution status data to continue the transmission and feedback, ensuring the integrity and security of the data.

[0067] In some embodiments, since the functions of each module in the intelligent forest fire prevention monitoring and scheduling system based on drones, as well as the deep learning algorithms and task optimization models called, are all compatible with other drone systems, the parameters can be adjusted according to different scenarios and task scales, supporting monitoring tasks in a wider range and complex environments, and having strong versatility. In addition, the system not only supports integration into existing forest monitoring drone systems, but can also be extended to other areas where monitoring tasks can be performed by drones, such as lakes, wetlands, coastal areas, etc. According to different monitoring scenarios and task requirements, the algorithms and models in the system are optimized, and through the integration with existing monitoring systems and drone stations, the performance and application value of the overall system are enhanced, supporting the optimization of multi-field monitoring tasks.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent forest fire prevention monitoring and dispatching system based on drones, comprising: The forest area division module, the monitoring task allocation module, the monitoring task scheduling module, and the data processing and monitoring module are characterized in that the forest area division module is used to divide the forest area to be monitored into multiple monitoring areas; The monitoring task allocation module is used to allocate corresponding forest monitoring tasks to the UAV according to the divided monitoring areas; The monitoring task scheduling module is used to call the task optimization model to schedule the forest monitoring task, wherein the optimization goal of the task optimization model is to minimize the completion time of the forest monitoring task; The data processing and monitoring module is used to process the forest monitoring data collected by the UAV when performing the forest monitoring task in real time, and to provide feedback on the execution status of the forest monitoring task.

2. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 1 is characterized in that: The process of the forest area division module dividing the forest area to be monitored into multiple monitoring areas includes: Determine the overall width of the forest area to be monitored; Dividing the overall width range into a plurality of equal or different sub-region ranges; The forest area formed by the sub-area range is used as the monitoring area of ​​the drone.

3. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 1 is characterized in that: The monitoring task scheduling module calls the task optimization model to schedule the forest monitoring task, including: When a single UAV is set to perform forest monitoring tasks in multiple monitoring areas, the first completion time of the single UAV to complete the forest monitoring task is calculated according to the task optimization model, and when the first completion time reaches the minimum value, the forest monitoring tasks of the multiple monitoring areas are scheduled; When multiple drones are set up to perform forest monitoring tasks for each monitoring area respectively, the second completion time for multiple drones to complete the forest monitoring tasks is calculated according to the task optimization model, and when the second completion time reaches the minimum value, the forest monitoring task of each drone in the monitoring area is scheduled.

4. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 3 is characterized in that: The task optimization model is updated in real time by acquiring the scheduling data of the forest monitoring task. The algorithm used by the task optimization model includes any one of the following: Decreasing total flight time algorithm; random iterative algorithm.

5. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 1 is characterized in that: The system also includes a UAV task execution module, which is used to convert the task allocation plan generated by the task optimization model in the monitoring task scheduling module into a specific action operation of the UAV.

6. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 5 is characterized in that: The specific action operations of the UAV include: path planning operation, command execution operation and task feedback operation; The path planning operation includes planning the optimal flight path of the UAV in the monitoring area according to the task allocation plan; The instruction execution operation includes converting the instruction of the optimal flight path into a flight control instruction of the UAV, wherein the flight control instruction is used to control the UAV to fly according to the optimal flight path; The task feedback operation includes real-time monitoring of the execution status of the UAV in performing the forest monitoring task, and feedback of the completion progress and flight data of the forest monitoring task.

7. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 1 is characterized in that: The data processing and monitoring module processes the forest monitoring data collected by the UAV when performing the forest monitoring task in real time, including: Preprocess the forest monitoring data collected by the UAV when performing forest monitoring tasks; Use deep learning algorithms to identify fire targets from preprocessed forest monitoring data; Based on the preprocessed forest monitoring data, a dynamic environmental model of the forest area is constructed.

8. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 7 is characterized in that: When the deep learning algorithm identifies that there is a fire target in the monitoring area, the following processing is further performed: Predict the location of fire targets; Predict the size of fire targets; Predict future development trends of fire targets.

9. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 1 is characterized in that: The system further comprises a safety mechanism module, and the operations performed by the safety mechanism module include: fault detection operation, emergency dispatch operation and data redundancy operation.

10. The intelligent forest fire prevention monitoring and dispatching system based on unmanned aerial vehicles according to claim 9 is characterized in that: The fault detection operation includes: real-time monitoring of the operating status of the system and performing fault detection on the system; The emergency dispatch operation includes: when the UAV is detected to be in a fire emergency situation, setting the system to enter an emergency dispatch mode, and adjusting the forest monitoring task allocation plan generated by the task optimization model in the emergency dispatch mode; The data redundancy operation includes: when allocating forest monitoring tasks and feedback data, backing up the forest monitoring tasks and feedback data according to a preset redundancy mechanism.

Images