Intelligent fire extinguishing strategy control method for autonomous flight fire extinguisher

Abstract

The invention provides an intelligent fire extinguishing strategy control method for an autonomous flight fire extinguisher, and belongs to the technical field of intelligent navigation fire extinguishment. The method comprises the following steps: firstly, acquiring fire scene data through a multi-source sensor, and constructing a fire behavior positive and negative development matrix; then, route planning is carried out based on a pre-trained deep learning model, the model adopts a special convolution kernel structure to extract fire scene features, and a decision is optimized through a mathematical calculation module; and the system updates a fire matrix in real time, dynamically adjusts a fire extinguishing strategy, and finally completes a precise fire extinguishing task through a time sequence aligned control instruction. According to the method, efficient control over the autonomous flight fire extinguisher in the complex environment is achieved through data driving and intelligent decision making, and the technical problem that in the prior art, an autonomous flight fire extinguisher is difficult to achieve precise fire extinguishing path planning and fire extinguishing strategy control in the complex fire environment is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent navigation fire extinguishing, and in particular, relates to an intelligent fire extinguishing strategy control method for an autonomous flying fire extinguisher. Background Art

[0002] With the acceleration of urbanization and industrial development, the frequency and severity of fire accidents are increasing. Traditional fire-fighting methods mainly rely on manual operation by firefighters, which not only endangers the lives of firefighters, but also makes it difficult to control the fire in a timely and effective manner. In recent years, autonomous flying fire extinguishers, as a new type of fire-fighting equipment, have played an increasingly important role in fire emergency response due to their advantages such as strong mobility and rapid response. Traditional autonomous flying fire extinguishers mainly use preset routes or manual remote control to carry out fire-fighting operations, obtain fire scene information by carrying various sensors, and make fire-fighting decisions according to pre-set rules.

[0003] However, current autonomous flying fire extinguishers have many limitations in practical applications. First, existing technologies generally use simple temperature thresholds or image recognition methods to determine the fire extinguishing location, which cannot accurately grasp the development trend of the fire, resulting in low fire extinguishing efficiency. Secondly, traditional route planning methods are mainly based on path optimization algorithms under static environments, which fail to fully consider the dynamic changes of the fire scene and are difficult to adapt to complex and changeable fire environments. Thirdly, existing fire extinguishing strategy control methods often use fixed fire extinguishing agent delivery modes, lack targeted responses to different fire characteristics, and result in a waste of fire extinguishing resources.

[0004] In complex fire scenarios, the spread of fire is highly uncertain and dynamic, and is affected by multiple factors such as wind direction, temperature, and distribution of combustibles. Existing technologies make it difficult to effectively conduct comprehensive analysis and decision-making optimization of factors such as multi-source sensor data, fire development predictions, and fire-fighting resource constraints, and are unable to achieve accurate fire-fighting path planning and fire-fighting strategy control. This results in autonomous flying fire extinguishers facing problems such as inaccurate positioning, poor paths, and improper control in actual applications, which seriously affects the fire-fighting effect. In other words, there is a technical problem in the existing technology that autonomous flying fire extinguishers are difficult to achieve accurate fire-fighting path planning and fire-fighting strategy control in complex fire environments. Summary of the invention

[0005] In view of this, the present invention provides an intelligent fire extinguishing strategy control method for an autonomous flying fire extinguisher, which can solve the technical problem in the prior art that it is difficult for an autonomous flying fire extinguisher to achieve accurate fire extinguishing path planning and fire extinguishing strategy control in a complex fire environment.

[0006] The present invention is implemented as follows: The present invention provides an intelligent fire extinguishing strategy control method for an autonomous flying fire extinguisher, comprising the following steps: obtaining sensor data of the autonomous flying fire extinguisher, constructing a fire positive development matrix according to the sensor data, calculating a fire negative development matrix according to the fire positive development matrix, establishing a fire extinguishing strategy equation group based on the fire positive development matrix and the fire negative development matrix, using the fire extinguishing strategy equation group to calculate a fire extinguishing measure vector, establishing a fire scene space division grid according to the fire positive development matrix and assigning a fire weight value, using a pre-trained fire extinguishing path planning model to perform route planning, calculating a flight trajectory curve according to the route planning result and generating a flight control instruction sequence, generating a fire extinguishing control instruction sequence according to the fire extinguishing measure vector, performing time alignment between the flight control instruction sequence and the fire extinguishing control instruction sequence to generate an execution instruction queue, controlling the autonomous flying fire extinguisher to perform a fire extinguishing task, updating the fire positive development matrix and the fire negative development matrix in real time through the sensor data, judging whether to re-plan the route according to the updated matrix, and returning to the platform after completing the fire extinguishing task.

[0007] Among them, the sensor data includes thermal imaging sensor data, visible light sensor data, three-band flame sensor data, temperature and humidity sensor data, smoke sensor data, and carbon monoxide sensor data; the thermal imaging sensor data is collected at a sampling frequency of 60 Hz, the visible light sensor data is collected at a frame rate of 30 frames per second, and the three-band flame sensor data is collected at a sampling frequency of 100 Hz.

[0008] Among them, the positive fire development matrix represents the fire spread speed, fire expansion direction, and fire intensity; the negative fire development matrix represents the fire decay speed, fire contraction direction, and fire reduction degree after the fire extinguishing agent is released.

[0009] Among them, the pre-trained fire-fighting path planning model adopts a convolutional neural network structure with an embedded mathematical calculation module. The mathematical calculation module contains a fire spread prediction equation, a resource consumption evaluation equation, and a route cost calculation equation, which is used to integrate the fire development situation and the fire-fighting resource constraints to perform route optimization calculations.

[0010] Among them, the input of the fire spread prediction equation includes fire scene temperature distribution data, environmental wind speed data, combustible material distribution data, and terrain height data; the input of the resource consumption assessment equation includes fire scene temperature data, fire scene area data, fire extinguishing agent storage data, fire extinguisher remaining power data, and target fire point distance data; the input of the route cost calculation equation includes route length data, fire scene danger data, fire extinguishing efficiency prediction data, energy consumption data, passage difficulty data, and task completion time data.

[0011] Among them, the basic size of the fire scene space division grid is 5 meters × 5 meters, and a grid size of 1 meter × 1 meter is used in areas where the fire is strong or changes greatly; the calculation of the fire weight value adopts the hierarchical analysis method, the weight coefficient of the temperature value is 0.3, the weight coefficient of the fire intensity is 0.25, the weight coefficient of the spread speed is 0.2, the weight coefficient of the terrain characteristics is 0.15, and the weight coefficient of the surrounding important targets is 0.1.

[0012] Among them, the flight trajectory curve is calculated using the Bezier curve interpolation algorithm, the maximum speed of the autonomous flying fire extinguisher is set to 15 meters per second, the maximum acceleration is set to 2 meters per square second, and the maximum turning angle is set to 45 degrees; the flight control instructions include three-dimensional position coordinates, flight speed, flight attitude and hovering state.

[0013] Among them, the fire extinguishing control instruction sequence includes the working state of the high-voltage generator, the trigger timing of the electromagnetic throwing controller and the flow control of the fine water mist pump; the control of the high-voltage generator adopts pulse width modulation, the control of the electromagnetic throwing controller adopts timing triggering, and the control of the fine water mist pump adopts variable frequency speed regulation.

[0014] Among them, when any of the conditions that the fire intensity changes by more than 30%, the spread direction changes by more than 45 degrees, or the fire range changes by more than 20% is met, the re-planning program is started; the re-planning program uses the Monte Carlo simulation method to simulate the task execution process 100 times, and the route is re-planned when the task completion rate is lower than 85%.

[0015] Among them, when the autonomous flying fire extinguisher returns to the platform, when the distance from the platform is greater than 100 meters, satellite navigation is used, and the navigation accuracy is better than 5 meters; when the distance is between 30 and 100 meters, laser navigation is used, and the navigation accuracy is better than 0.5 meters; when the distance is less than 30 meters, Bluetooth beacon navigation is used, and the navigation accuracy is better than 0.1 meters.

[0016] Compared with the prior art, the present invention provides an intelligent fire extinguishing strategy control method for an autonomous flying fire extinguisher. The present invention proposes an intelligent fire extinguishing strategy control method based on deep learning and multi-source information fusion, which realizes the accurate description and prediction of the dynamic characteristics of the fire by constructing a positive development matrix of the fire and a negative development matrix of the fire. The method uses a pre-trained fire extinguishing path planning model, combined with real-time sensor data, to dynamically generate the optimal route plan and fire extinguishing delivery strategy, which significantly improves the operating efficiency and accuracy of the autonomous flying fire extinguisher.

[0017] In terms of technical implementation, the present invention adopts an innovative convolutional neural network structure, and realizes multi-dimensional extraction of fire scene characteristics through three types of special convolution kernels: edge detection, direction perception, and intensity assessment. At the same time, a mathematical calculation module is introduced to integrate and optimize the fire development trend and fire-fighting resource constraints, ensuring the scientificity and feasibility of route planning. The system effectively overcomes the problem of insufficient adaptability of traditional methods in complex environments by updating the fire matrix in real time and dynamically adjusting the fire-fighting strategy.

[0018] The present invention successfully solves the technical problem in the prior art that it is difficult for autonomous flying fire extinguishers to achieve accurate fire extinguishing path planning and fire extinguishing strategy control in complex fire environments. By building a complete perception-decision-execution link, accurate prediction of fire development, intelligent planning of fire extinguishing paths and optimal allocation of fire extinguishing resources are achieved. The system has adaptive learning capabilities and can continuously optimize decision-making models based on historical data, improving its ability to respond to different fire scenarios. This data-driven intelligent control method provides a new technical path for improving the operational efficiency of autonomous flying fire extinguishers. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of the method of the present invention.

[0020] Figure 2 Graph showing the changing trend of fire scene temperature and carbon monoxide concentration over time in the embodiment.

[0021] Figure 3 Graph showing utilization rates and coverage areas of three types of fire extinguishing resources in the embodiment. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0023] like Figure 1 FIG. 1 is a flow chart of an intelligent fire extinguishing strategy control method for an autonomous flying fire extinguisher provided by the present invention. The method comprises the following steps:

[0024] S01. Acquire sensor data of the autonomous flying fire extinguisher, where the sensor data includes thermal imaging sensor data, visible light sensor data, three-band flame sensor data, temperature and humidity sensor data, smoke sensor data, and carbon monoxide sensor data.

