Live wire situation awareness and autonomous breakthrough path generation method for forest fire fighting truck

Abstract

The invention provides a live wire situation awareness and autonomous breakthrough path generation method for a forest fire fighting truck, and relates to the technical field of fire fighting, and the method comprises the steps: segmenting a fire forest region into positioning spaces based on a satellite remote sensing image, extracting the maximum gradient, thermal radiation and other parameters of each space, screening feasible breakthrough path spaces around the fire fighting truck, and carrying out the positioning of the positioning spaces; and repeating the operation to a safe area to form an initial path, and predicting thermal radiation intensity to optimize and generate a final path. According to the method, the situation is sensed in real time through satellite remote sensing, paths are screened in multiple dimensions, fire behavior and terrain dynamic adjustment and thermal radiation prediction optimization are combined, situation sensing accuracy and path dynamic adaptability are achieved, multiple risks are avoided, safe, efficient and autonomous breakthrough of the fire fighting truck is guaranteed, and the rescue survival guarantee capacity is improved.

Description

Technical Field

[0001] This invention relates to the field of fire protection technology, and more specifically, to a method for fire situation awareness and autonomous escape path generation for forest fire trucks. Background Technology

[0002] With global warming, forest fires are becoming more frequent and are characterized by rapid spread, wide impact, and complex on-site environments. Within fire zones, dense smoke obscures the landscape, and rugged terrain makes it difficult to accurately capture real-time fire dynamics due to limited coverage of traditional ground monitoring stations and blind spots in manual observation. Forest fire trucks, as core equipment for fire rescue, are vulnerable to being surrounded by fire if they cannot promptly grasp the fire situation and plan safe escape routes, threatening the safety of vehicles and personnel. Therefore, there is an urgent need to build an efficient autonomous escape route generation system based on high-precision, wide-coverage monitoring technology.

[0003] Traditional methods for planning escape routes for forest fire trucks often rely on pre-set terrain maps or simple manual judgment, lacking the ability to perceive the real-time situation of the fire line and dynamically extract key parameters such as the heat radiation intensity and obstacle distribution of the fire area. Route selection only considers the terrain slope and does not take into account multiple dimensions such as the fire truck's obstacle clearing capabilities and the direction of fire spread. Furthermore, it does not consider the trend of fire changes over time and does not predict the future heat radiation intensity of the route space, making the planned route prone to becoming a dangerous area due to the spread of fire, and significantly reducing the safety and efficiency of the escape.

[0004] Therefore, it is necessary to design a fire situation awareness and autonomous escape path generation method for forest fire trucks to solve the problems of inaccurate and real-time fire situation awareness, single-dimensional escape path selection, lack of dynamic path adjustment based on fire situation and terrain, and lack of thermal radiation prediction optimization in existing technologies. These problems result in low safety factor and poor efficiency of forest fire trucks in escaping, making it difficult to cope with complex fire environments. Summary of the Invention

[0005] In view of this, the present invention proposes