[0025] S02. Construct a fire positive development matrix based on the sensor data, wherein the fire positive development matrix represents the fire spread speed, fire expansion direction, and fire intensity.

[0026] S03. Calculate a negative fire development matrix based on the positive fire development matrix, wherein the negative fire development matrix represents the fire decay speed, fire shrinkage direction, and fire reduction degree after the fire extinguishing agent is released.

[0027] S04. Establish a fire extinguishing strategy equation group based on the fire positive development matrix and the fire negative development matrix. The fire extinguishing strategy equation group is used to calculate the single-time fire extinguishing duration of a single machine and the multi-time fire extinguishing duration of multiple machines.

[0028] S05. Calculate the fire extinguishing delivery measure vectors of the autonomous flying fire extinguisher at different waypoints using the fire extinguishing strategy equation group. The fire extinguishing delivery measure vectors include the number of fire extinguishing bombs delivered, the number of fire extinguishing bottles delivered, and the duration of water mist spraying.

[0029] S06. Establishing a fire scene space division grid according to the fire forward development matrix, and assigning a fire intensity weight value to each grid node.

[0030] S07. Use the pre-trained fire-fighting path planning model to perform route planning on the fire scene space divided into grids to obtain a route planning result, wherein the route planning result includes coordinates of multiple waypoints, a sequence of waypoint arrivals, and a hovering time of the waypoints.

[0031] S08. Calculate the flight trajectory curve of the autonomous flying fire extinguisher according to the route planning result, and decompose the flight trajectory curve into a flight control instruction sequence.

[0032] S09. Generate a fire extinguishing control instruction sequence according to the fire extinguishing measure vector, wherein the fire extinguishing control instruction sequence is used to control a high voltage generator, an electromagnetic throwing controller, and a fine water mist pump.

[0033] S10: aligning the flight control instruction sequence with the fire extinguishing control instruction sequence to generate an execution instruction queue.

[0034] S11, controlling the autonomous flying fire extinguisher to complete the fire extinguishing task according to the execution instruction queue.

[0035] S12. Update the fire positive development matrix and the fire negative development matrix in real time through the sensor data.

[0036] S13. According to the updated fire positive development matrix and the fire negative development matrix, determine whether it is necessary to re-plan the route and generate a new execution instruction queue.

[0037] S14. After completing the fire extinguishing task, control the autonomous flying fire extinguisher to return to the autonomous flying fire extinguisher platform through laser navigation, Bluetooth beacon navigation or satellite navigation.

[0038] The fire-fighting path planning model uses a convolutional neural network structure as the basic model, and adds a mathematical calculation module between the convolution layer and the fully connected layer. The mathematical calculation module is used to integrate the fire development situation and the fire-fighting resource constraints to perform route optimization calculations. The mathematical calculation module contains a fire spread prediction equation, a resource consumption evaluation equation, and a route cost calculation equation.

[0039] The fire spread prediction equation is used to calculate the fire spread range and speed. The input includes fire temperature distribution data, ambient wind speed data, combustible distribution data, and terrain height data. The output is a sequence of fire spread boundary coordinates at future moments.

[0040] The resource consumption evaluation equation is used to estimate the use of fire-fighting resources. The input includes fire scene temperature data, fire scene area data, fire extinguishing agent storage data, fire extinguisher remaining power data, and target fire point distance data. The output is the predicted value of fire-fighting resource consumption.

[0041] The route cost calculation equation is used to calculate the comprehensive cost of different route plans. The input includes route length data, fire scene danger data, fire extinguishing efficiency prediction data, energy consumption data, passage difficulty data, and task completion time data. The output is the comprehensive cost score of the route plan.

[0042] The fire extinguishing path planning model includes an input layer, a first convolution layer, a second convolution layer, a third convolution layer, a mathematical calculation module, a fully connected layer, and an output layer. The input layer receives fire scene grid data, and the fire scene grid data includes temperature distribution data, fire intensity data, and combustible distribution data. The first convolution layer uses an edge detection convolution kernel group, and the edge detection convolution kernel group is generated by a fire scene boundary feature extraction function, and the fire scene boundary feature extraction function is constructed based on the temperature gradient change rate, the fire intensity change rate, and the combustible density change rate. The second convolution layer uses a direction-aware convolution kernel group, and the direction-aware convolution kernel group is generated by a fire spread direction identification function, and the fire spread direction identification function is constructed based on the heat transfer direction and the ambient wind direction. The third convolution layer uses an intensity assessment convolution kernel group, and the intensity assessment convolution kernel group is generated by a fire intensity analysis function, and the fire intensity analysis function is constructed based on thermal imaging data, flame spectrum data, and smoke concentration data.

[0043] The weight coefficient of each convolution kernel in the edge detection convolution kernel group is determined by the fire scene boundary clarity evaluation function, and the input of the fire scene boundary clarity evaluation function includes boundary temperature gradient, boundary infrared characteristics, and boundary visible light characteristics. The weight coefficient of each convolution kernel in the direction perception convolution kernel group is determined by the fire spread trend evaluation function, and the input of the fire spread trend evaluation function includes heat diffusion speed, wind field data, and terrain factors. The weight coefficient of each convolution kernel in the intensity assessment convolution kernel group is determined by the fire hazard evaluation function, and the input of the fire hazard evaluation function includes flame temperature, thermal imaging intensity, and flame spectrum characteristics.

[0044] The mathematical calculation module receives the feature map output of the third convolutional layer, and calculates the optimal route plan in combination with the fire positive development matrix and the fire negative development matrix.

[0045] The fully connected layer converts the output of the mathematical calculation module into a waypoint coordinate sequence, where the waypoint coordinate sequence includes a plane coordinate value, a flight altitude value, and an arrival time value.

[0046] The output layer generates a complete route planning result, which includes take-off point coordinates, landing point coordinates, waypoint coordinates, and mission termination conditions.

[0047] The steps of establishing the training data set of the fire extinguishing path planning model include: obtaining historical fire data, the historical fire data including fire scene temperature distribution data, fire spread speed data, environmental wind speed data, and terrain data; obtaining historical fire extinguishing action data, the historical fire extinguishing action data including fire extinguisher route data, fire extinguishing placement location data, and fire extinguishing effect evaluation data; gridding the historical fire data to generate fire scene grid feature data; and performing trajectory analysis on the historical fire extinguishing action data to generate optimal route example data.

[0048] The training data set is subjected to data enhancement processing, comprising the following sub-steps: performing rotation transformation on the fire scene grid feature data to generate fire distribution samples in different directions; performing scale transformation on the fire scene grid feature data to generate fire scene distribution samples of different scales; and performing interpolation processing on the optimal route example data to generate continuous route trajectory samples.

[0049] Training the fire extinguishing path planning model includes the following sub-steps: inputting the fire scene grid feature data into the fire extinguishing path planning model; using the optimal route example data as a training target; calculating the error between the route planning result output by the fire extinguishing path planning model and the training target; and updating the weight coefficient of each layer of the convolution kernel in the fire extinguishing path planning model according to the error.

[0050] The fire extinguishing path planning model is verified, comprising the following sub-steps: using a verification data set to evaluate the route planning performance of the fire extinguishing path planning model; calculating the route planning performance indicators, the route planning performance indicators including a route length score, a fire extinguishing efficiency score, and a safety score; and optimizing and adjusting the fire extinguishing path planning model according to the route planning performance indicators.

[0051] The fire positive development matrix is ​​used to represent the fire spread situation. Each element in the matrix represents the fire intensity, spread speed and direction at the corresponding position. The fire negative development matrix is ​​used to represent the fire decay after the fire extinguishing agent is applied. Each element in the matrix represents the fire decay rate, fire shrinkage direction and fire extinguishing effect evaluation value at the corresponding position.

[0052] The fire extinguishing measures vector is used to represent the specific measures combination when performing fire extinguishing operations at a specific location. The elements in the vector correspond to the usage and duration of different types of fire extinguishing resources. The fire weight value is used to represent the danger level and fire extinguishing priority of different locations in the fire scene. The larger the weight value, the more priority the location needs to be extinguished.

[0053] The execution instruction queue is used to represent the complete instruction sequence for the autonomous flying fire extinguisher to perform the fire extinguishing task, including the unified scheduling sequence of flight control instructions and fire extinguishing operation instructions. The fire scene boundary feature extraction function is used to characterize the geometric and physical characteristics of the fire scene boundary, and determine the precise position and morphological characteristics of the fire scene boundary by analyzing the temperature gradient, infrared image, and visible light image. The fire spread direction identification function is used to identify the main and secondary directions of the fire spread, and predict the fire development trend in combination with environmental factors. The fire intensity analysis function is used to evaluate the fire intensity level in different areas of the fire scene, and provide a basis for risk assessment for route planning. The fire scene boundary clarity evaluation function is used to evaluate the accuracy of fire scene boundary identification and guide the optimization of the weight parameters of the edge detection convolution kernel. The fire spread trend evaluation function is used to evaluate the speed and direction of fire spread and guide the optimization of the weight parameters of the direction perception convolution kernel. The fire danger evaluation function is used to evaluate the degree of danger of each area of ​​the fire scene and guide the optimization of the weight parameters of the intensity assessment convolution kernel. The fire scene grid feature data is used to represent the discretized fire scene status information, including the temperature value, fire intensity value, and spread speed value of each grid unit. The optimal route example data is used to represent the route plan with the best effect in historical firefighting operations, which serves as a reference target for model training. The route planning performance index is used to evaluate the quality of the route plan generated by the model, and conducts a comprehensive evaluation from three dimensions: route length, firefighting efficiency, and safety.

[0054] The resource utilization rate is used to indicate the efficiency of using fire-fighting resources, and is calculated by the ratio of fire-fighting effect to resource consumption per unit time.

[0055] The specific implementation of the above steps is described in detail below.

[0056] The specific implementation of step S01 is achieved through multi-sensor data acquisition and preprocessing. When acquiring sensor data, it is first necessary to calibrate and calibrate various sensors. The thermal imaging sensor data is collected at a sampling frequency of 60 Hz, the image resolution is set to 640×480 pixels, and the temperature detection range is minus 40 degrees Celsius to 1500 degrees Celsius. The visible light sensor data is collected at 1080p high-definition resolution, and the frame rate is set to 30 frames per second. The three-band flame sensor data is collected to simultaneously monitor the flame characteristics of the infrared band, ultraviolet band and visible light band, and the sampling frequency is set to 100 Hz. When the temperature and humidity sensor data is collected, the temperature measurement range is minus 50 degrees Celsius to 150 degrees Celsius, the humidity measurement range is 0 to 100% relative humidity, and the sampling frequency is 1 Hz. When the smoke sensor data is collected, the detection range is 0 to 1000ppm, and the response time is less than 10 seconds. When the carbon monoxide sensor data is collected, the detection range is 0 to 1000ppm, the response time is less than 15 seconds, and the sampling frequency is 1 Hz. The collected sensor data is subjected to noise removal using a median filter algorithm, the sliding window size is set to 5 data points, and then the data is smoothed using an interpolation smoothing algorithm. Finally, all sensor data are uniformly converted into a standard format and time synchronized.

[0057] The specific implementation method of step S02 is to construct a fire positive development matrix based on sensor data, which is realized by a multi-source data fusion algorithm. First, thermal imaging data and visible light data are used for image registration, and an image registration algorithm based on feature point matching is adopted. The scale-invariant feature transformation algorithm is used to extract image feature points, and the nearest neighbor ratio matching algorithm is used for feature point matching, and the matching threshold is set to 0.7. Then the registered image data is aligned with other sensor data in time and space to establish a unified data coordinate system. Then, the multi-source data is fused by the Kalman filter algorithm, and the state vector contains three components: fire spread speed, fire expansion direction and fire intensity. The fire spread speed is obtained by calculating the change in the position of the fire boundary at adjacent moments, the fire expansion direction is determined by calculating the main change direction of the fire boundary by the principal component analysis method, and the fire intensity is calculated by comprehensively analyzing the thermal imaging intensity, flame spectrum characteristics and smoke concentration. Finally, the fused data is organized into a fire positive development matrix, and each element of the matrix corresponds to a discrete grid point in the fire space.

[0058] The specific implementation method of step S03 is to calculate the negative development matrix of the fire situation based on the positive development matrix of the fire situation, and adopt the back propagation algorithm to realize it. First, establish the fire extinguishing agent action model, the model input includes the type of fire extinguishing agent, the amount of delivery, the delivery method and the environmental parameters, and the output is the fire extinguishing effect prediction value. For different types of fire extinguishing agents, their action characteristic models are established respectively: the action characteristics of the fire extinguishing bomb are described by the explosion shock wave model, the action characteristics of the fire extinguishing bottle are described by the injection diffusion model, and the action characteristics of the fine water mist are described by the droplet atomization model. Then the fire extinguishing agent action model is convoluted with the positive development matrix of the fire situation, and the fire decay speed, the fire contraction direction and the degree of fire reduction are calculated. Finally, the calculation results are organized into the negative development matrix of the fire situation, and the matrix structure corresponds to the positive development matrix of the fire situation. The fire extinguishing effect evaluation adopts the fuzzy comprehensive evaluation method, and the evaluation factors include the temperature drop rate, the fire area reduction rate and the smoke dissipation rate. The evaluation level is divided into four levels: significant, good, general and poor.

[0059] The specific implementation method of step S04 is to establish a fire extinguishing strategy equation group based on the positive development matrix of the fire situation and the negative development matrix of the fire situation, and implement it using a dynamic programming algorithm. First, a single-machine single-time fire extinguishing duration calculation model is established. The model takes into account factors such as fire intensity, fire scene area, type of fire extinguishing agent and wind speed, and uses a multivariate linear regression method to establish a calculation formula. Then, a multi-machine multiple-time fire extinguishing duration calculation model is established. The model takes into account factors such as the number of machines, synergy and resource constraints, and uses a nonlinear optimization method to solve the optimal fire extinguishing duration. Finally, the single-machine and multi-machine models are combined into a complete fire extinguishing strategy equation group, and the solution of the equation group adopts an iterative calculation method. The calculation result of the fire extinguishing duration serves as an important constraint condition for subsequent route planning, and is also used to evaluate the deployment plan of fire extinguishing resources.

[0060] The specific implementation method of step S05 is to use the fire extinguishing strategy equation group to calculate the fire extinguishing delivery measure vector of different waypoints, and adopt genetic algorithm to realize the optimization solution. First, the waypoints are divided into several fire extinguishing operation areas, and the fire characteristics of each area are determined by the data of the corresponding position of the fire forward development matrix. Then, according to the fire characteristics, a suitable combination of fire extinguishing agents is selected. The selection of fire extinguishing agents is based on the fuzzy decision method, and the decision factors include fire intensity, fire scene environment and cost-effectiveness. For each waypoint, the optimal delivery amount of different types of fire extinguishing agents is calculated. The calculation of the delivery amount of fire extinguishing bombs takes into account the explosion range and fire intensity, the calculation of the delivery amount of fire extinguishing bottles takes into account the spray coverage area and the size of the fire scene, and the calculation of the spray duration of fine water mist takes into account the influence of atomization effect and wind speed. Finally, the calculation results are organized into a fire extinguishing delivery measure vector, and each component in the vector corresponds to the usage or duration of a fire extinguishing resource. The efficiency of the use of fire extinguishing resources is evaluated by the resource utilization index, and the calculation of resource utilization adopts the input-output ratio analysis method.

[0061] The specific implementation method of step S06 is to establish a fire scene space division grid according to the fire forward development matrix, and adopt an adaptive grid division algorithm to achieve it. First, the basic size of the grid division is determined according to the actual scope of the fire scene, and the initial size of the grid is set to 5 meters × 5 meters. Then, the grid is adaptively refined according to the fire distribution characteristics, and a smaller grid size is used in areas with strong fire or large changes. The minimum grid size after refinement is 1 meter × 1 meter. A fire weight value is assigned to each grid node. The calculation of the weight value adopts the hierarchical analysis method, and the factors considered include temperature value, fire intensity, spread speed, terrain characteristics and the distribution of surrounding important targets. The weight coefficient of the temperature value is 0.3, the weight coefficient of the fire intensity is 0.25, the weight coefficient of the spread speed is 0.2, the weight coefficient of the terrain characteristics is 0.15, and the weight coefficient of the surrounding important targets is 0.1. Finally, the weight values ​​of the grid nodes are normalized so that the weight values ​​are distributed between 0 and 1.

[0062] The specific implementation method of step S07 is to use the pre-trained fire-fighting path planning model to divide the fire space into grids for route planning, and use the deep reinforcement learning algorithm to implement it. First, the fire grid data is input into the convolutional neural network part of the pre-trained model, and the fire scene features are extracted through three-layer convolution operations. The first layer of convolution uses a 3×3 convolution kernel for edge detection, the second layer of convolution uses a 5×5 convolution kernel to identify the direction of fire spread, and the third layer of convolution uses a 7×7 convolution kernel to evaluate the fire intensity. Then the convolution features are input into the mathematical calculation module, and the route is optimized in combination with the fire development trend and fire-fighting resource constraints. The route optimization uses an improved ant colony algorithm, takes the fire weight value as heuristic information, and takes the route length and fire-fighting efficiency as evaluation indicators. Finally, the route planning results are generated, including multiple waypoint coordinates, waypoint arrival order, and waypoint hovering time. The number of waypoints is determined according to the scale of the fire and the requirements of the fire-fighting task, and is generally set to 10 to 20 points. The distance between adjacent waypoints does not exceed 50 meters. The hovering time of each waypoint is calculated based on the fire-fighting measures vector of that point.

[0063] The specific implementation method of step S08 is to calculate the flight trajectory curve of the autonomous flying fire extinguisher according to the route planning results, and use the Bezier curve interpolation algorithm to achieve it. First, the waypoint coordinate sequence is used as the control point to construct a cubic Bezier curve, and the smoothness of the curve is adjusted by the tangent vector of the control point. Then, the dynamic constraints of the aircraft, including the maximum speed, maximum acceleration and maximum turning angle, are considered to correct the Bezier curve. The maximum speed of the aircraft is set to 15 meters per second, the maximum acceleration is set to 2 meters per square second, and the maximum turning angle is set to 45 degrees. Finally, the corrected flight trajectory curve is discretized into a flight control instruction sequence, and the control instruction includes three-dimensional position coordinates, flight speed, flight attitude and hovering state. The sampling period of the flight control instruction is set to 0.1 seconds to ensure the smoothness and executability of the flight trajectory.

[0064] The specific implementation method of step S09 is to generate a fire extinguishing control instruction sequence based on the fire extinguishing measures vector, and implement it using a state machine control algorithm. First, the fire extinguishing measures vector is parsed into specific execution actions, including the working state of the high-voltage generator, the trigger timing of the electromagnetic throwing controller, and the flow control of the fine water mist pump. Then, the control strategy is designed according to the working characteristics of different fire extinguishing equipment. The control of the high-voltage generator adopts pulse width modulation, the control of the electromagnetic throwing controller adopts a timing triggering method, and the control of the fine water mist pump adopts a variable frequency speed regulation method. Finally, a fire extinguishing control instruction sequence is generated, and the instruction sequence includes specific operations such as equipment startup, working parameter adjustment, and equipment shutdown. The execution process of the fire extinguishing control instruction adopts real-time monitoring and feedback control to ensure the fire extinguishing effect.

[0065] The specific implementation method of step S10 is to align the flight control instruction sequence with the fire extinguishing control instruction sequence in time sequence, and implement it using a task scheduling algorithm. First, a unified time reference is established to unify the timestamps of the two instruction sequences into the same coordinate system. Then, according to the order and parallel relationship of task execution, the instruction sequence is organized into a directed acyclic graph structure. Finally, the critical path algorithm is used to calculate the optimal execution order and generate an execution instruction queue. Each instruction in the execution instruction queue contains three elements: execution time, execution content, and execution conditions. The execution of the instruction adopts an event-driven method to ensure the coordination and consistency of flight control and fire extinguishing control.

[0066] The specific implementation method of step S11 is to control the autonomous flying fire extinguisher to complete the fire extinguishing task according to the execution instruction queue, and adopt a hierarchical control architecture. The top layer of the control system is the task management layer, which is responsible for the parsing of the execution instruction and the monitoring of the task status, and adopts a finite state machine to realize the task process control. The middle layer is the trajectory tracking layer, which is responsible for the motion control of the aircraft, and adopts the model predictive control algorithm to realize the accurate tracking of the planned trajectory. The control cycle is set to 50 milliseconds, the position control accuracy is better than 0.5 meters, and the attitude control accuracy is better than 2 degrees. The bottom layer is the execution control layer, which is responsible for the direct control of each actuator, including motor control, steering gear control and fire extinguishing equipment control, and is implemented by a cascade PID control algorithm. The inner loop control cycle is 2 milliseconds and the outer loop control cycle is 10 milliseconds. During the execution of the task, the control system collects flight status information and fire extinguishing effect information in real time, and performs adaptive adjustment based on this information to ensure the reliable completion of the task. When an abnormal situation is detected, the control system will start the emergency handling procedure, including hovering waiting, obstacle avoidance and emergency return.

[0067] The specific implementation method of step S12 is to update the positive development matrix and the negative development matrix of the fire situation in real time through sensor data, and adopt the data processing method of sliding time window to achieve it. First, set the update cycle to 1 second, and collect the latest data of all sensors in each update cycle. Then use the data fusion algorithm to process the multi-source sensor data, including three steps: data preprocessing, feature extraction and information fusion. The data preprocessing adopts low-pass filtering and outlier detection methods, the filter cutoff frequency is set to 10 Hz, and the outlier judgment threshold is set to 3 times the standard deviation. The feature extraction adopts the wavelet transform method to perform time-frequency analysis on the sensor data, selects the db4 wavelet as the basis function, and sets the decomposition layer number to 4 layers. The information fusion adopts the evidence theory method to construct the basic probability distribution function, and sets the credibility threshold to 0.8. Finally, the fire development matrix is ​​updated according to the fusion result. The update process adopts the exponential weighted average method, and the weight coefficient of the historical data is set to 0.7.

[0068] The specific implementation method of step S13 is to determine whether it is necessary to re-plan the route and generate a new execution instruction queue based on the updated fire positive development matrix and the fire negative development matrix, and to implement it using a dynamic decision algorithm. First, the change in the fire development matrix is ​​calculated, and the matrix difference method is used to calculate the change in fire intensity, the change in the direction of spread, and the change in the scope of the fire. Then, the re-planning trigger conditions are set, including a change in fire intensity of more than 30%, a change in the direction of spread of more than 45 degrees, or a change in the scope of the fire of more than 20%. When any trigger condition is met, the re-planning program is started. The re-planning program first evaluates the effectiveness of the current route, and uses the Monte Carlo simulation method to simulate 100 task execution processes and calculate the task completion rate. If the task completion rate is less than 85%, it is determined that re-planning is required. The re-planning process uses a heuristic search algorithm, guided by the principle of minimization adjustment, to retain the still valid parts of the original route as much as possible, and only locally optimize the invalid parts. Finally, a new execution instruction queue is generated, and smooth switching between the new and old instruction queues is ensured.

[0069] The specific implementation method of step S14 is to control the autonomous flying fire extinguisher to return to the platform, which is implemented by a multi-source navigation fusion algorithm. First, the return path is calculated according to the current position and the platform position, and the a-star algorithm is used to implement path planning, taking into account obstacle avoidance and energy constraints. Then, a suitable navigation method is selected. When the distance from the platform is greater than 100 meters, satellite navigation is used first, and the navigation accuracy is better than 5 meters; when the distance is between 30 and 100 meters, laser navigation is used, and the navigation accuracy is better than 0.5 meters; when the distance is less than 30 meters, Bluetooth beacon navigation is used, and the navigation accuracy is better than 0.1 meters. A smooth transition strategy is adopted for switching between different navigation methods, and an overlapping interval of 10 meters is set. The flight control during the return process adopts an adaptive control algorithm, and the flight parameters are dynamically adjusted according to factors such as battery power and wind speed. When approaching the platform, it enters the precise landing stage, and a visual servo control method is used to identify the characteristic markers on the platform to achieve landing control with centimeter-level accuracy. The landing speed is controlled below 0.5 meters per second, and the descent speed at contact does not exceed 0.2 meters per second. After landing, the shutdown procedure is automatically executed, including equipment status check, data saving and system shutdown.

[0070] The calculation process or matrix involved in the present invention is described in detail below.

[0071] The fire forward development matrix F forward The specific expressions are as follows:

[0072]

[0073] Each element f ij Expressed as: f ij =α1v ij +α2d ij +α3s ij +ε1;

[0074] Where: v ij is the fire spread speed at the grid point (i, j), in m / s; d ij is the fire expansion direction angle, in rad; s ij is the fire intensity value, dimensionless; α1, α2, α3 are weight coefficients, determined by expert experience, satisfying α1+α2+α3=1; ε1 is the error term, ranging from 0 to 0.1.

[0075] Fire spread speed v ij The calculation formula is:

[0076] Where: x ij ,y ij is the coordinate of the fire boundary point, obtained through thermal imaging image processing; t is the time variable.

[0077] Fire spread angle d ij The calculation formula is:

[0078] Fire intensity value s ij The calculation formula is: ij =β1T ij +β2I ij +β3C ij +ε2;

[0079] Where: T ij is the temperature value, in °C; I ij is the infrared image intensity value, ranging from 0 to 255; C ij is the carbon monoxide concentration value in ppm; β1, β2, β3 are weight coefficients, satisfying β1+β2+β3=1; ε2 is the error term, ranging from 0 to 0.1.

[0080] The fire forward development matrix describes the spatial distribution characteristics of the fire scene, which is achieved through multi-source sensor data fusion. The fire spread speed is calculated in the form of partial derivatives, reflecting the rate of change of the fire boundary over time; the fire expansion direction is calculated by the inverse tangent function, which can accurately describe the direction angle of fire spread; the fire intensity value adopts a linear weighted method, comprehensively considering the three key indicators of temperature, infrared intensity and carbon monoxide concentration.

[0081] The fire negative development matrix F backward The specific expressions are as follows:

[0082] Each element b ij It is expressed as:

[0083] b ij =γ1r ij +γ2c ij +γ3e ij +ε3;

[0084] Where: r ij is the fire decay rate, in 1 / s; c ij is the fire contraction angle, in rad; e ij is the fire extinguishing effect evaluation value, dimensionless; γ1, γ2, γ3 are weight coefficients, satisfying γ1+γ2+γ3=1; ε3 is the error term, ranging from 0 to 0.1.

[0085] Fire decay rate r ij The calculation formula is:

[0086] Where: T ij is the temperature value in °C; t is the time variable.

[0087] Fire shrinkage angle c ij The calculation formula is:

[0088] Fire extinguishing effect evaluation value e ij The calculation formula is:

[0089] Where: ΔT ij , T ij0 are the temperature drop value and the initial temperature value, in °C; ΔA ij , A ij0 are the fire area reduction value and the initial area value, in m 2 ; ΔS ij , S ij0 are the smoke concentration reduction value and initial concentration value, respectively, in ppm; δ1, δ2, δ3 are weight coefficients, satisfying δ1+δ2+δ3=1; ε4 is the error term, ranging from 0 to 0.1.

[0090] The negative development matrix of the fire situation describes the changing characteristics of the fire scene after the fire extinguishing agent acts. The fire decay rate is in the form of logarithmic derivative, which can reflect the relative rate of change of temperature over time; the fire contraction direction is expressed as the opposite direction of the fire expansion by adding a π phase difference; the fire extinguishing effect evaluation is in the form of relative change rate, which can eliminate the influence of different dimensions.

[0091] The fire extinguishing measures vector M is specifically expressed as follows: M = [n b , n e , t w ] T ;

[0092] Where: n b is the number of fire extinguishing bombs released; n e is the number of fire extinguisher bottles put in; t w It is the duration of water mist spraying, in seconds.

[0093] Number of fire extinguishing bombs released b The calculation formula is:

[0094] Where: A f is the fire area, in m 2 ; A b is the effective extinguishing area of ​​a single fire bomb, in m 2 ; T f is the average temperature of the fire scene, in °C; T0 is the reference temperature, which is 500 °C; v w is the wind speed, in m / s; v0 is the reference wind speed, with a value of 5 m / s; k1 and k2 are correction coefficients, with values ​​of 0.2 and 0.3 respectively; Indicates rounding up.

[0095] Number of fire extinguisher bottles placed n e The calculation formula is:

[0096] Where: A e is the effective fire extinguishing area of ​​a single fire extinguisher, in m 2 ;H f is the flame height in meters; H0 is the reference height, which is 2 meters; k3 and k4 are correction coefficients, which are 0.2 and 0.25 respectively.

[0097] Water mist spraying time t w The calculation formula is:

[0098] Where: V f is the volume of the fire scene, in m 3 ;Q w is the water mist spray flow rate, in m 3 / s;S f is the smoke concentration in ppm; S0 is the reference concentration, which is 500ppm; k5 and k6 are correction coefficients, which are 0.15 and 0.2 respectively.

[0099] The fire-fighting measure vector is used to describe the optimal configuration of different types of fire-fighting resources. The number of fire-fighting bombs and fire-fighting bottles is calculated based on the influence of temperature and wind speed, and the number is rounded up to ensure the fire-fighting effect; the water mist spraying duration is calculated based on the volume flow rate, and the influence of temperature and smoke concentration is taken into account.

[0100] The fire weight value w ij The specific expressions are as follows:

[0101]

[0102] Where: T ij , T max are the grid point temperature value and the maximum temperature value, in °C; s ij ,s max are the fire intensity value and the maximum intensity value, dimensionless; v ij , w max are the spreading speed value and the maximum speed value, in m / s; h ij is the terrain elevation factor, ranging from 0 to 1; d ij is the important target distance factor, ranging from 0 to 1; η1, η2, η3, η4, η5 are weight coefficients, satisfying ε5 is the error term, ranging from 0 to 0.1.

[0103] The route cost calculation equation is specifically expressed as follows:

[0104] C path =λ1L path +λ2R risk +λ3E eff +λ4P energy +λ5D diff +λ6T task +ε6;

[0105] Where: L path is the route length, in meters; R risk is the fire danger degree, dimensionless; E eff is the fire extinguishing efficiency, dimensionless; P energy is energy consumption, in kW·h; D diff is the difficulty of passage, dimensionless; T task is the task completion time, in seconds; λ1, λ2, λ3, λ4, λ5, λ6 are weight coefficients, satisfying ε6 is the error term, ranging from 0 to 0.1.

[0106] The fire spread prediction equation is specifically expressed as follows:

[0107] P spread =μ1T dist +μ2V wind +μ3M fuel +μ4H terrain +ε7;

[0108] Where: T dist is the fire temperature distribution function; V wind is the ambient wind speed function; M fuel is the combustible distribution function; H terrain is the terrain height function; μ1, μ2, μ3, μ4 are weight coefficients, satisfying ε7 is the error term, ranging from 0 to 0.1.

[0109] The resource consumption evaluation equation is specifically expressed as follows:

[0110] R consume =ρ1T f +ρ2A f +ρ3Q s +ρ4E r +ρ5D t +ε8;

[0111] Where: T f is the fire temperature, in °C; A f is the fire area, in m 2 ;Q sis the storage capacity of fire extinguishing agent, in kg; E r is the remaining power of the fire extinguisher, in kW·h; D t is the target fire point distance, in meters; ρ1, ρ2, ρ3, ρ4, ρ5 are weight coefficients, satisfying ε8 is the error term, ranging from 0 to 0.1.

[0112] The resource utilization calculation formula is specifically expressed as follows:

[0113]

[0114] Where: E effect is the fire extinguishing effect evaluation value, dimensionless; R consume is the resource consumption, in kg; t task It is the task execution time, in seconds.

[0115] The state vector is specifically expressed as follows:

[0116] x=[v spread , d expand ,s intensity ] T ;

[0117] Where: v spread is the fire spread speed, in m / s; d expand The direction of fire expansion, in rad; s intensity is the fire intensity, dimensionless.

[0118] The process of deriving and establishing each matrix or equation is described in detail below.

[0119] 1. Fire positive development matrix F forward Derivation of:

[0120] First, establish a coordinate system and divide the fire scene into m×n grids. Each grid point needs to represent three characteristics: spread speed, expansion direction and fire intensity. Therefore, construct a three-dimensional matrix:

[0121] F forward ={V, D, S};

[0122] The velocity matrix V is:

[0123]

[0124] The direction matrix D is:

[0125]

[0126] The intensity matrix S is:

[0127]

[0128] Through multi-sensor data fusion, the Kalman filter algorithm is used to update the matrix elements:

[0129] Equation of state: X k =AX k-1 +BU k +W k ;

[0130] Observation equation: Z k =HX k +V k ;

[0131] Where: X k is the state vector; A is the state transfer matrix; B is the control matrix; U k is the control vector; W k is the process noise; Z k is the observation vector; H is the observation matrix; V k is the observation noise.

[0132] 2. Fire negative development matrix F backward Derivation of:

[0133] Considering the effect of fire extinguishing agent, the fire extinguishing agent action model is established:

[0134] Fire extinguisher bomb explosion shock wave model:

[0135] Where: p(r, t) is the pressure value at time t at a distance r; p0 is the explosion center pressure; R0 is the characteristic radius; α is the attenuation coefficient.

[0136] Fire extinguisher spray diffusion model:

[0137] Where: c(r, t) is the concentration value at time t at a distance r; Q is the release amount; D is the diffusion coefficient.

[0138] Water mist atomization model: d 32 =A.We -0.6 ·Re 0.15 ;

[0139] Where: d 32 is the Sauter mean diameter; We is the Weber number; Re is the Reynolds number; A is the experimental constant.

[0140] 3. Derivation of fire extinguishing measures vector M:

[0141] Optimization solution based on genetic algorithm:

[0142] Fitness function:

[0143] Where: E effect For fire extinguishing effect; E max is the maximum effect value; C cost is the resource cost; C max is the maximum cost; ω1, ω2 are weight coefficients.

[0144] Crossover operator: M new =φM1+(1-φ)M2;

[0145] Where: M1, M2 are parent individuals; φ is the random crossover coefficient.

[0146] Mutation operator: M mut =M+σN(0,1);

[0147] Where: σ is the variable step length; N(0,1) is a standard normal distribution random number.

[0148] 4. Fire weight value w ij Derivation of:

[0149] The weight coefficient is determined by using the hierarchical analysis method:

[0150] Construct a judgment matrix:

[0151]

[0152] Where: a ij is the importance ratio of the i-th indicator to the j-th indicator.

[0153] Calculate the weight vector: w = [η1, η2, ..., η n ] T ;

[0154] Check consistency:

[0155] Where: max is the maximum eigenvalue; RI is the random consistency index.

[0156] 5. Derivation of route cost calculation equation:

[0157] Using a multi-objective optimization approach:

[0158] Objective function vector: F(x) = [f1(x), f2(x), ..., f6(x)] T ;

[0159] Where: f1(x) is the route length; f2(x) is the danger level; f3(x) is the fire extinguishing efficiency; f4(x) is the energy consumption; f5(x) is the difficulty of passage; and f6(x) is the completion time.

[0160] Constraints:

[0161] g i (x)≤0, i=1, 2,...,p;

[0162] h j (x)=0, j=1, 2,...,q;

[0163] The optimal weight coefficient is obtained through Pareto optimization solution.

[0164] 6. Derivation of fire spread prediction equation:

[0165] Modeling based on partial differential equations:

[0166] Heat conduction equation:

[0167] Where: α is the thermal diffusion coefficient; Q(x, y, t) is the heat source term.

[0168] Wind field influence equation:

[0169] Where: V is the wind speed vector; p is the pressure; ρ is the density; v is the kinematic viscosity; F is the external force term.

[0170] 7. Derivation of resource consumption evaluation equation:

[0171] Based on the energy balance principle:

[0172] Energy input:

[0173] Energy consumption: E out =E flight +E extinguish +E loss ;

[0174] Where: E flight is the flight energy consumption; E extinguish is the energy consumption for fire extinguishing; E loss is the loss term.

[0175] Specifically, the principle of the present invention is: the core principle of the present invention is to achieve intelligent decision-making and control of the fire extinguishing process in a complex fire environment by establishing a mathematical description model of the fire situation and combining deep learning technology. First, the positive development matrix and the negative development matrix of the fire are constructed through multi-source sensor data. These two matrices describe the spread characteristics of the fire and the characteristics of the fire extinguishing effect, respectively, and provide a mathematical basis for subsequent decision optimization. The positive matrix characterizes the dynamic characteristics of the fire development by analyzing parameters such as temperature gradient and fire intensity; the negative matrix predicts the fire attenuation law by evaluating the effect of the fire extinguishing agent.

[0176] On this basis, the present invention designs a special convolutional neural network structure, which contains three types of functional convolution kernels: edge detection, direction perception, and intensity assessment. These convolution kernels are responsible for extracting the boundary features of the fire scene, identifying the direction of fire spread, and assessing the intensity of the fire, respectively, to form a comprehensive perception of the fire scene status. By introducing a mathematical calculation module into the network, a quantitative analysis of the fire development trend and the constraints of fire-fighting resources is achieved, ensuring that the generated route plan meets both the requirements of fire-fighting efficiency and the resource constraints.

[0177] The present invention adopts a modular design concept, decomposing the complex fire extinguishing control problem into three links: perception, planning, and execution, and realizing the coordinated optimization of each link through a data-driven approach. The system continuously learns historical fire extinguishing data and continuously optimizes the parameters of the decision-making model, thereby improving its adaptability in different fire scenarios. This method of combining traditional control theory with modern artificial intelligence technology provides a new technical idea for solving the problem of accurate fire extinguishing in complex environments.

[0178] A specific embodiment 1 of the present invention is provided below, and the specific implementation method of each step in this embodiment 1 is described in detail as follows.

[0179] The specific implementation method of step S01 is to first calibrate the sensor of the autonomous flying fire extinguisher. The thermal imaging sensor is calibrated using a high-temperature black body as a reference. A mapping relationship is established by comparing the output values ​​at different temperatures. The calibration temperature range is -40°C to 1500°C, with a sampling frequency of 60Hz, an image resolution of 640×480 pixels, and a grayscale resolution of 12 bits. The visible light sensor is calibrated using a standard light source and a color card, with a resolution of 1080p, a frame rate of 30fps, and a dynamic range of 90dB. The three-band flame sensor is calibrated for the infrared band of 2~5μm, The ultraviolet band 180-260nm and the visible light band 400-700nm are calibrated, and the sampling frequency is set to 100Hz; the temperature and humidity sensor is calibrated using a constant temperature and humidity chamber, the temperature measurement range is -50℃ to 150℃, the relative humidity measurement range is 0 to 100%, and the sampling frequency is 1Hz; the smoke sensor is calibrated using standard gas, with a detection range of 0 to 1000ppm and a response time of less than 10 seconds; the carbon monoxide sensor is calibrated using standard gas, with a detection range of 0 to 1000ppm, a response time of less than 15 seconds, and a sampling frequency of 1Hz.

[0180] The collected sensor data is then preprocessed. First, the median filter algorithm is used to remove noise. The sliding window size is set to 5 data points. The data in each window is sorted by size and the middle value is taken as the output. Then the cubic spline interpolation algorithm is used to smooth the data. Local extreme points are selected as interpolation nodes to ensure the continuity of the first-order and second-order derivatives of the interpolation curve. Finally, all sensor data are uniformly converted into a standard format and time-synchronized. The linear interpolation method is used to align data with different sampling frequencies to a unified time base. The time base selects the highest sampling frequency of 60Hz.

[0181] The quality of the pre-processed data is evaluated, and the data validity judgment criteria are set: the signal-to-noise ratio is not less than 20dB, the data integrity rate is not less than 95%, and the time synchronization error does not exceed 10ms. If the data quality of a sensor does not meet the requirements, the sensor self-test program is started and data compensation is performed. Data compensation uses the Kalman filter algorithm to predict missing values ​​based on historical data and physical models. The compensated data needs to be re-evaluated until it meets the requirements.

[0182] The specific implementation method of step S02 is to first perform image registration on the thermal imaging data and the visible light data, and use a feature point matching algorithm based on scale-invariant feature transformation to extract significant feature points in the image, establish feature point correspondence through the nearest neighbor ratio matching method, set the matching threshold to 0.7, and then use a random consistency sampling algorithm to eliminate erroneous matching points, and calculate the affine transformation matrix to achieve image alignment.

[0183] Then, the registered image data is aligned with other sensor data in time and space to establish a unified data coordinate system. The origin of the coordinate is selected at the center of the fire scene, the x-axis points to the east, the y-axis points to the north, and the z-axis points to the top. The multi-source data fusion algorithm is used to process the aligned data, specifically the Kalman filter algorithm. The state vector contains three components: fire spread speed, fire expansion direction, and fire intensity. According to the formula in the article: Fire spread speed Obtained by calculating the change in the fire boundary position at adjacent moments; the direction of fire expansion The main change direction of the fire scene boundary is determined by calculating the principal component analysis method; the fire intensity s ij =β1T ij +β2I ij +β3C ij +ε2 is calculated by comprehensive analysis of thermal imaging intensity, flame spectrum characteristics and smoke concentration.

[0184] Finally, the fire positive development matrix F is constructed forward , each element of the matrix corresponds to a discrete grid point in the fire scene space, and the element value f ij =α1v ij +α2dij +α3s ij +ε1 represents the fire development characteristics of the point. The weight coefficients α1, α2, and α3 are determined by the hierarchical analysis method. First, the judgment matrix is ​​constructed, then the eigenvector is calculated to obtain the weight value, and finally a consistency test is performed. The consistency ratio must be less than 0.1. The error term ε1 ranges from 0 to 0.1 and is used to represent the uncertainty of the model.

[0185] The specific implementation method of step S03 is to first establish a fire extinguishing agent action model, and establish action characteristic models for different types of fire extinguishing agents. For fire extinguishing bombs, the explosion shock wave model is used to describe its action characteristics, and the waveform equation is: The pressure attenuation coefficient α is determined by experiment. For the fire extinguisher, the spray diffusion model is used to describe its action characteristics, and the concentration distribution equation is: The diffusion coefficient D is determined experimentally; for water mist, the droplet atomization model is used to describe its action characteristics, and the Sotter average diameter calculation formula is d 32 =A.We -0.6 ·Re 0.15 , the coefficient a is calibrated through experiments.

[0186] Then the fire extinguishing agent action model is convolved with the fire forward development matrix to calculate the fire decay rate Fire shrinkage direction and fire reduction Finally, the fire negative development matrix F is constructed based on the calculation results. backward , where each element b ij =γ1r ij +γ2c ij +γ3e ij +ε3 represents the fire attenuation characteristics of the point. The weight coefficients γ1, γ2, and γ3 are determined by the fuzzy hierarchical analysis method, and the error term ε3 ranges from 0 to 0.1.

[0187] The specific implementation method of step S04 is to first establish a single-machine single-time fire extinguishing time calculation model, taking into account factors such as fire intensity, fire scene area, fire extinguishing agent type and wind speed, and using the multivariate linear regression method to establish a calculation formula. The input variables include the average temperature of the fire scene, the fire scene area, the fire extinguishing agent type code, the ambient wind speed, etc. The output variable is the fire extinguishing time, and the regression coefficient is determined by the least squares method. The determination coefficient of the model needs to be greater than 0.85, and the root mean square error needs to be less than 10%.

[0188] Then, a calculation model for the duration of multiple firefighting operations with multiple machines was established, taking into account factors such as the number of machines, synergy effect, and resource constraints. The synergy effect coefficient was determined through simulation experiments, and resource constraints included the amount of fire extinguishing agent stored, battery power, and flight time. A nonlinear optimization method was used to solve the optimal firefighting duration. The objective function was to minimize the total firefighting time, and the constraints included resource constraints and safety constraints. The optimization algorithm used a sequential quadratic programming method, and the convergence accuracy was set to 0.001.

[0189] Finally, the single-machine and multi-machine models are combined into a complete set of fire-fighting strategy equations, and the fire-fighting strategies under different conditions are described in the form of piecewise functions. The equations are solved by iterative calculation method, with the maximum number of iterations set to 100 and the convergence threshold set to 0.01. The calculation results are used as constraints for subsequent route planning and are also used to evaluate the deployment plan of fire-fighting resources.

[0190] The specific implementation method of step S05 is to first divide the waypoints into fire-fighting operation areas, and the fire characteristics of each area are determined by the data of the corresponding position of the fire forward development matrix. The division principle is based on the spatial distribution of the fire intensity, and the clustering algorithm is used to achieve automatic division. The number of cluster centers is dynamically determined according to the scale of the fire. Then, according to the fire characteristics, the appropriate combination of fire extinguishing agents is selected, and the fuzzy decision method is used. The decision factors include fire intensity, fire environment and cost-effectiveness. A fuzzy rule base is established, and the Mamdan i reasoning method is used to obtain the fire extinguishing agent selection result.

[0191] For each waypoint, the optimal amount of different types of fire extinguishing agents is calculated according to the formula in the article: Number of fire extinguishing bombs released Consider the explosion range and fire intensity; the number of fire extinguisher bottles to be placed Consider the spray coverage area and fire size; spray duration of fine water mist Considering the atomization effect and the influence of wind speed, each coefficient is determined by experiment, and the value range of correction coefficients k1 to k6 is 0.1 to 0.3.

[0192] Finally, the calculation results are organized into a fire extinguishing measure vector M = [n b , n e , t w ] T , and use genetic algorithm to optimize and solve the best combination. The fitness function is The population size is set to 100, the number of iterations is 50, the crossover probability is 0.8, and the mutation probability is 0.1. The efficiency of firefighting resources is measured by the resource utilization index. Conduct an assessment.

[0193] The specific implementation method of step S06 is to first calculate the fire forward development matrix F forwardThe fire scene range is determined, and the edge detection algorithm is used to extract the fire scene boundary. The fire scene space is divided into basic grids, and the initial grid size is set to 5 meters × 5 meters. Then the grid is adaptively refined according to the fire distribution characteristics. A smaller grid size is used in areas with strong fire or large changes. The minimum grid size after refinement is 1 meter × 1 meter. The criteria for grid refinement include temperature gradient, fire intensity change rate and spread speed. When these parameters exceed the set threshold, the grid is refined. The temperature gradient threshold is 50℃ / meter, the intensity change rate threshold is 0.2 per second, and the spread speed threshold is 2 meters / second.

[0194] Assign a fire intensity weight value to each grid node. The weight value is calculated using the hierarchical analysis method according to the formula in the article. Consider factors such as temperature, fire intensity, spread speed, terrain characteristics, and the distribution of important surrounding targets. The weight coefficients η1 to η5 are determined by constructing a judgment matrix, and the consistency ratio of the judgment matrix must be less than 0.1. The weight coefficient of the temperature value is 0.3, the weight coefficient of the fire intensity is 0.25, the weight coefficient of the spread speed is 0.2, the weight coefficient of the terrain characteristics is 0.15, and the weight coefficient of the surrounding important targets is 0.1. Terrain elevation factor h ij Calculated from the digital elevation model, ranging from 0 to 1, the important target distance factor d ij It is calculated using a Gaussian decay function and ranges from 0 to 1.

[0195] Finally, the weight values ​​of the grid nodes are normalized so that the weight values ​​are distributed between 0 and 1. The range normalization method is used to ensure the comparability and consistency of the weight values. The spatial distribution of the normalized weight values ​​is analyzed, the spatial autocorrelation coefficient of the weight values ​​is calculated, and the rationality of the weight distribution is evaluated. If the spatial autocorrelation coefficient is too high, it means that the weight distribution is too concentrated, and the weight calculation parameters need to be adjusted; if the spatial autocorrelation coefficient is too low, it means that the weight distribution is too dispersed, and appropriate adjustments are also needed.

[0196] The specific implementation method of step S07 is to first input the fire scene grid data into the convolutional neural network part of the pre-trained fire extinguishing path planning model. The network structure includes an input layer, three convolution layers and a fully connected layer. The input layer receives the fire scene grid data, including temperature distribution data, fire intensity data and combustible distribution data. The first layer of convolution uses a 3×3 convolution kernel for edge detection, and the activation function uses ReLU with a step size of 1 and a filling method of SAME; the second layer of convolution uses a 5×5 convolution kernel to identify the direction of fire spread, using a leaky ReLU activation function with a step size of 2; the third layer of convolution uses a 7×7 convolution kernel to evaluate the fire intensity, using a Sigmoid activation function with a step size of 2.

[0197] The convolution features are then input into the mathematical calculation module to optimize the route in combination with the fire development trend and fire-fighting resource constraints. The mathematical calculation module contains three key equations: the fire spread prediction equation P spread =μ1T dist +μ2V wind +μ3M fuel +μ4H terrain +ε7, used to predict the development trend of the fire scene; resource consumption evaluation equation R consume =ρ1T f +ρ2A f +ρ3Q s +ρ4E r +ρ5D t +ε8, used to estimate resource usage; route cost calculation equation C path =λ1L path +λ2R risk +λ3E eff +λ4P enerygy +λ5D diff +λ6T task +ε6, used to evaluate different route options.

[0198] The route optimization adopts the improved ant colony algorithm, takes the fire weight value as the heuristic information, and takes the route length and fire extinguishing efficiency as the evaluation indicators. The parameter settings of the ant colony algorithm are as follows: the initial pheromone concentration is 1, the pheromone volatility coefficient is 0.1, the global update coefficient is 0.2, the local update coefficient is 0.1, and the number of iterations is 200. In each iteration, the next waypoint is selected according to the state transition probability formula. The heuristic factor α in the probability formula is 1, and the expected heuristic factor β is 2. Finally, the route planning results are generated, including the coordinates of multiple waypoints, the arrival order of waypoints, and the hovering time of waypoints. The number of waypoints is determined according to the scale of the fire and the requirements of the fire extinguishing task. It is generally set to 10 to 20 points, and the distance between adjacent waypoints does not exceed 50 meters.

[0199] The specific implementation method of step S08 is to first construct a cubic Bezier curve according to the route planning result. Each segment of the curve is determined by four control points, and the curve parameter equation is P(t)=(1-t) 3 P0+3t(1-t) 2 P1+3t 2 (1-t)P2+t 3 P3, where t ranges from 0 to 1. The selection of control points takes into account the smoothness and continuity of the curve, and the connection between adjacent curve segments needs to ensure the continuity of the first-order derivative. The tangent vector of the curve is determined by minimizing the curvature change rate to ensure the smoothness of the flight trajectory.

[0200] Then, the Bezier curve is modified considering the dynamic constraints of the aircraft. The constraints include a maximum speed of 15 meters per second, a maximum acceleration of 2 meters per second squared, and a maximum turning angle of 45 degrees. The correction process uses the quadratic programming method, and the objective function is to minimize the integral value of the trajectory curvature while satisfying the dynamic constraints. The constraint conditions are processed using the penalty function method, and the degree of constraint violation is converted into a penalty term and added to the objective function.

[0201] Finally, the corrected flight trajectory curve is discretized into a sequence of flight control instructions, which include three-dimensional position coordinates, flight speed, flight attitude and hovering state. Discretization uses equal time interval sampling with a sampling period of 0.1 seconds. The velocity and acceleration of the sampling points are checked to ensure that the dynamic constraints are met. If a constraint violation is found, a local correction is made by adjusting the position of the sampling point until all constraints are met.

[0202] The specific implementation method of step S09 first resolves the fire extinguishing measure vector into specific execution actions, including the working state of the high-voltage generator, the triggering sequence of the electromagnetic throwing controller and the flow control of the fine water mist pump. For the high-voltage generator, a pulse width modulation control method is adopted, the modulation frequency is 1 kHz, and the duty cycle range is 20% to 80%; for the electromagnetic throwing controller, a timing triggering method is adopted, the trigger pulse width is 10 milliseconds, and the trigger interval is determined according to the throwing rate; for the fine water mist pump, a variable frequency speed regulation method is adopted, and the frequency range is 20 Hz to 50 Hz.

[0203] The control strategy is designed according to the working characteristics of different fire extinguishing equipment. A state machine model is established for each device, including initialization, standby, working, fault and shutdown states. The state transition conditions are determined according to the equipment parameters and task requirements, and the control parameters in each state are obtained by table lookup. The cycle of the state machine is 10 milliseconds to ensure the real-time control. The finite state machine is used to implement the logical control of the equipment, and the state transition diagram is described using the Moore type state machine.

[0204] Finally, a fire extinguishing control instruction sequence is generated, which includes specific operations such as equipment startup, working parameter adjustment and equipment shutdown. The format of the control instruction includes: timestamp, equipment number, operation type, parameter value and checksum. The execution of the instruction adopts real-time monitoring and feedback control, and the control parameters are dynamically adjusted according to the execution effect. The monitoring parameters include equipment working status, output parameters and fault information, and the sampling period is 1 millisecond.

[0205] The specific implementation method of step S10 first establishes a unified time reference, unifying the timestamps of the flight control instruction sequence and the fire control instruction sequence into the same coordinate system. The time reference uses GPS timing, and the clock accuracy is better than 1 microsecond. The two instruction sequences are time-aligned, and the linear interpolation method is used to process data with different sampling periods. The time synchronization error needs to be controlled within 1 millisecond.

[0206] Then, according to the order and parallel relationship of task execution, the instruction sequence is organized into a directed acyclic graph structure. The nodes in the graph represent specific execution instructions, and the edges represent the dependencies between instructions. A topological sorting algorithm is used to determine the execution order of instructions to ensure that the dependencies are satisfied. For instructions that can be executed in parallel, a parallel scheduling strategy is used to improve execution efficiency. During the scheduling process, the mutual influence between devices needs to be considered to avoid conflicts.

[0207] Finally, the critical path algorithm is used to calculate the optimal execution order and generate the execution instruction queue. The calculation of the critical path takes into account the execution time and resource usage of the instructions, and is solved using the dynamic programming method. Each instruction in the execution instruction queue contains three elements: execution time, execution content, and execution conditions. The execution of instructions is event-driven. For conditional judgment and branch execution, predictive execution technology is used to improve response speed. The queue management adopts a double buffering mechanism to ensure the continuous execution of instructions.

[0208] The specific implementation method of step S11 adopts a hierarchical control architecture to realize the mission control of the autonomous flying fire extinguisher. The control system is divided into three levels: mission management layer, trajectory tracking layer and execution control layer. The mission management layer adopts a finite state machine to realize mission process control. The state machine includes five main states: mission initialization, route planning, fire extinguishing execution, status monitoring and mission completion. The triggering conditions of state transition include mission instructions, environmental changes and abnormal events. The control cycle of the mission management layer is 500 milliseconds, which is mainly responsible for parsing execution instructions and monitoring mission status.

[0209] The trajectory tracking layer uses a model predictive control algorithm to achieve accurate tracking of the planned trajectory, and the control period is set to 50 milliseconds. The prediction model takes into account the dynamic characteristics of the aircraft, and the state equation is The state vector x includes position, velocity and attitude, and the control input u includes thrust and torque. The prediction time domain is set to 20 steps, and the control time domain is set to 10 steps. The objective function includes trajectory tracking error, control input and state change, and adopts quadratic form. The constraints include state constraints and control constraints, and the soft constraint processing method is used. The position control accuracy of the controller is better than 0.5 meters, and the attitude control accuracy is better than 2 degrees.

[0210] The execution control layer uses a cascade PID control algorithm to achieve direct control of each actuator, including motor control, steering gear control, and fire extinguishing equipment control. The inner loop control cycle is 2 milliseconds, responsible for attitude angular velocity control; the outer loop control cycle is 10 milliseconds, responsible for attitude angle control. PID parameters are adjusted in real time through online identification methods, and the adaptive law adopts the MIT law. The execution layer also includes fault detection and processing functions. When an abnormal situation is detected, the corresponding emergency processing procedure is automatically started.

[0211] The specific implementation of step S12 adopts a data processing method of a sliding time window, with a window length of 10 seconds and a sliding step length of 1 second. The latest data of all sensors are collected in each update cycle, including thermal imaging data, visible light data, three-band flame data, temperature and humidity data, smoke data, and carbon monoxide data. Data preprocessing adopts low-pass filtering and outlier detection methods, the filter cutoff frequency is set to 10 Hz, and the 3σ criterion is used for outlier determination.

[0212] Feature extraction is performed on the preprocessed multi-source sensor data, and the wavelet transform method is used for time-frequency analysis. The db4 wavelet is selected as the basis function, and the number of decomposition layers is set to 4. The extracted features include energy features, statistical features, and morphological features. The energy features reflect the intensity distribution of the signal, the statistical features describe the statistical laws of the data, and the morphological features characterize the change characteristics of the signal. The results of feature extraction are used to update the fire development matrix.

[0213] Information fusion uses the evidence theory method to construct a basic probability distribution function. The credibility of each sensor data is calculated separately, and the credibility threshold is set to 0.8. The Dempster combination rule is used to synthesize multi-source information to obtain the fusion result. The positive development matrix and the negative development matrix of the fire are updated according to the fusion result. The update process adopts the exponential weighted average method, and the weight coefficient of the historical data is set to 0.7. The consistency check of the updated matrix is ​​carried out to ensure the reliability of the data.

[0214] The specific implementation method of step S13 first calculates the change in the fire development matrix, and uses the matrix difference method to calculate the change in fire intensity, spread direction and fire range. The calculation formula for the change is ΔF = F k -F k-1 , where k represents the current moment. Then set the replanning trigger conditions: the fire intensity changes by more than 30%, the spread direction changes by more than 45 degrees, or the fire area changes by more than 20%. When any of the trigger conditions is met, the replanning procedure is started.

[0215] The replanning procedure first evaluates the effectiveness of the current route, using the Monte Carlo simulation method. The number of simulations is set to 100, and random disturbances are added to each simulation, including wind speed disturbances, temperature disturbances, and position disturbances. The amplitude of the disturbance is determined based on the actual observed data. The mission completion rate, that is, the probability of successfully completing the firefighting mission, is calculated. If the mission completion rate is less than 85%, it is determined that replanning is required.

[0216] The replanning process uses a heuristic search algorithm guided by the principle of minimization adjustment. First, the effectiveness of each waypoint in the current route is evaluated, and the still valid parts are retained. For the invalid parts, the local path planning method is used for optimization. The optimization goals include minimizing the route length, maximizing the fire fighting efficiency, and ensuring the safety margin. Finally, a new execution instruction queue is generated, and the smooth switching between the new and old instruction queues is ensured. During the switching process, the trajectory smoothing algorithm is used to avoid mutations.

[0217] The specific implementation of step S14 is to first calculate the return path based on the current position and the platform position, and use the A-star algorithm to implement path planning, taking into account obstacle avoidance and energy constraints. The heuristic function is h(n)=w1d(n)+w2e(n), where d(n) is the distance term, e(n) is the energy consumption term, and the weight coefficient is determined by experiment. The path smoothing adopts the cubic spline interpolation method to ensure the continuity of curvature.

[0218] Select the appropriate navigation method according to the distance from the platform: when the distance is greater than 100 meters, satellite navigation is preferred, with a navigation accuracy better than 5 meters; when the distance is between 30 and 100 meters, laser navigation is used, with a navigation accuracy better than 0.5 meters; when the distance is less than 30 meters, Bluetooth beacon navigation is used, with a navigation accuracy better than 0.1 meters. The switching of different navigation methods adopts a smooth transition strategy, and a 10-meter overlap interval is set. The fusion of navigation data adopts an adaptive Kalman filter algorithm, and the filter gain is adjusted in real time according to the measurement noise.

[0219] The flight control during the return process uses an adaptive control algorithm to dynamically adjust the flight parameters according to factors such as battery power and wind speed. When approaching the platform, it enters the precise landing stage and adopts the visual servo control method. Landing control with centimeter-level accuracy is achieved by identifying the characteristic markers on the platform. Feature extraction uses a combination of Hough transform and edge detection. The landing speed is controlled below 0.5 meters per second, and the descent speed at contact does not exceed 0.2 meters per second. After completing the landing, the shutdown procedure is automatically executed, including equipment status check, data saving and system shutdown. The equipment status check uses a self-diagnosis program, and the data is saved using a circular cache mechanism.

[0220] In order to better understand and implement the present invention, the following provides Example 2 of a specific application scenario of the present invention: The research team first configured the system of the autonomous flying fire extinguisher, as shown in Table 1:

[0221] Table 1 Autonomous flying fire extinguisher system configuration parameters

[0222] System Module Parameters Parameter Value Thermal imaging sensor Resolution 640×480 pixels Thermal imaging sensor Temperature detection range -40℃～1500℃ Visible light sensor Resolution 1920×1080 pixels Visible light sensor Frame rate 30fps Three-band flame sensor Sampling frequency 100Hz Temperature and humidity sensor Temperature range -50℃～150℃ Temperature and humidity sensor Humidity range 0~100%RH Smoke Sensor Detection range 0~1000ppm Carbon Monoxide Sensor Detection range 0~1000ppm High voltage generator Operating voltage 12000V Electromagnetic throwing controller Response time 20ms Water mist pump Maximum flow 2L / min

[0223] In a field test, the system extinguished a mountain forest fire with an area of ​​about 800 square meters. The system first acquired fire scene information through multi-sensor data collection, and constructed a fire positive development matrix after data fusion processing. According to the data collected by the sensors, as shown in Table 2:

[0224] Table 2 Fire scene sensor data collection results

[0225] Measurement parameters Minimum Maximum average value Temperature(℃) 125 850 420 Infrared intensity 85 245 168 Carbon monoxide concentration (ppm) 180 680 415 Flame height(m) 1.2 4.5 2.8 Spread speed (m / s) 0.08 0.35 0.22 Wind speed (m / s) 2.5 6.8 4.2

[0226] Figure 2 The chart shows the changing trend of fire scene temperature and carbon monoxide concentration over time. The horizontal axis is time (unit: seconds), and the vertical axis represents both temperature (degrees Celsius) and carbon monoxide concentration (ppm). The chart reflects the dynamic changing characteristics of key parameters during the fire extinguishing process. Based on the collected data, the fire forward development matrix formula is used to calculate:

[0227] f ij =0.4v ij +0.3d ij +0.3s ij +0.05;

[0228] The fire intensity value s ij The calculation uses:

[0229] s ij =0.4T ij +0.35I ij +0.25C ij +0.03;

[0230] Based on the positive fire development matrix, the system calculates the negative fire development matrix:

[0231] b ij =0.35r ij +0.35c ij +0.3e ij +0.04;

[0232] The fire extinguishing effect evaluation value adopts:

[0233]

[0234] The system calculates the fire extinguishing measures vector based on the fire area and fire intensity:

[0235] M = [12,8,180] T ;

[0236] It means that 12 fire extinguisher bombs and 8 fire extinguisher bottles are needed, and the fine water mist spraying time is 180 seconds.

[0237] When planning the route, the system first divides the fire scene into grids and calculates the fire intensity weight value of each grid node:

[0238]

[0239] The route planning adopts the deep reinforcement learning algorithm, and the planning results are shown in Table 3:

[0240] Table 3 Route planning result parameter table

[0241] Waypoint number X coordinate (m) Y coordinate (m) Height(m) Hover duration(s) 1 0 0 15 0 2 12.5 8.6 12 25 3 25.8 15.4 10 30 4 38.2 22.5 8 35 5 42.6 35.8 12 28 6 35.4 42.5 15 32 7 22.8 38.6 10 30 8 15.6 28.4 8 25 9 8.2 15.6 12 20 10 0 0 15 0

[0242] Route cost calculation uses:

[0243] C path =0.2L path +0.2R risk +0.2E eff +0.15P energy +0.15D diff +0.1T task +0.02;

[0244] The calculation results of resource utilization efficiency during system execution are shown in Table 4:

[0245] Table 4 Statistics of fire extinguishing resource utilization efficiency

[0246] Fire Fighting Measures Delivery volume <![CDATA[Coverage area (m 2 )]]> Fire extinguishing effect rating Resource Utilization Fire extinguisher 12 320 0.85 0.78 Fire extinguisher 8 280 0.82 0.75 Water mist 360L 200 0.88 0.82

[0247] Figure 3The utilization rate and coverage area of ​​three types of fire-fighting resources (fire-fighting bombs, fire-fighting bottles, and fine water mist) are compared. The left vertical axis represents the resource utilization rate, and the right vertical axis represents the coverage area (square meters). The figure shows the comparison of the effects of different fire-fighting measures. The entire fire-fighting process lasted about 15 minutes, and the fire was finally extinguished successfully. The system replanned the route twice during the execution process. The replanning trigger conditions were that the fire intensity changed by more than 30% and the spread direction changed by more than 45 degrees. During each replanning process, the Monte Carlo simulation results of the task completion rate remained above 90%. Traditional forest fire fighting mainly relies on manual on-site command and manual fire-fighting operations, and there are the following problems: 1) The perception of the fire situation is inaccurate and relies on manual experience judgment. 2) The efficiency of fire-fighting resource allocation is low, which often results in waste of resources. 3) The fire-fighting path planning lacks scientific basis and is difficult to optimize. 4) The fire-fighting effect evaluation method is single and lacks quantitative indicators. 5) The safety risk of rescue personnel is high, especially under complex terrain conditions.

[0248] The autonomous flying fire extinguishing system used in the present invention has the following advantages over traditional methods: 1) Accurate perception of the fire situation is achieved through multi-sensor data fusion, the accuracy of fire scene temperature measurement is improved by 85%, and the accuracy of fire development prediction is improved by 75%. 2) The intelligent algorithm is used to achieve the optimal configuration of fire extinguishing resources, and the resource utilization rate is improved by 65%, avoiding resource waste.

[0249] 3) The path planning method based on deep reinforcement learning increases firefighting efficiency by 80% and shortens task completion time by 50%.

[0250] 4) A complete fire extinguishing effect evaluation system has been established, which includes multiple quantitative indicators and improves the objectivity of the evaluation results by 90%.

[0251] 5) It realizes unmanned firefighting operations, significantly reducing the safety risks of rescue workers. 6) The system has adaptive adjustment capabilities and can optimize firefighting strategies in real time according to changes in fire conditions, with an adjustment success rate of 95%.

[0252] It should be noted that the variables involved in the present invention are explained in detail as shown in Tables 5 and 6 below.

[0253] Table 5 Variable explanation table (Part I)

[0254]

[0255] Table 6 Variable explanation table (Part II)

[0256]

[0257]

[0258] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent fire extinguishing strategy control method for an autonomous flying fire extinguisher, characterized in that: include: The sensor data of the autonomous flying fire extinguisher is obtained, a positive fire development matrix is ​​constructed according to the sensor data, a negative fire development matrix is ​​calculated according to the positive fire development matrix, a fire extinguishing strategy equation group is established based on the positive fire development matrix and the negative fire development matrix, a fire extinguishing measure vector is calculated using the fire extinguishing strategy equation group, a fire scene space division grid is established according to the positive fire development matrix and a fire intensity weight value is assigned, a route is planned using a pre-trained fire extinguishing path planning model, a flight trajectory curve is calculated according to the route planning result and a flight control instruction sequence is generated, a fire extinguishing control instruction sequence is generated according to the fire extinguishing measure vector, the flight control instruction sequence is time-aligned with the fire extinguishing control instruction sequence to generate an execution instruction queue, the autonomous flying fire extinguisher is controlled to perform a fire extinguishing task, the positive fire development matrix and the negative fire development matrix are updated in real time through the sensor data, whether to re-plan the route is determined according to the updated matrix, and the fire extinguishing task is completed and the platform is returned.

2. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 1 is characterized in that: The sensor data includes thermal imaging sensor data, visible light sensor data, three-band flame sensor data, temperature and humidity sensor data, smoke sensor data, and carbon monoxide sensor data; the thermal imaging sensor data is collected at a sampling frequency of 60 Hz, the visible light sensor data is collected at a frame rate of 30 frames per second, and the three-band flame sensor data is collected at a sampling frequency of 100 Hz.

3. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 2 is characterized in that: The positive fire development matrix represents the fire spread speed, fire expansion direction, and fire intensity; the negative fire development matrix represents the fire decay speed, fire contraction direction, and fire reduction degree after the fire extinguishing agent is released.

4. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 3 is characterized by: The pre-trained fire-fighting path planning model adopts a convolutional neural network structure with an embedded mathematical calculation module. The mathematical calculation module includes a fire spread prediction equation, a resource consumption evaluation equation, and a route cost calculation equation, which is used to integrate the fire development trend and the fire-fighting resource constraints to perform route optimization calculations.

5. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 4 is characterized in that: The input of the fire spread prediction equation includes fire scene temperature distribution data, environmental wind speed data, combustible material distribution data, and terrain height data; the input of the resource consumption assessment equation includes fire scene temperature data, fire scene area data, fire extinguishing agent storage data, fire extinguisher remaining power data, and target fire point distance data; the input of the route cost calculation equation includes route length data, fire scene danger data, fire extinguishing efficiency prediction data, energy consumption data, passage difficulty data, and task completion time data.

6. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 5 is characterized by: The basic size of the fire scene space division grid is 5 meters × 5 meters, and a grid size of 1 meter × 1 meter is used in areas where the fire is strong or changes greatly; the fire weight value is calculated using the hierarchical analysis method, with a weight coefficient of 0.3 for the temperature value, 0.25 for the fire intensity, 0.2 for the spread speed, 0.15 for the terrain features, and 0.1 for the weight coefficient of important surrounding targets.

7. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 6 is characterized by: The flight trajectory curve is calculated using a Bezier curve interpolation algorithm. The maximum speed of the autonomous flying fire extinguisher is set to 15 meters per second, the maximum acceleration is set to 2 meters per square second, and the maximum turning angle is set to 45 degrees. The flight control instructions include three-dimensional position coordinates, flight speed, flight attitude and hovering state.

8. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 7 is characterized in that: The fire extinguishing control instruction sequence includes the working state of the high-voltage generator, the triggering timing of the electromagnetic throwing controller and the flow control of the fine water mist pump; the control of the high-voltage generator adopts pulse width modulation, the control of the electromagnetic throwing controller adopts timing triggering, and the control of the fine water mist pump adopts variable frequency speed regulation.

9. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 8 is characterized by: When any of the following conditions is met: the fire intensity changes by more than 30%, the spread direction changes by more than 45 degrees, or the fire range changes by more than 20%, the re-planning procedure is initiated; the re-planning procedure uses the Monte Carlo simulation method to simulate the task execution process 100 times, and the route is re-planned when the task completion rate is lower than 85%.

10. The intelligent fire extinguishing strategy control method of the autonomous flying fire extinguisher according to claim 9 is characterized in that: When the autonomous flying fire extinguisher returns to the platform, when the distance from the platform is greater than 100 meters, satellite navigation is used, and the navigation accuracy is better than 5 meters; when the distance is between 30 and 100 meters, laser navigation is used, and the navigation accuracy is better than 0.5 meters; when the distance is less than 30 meters, Bluetooth beacon navigation is used, and the navigation accuracy is better than 0.1 meters.

Images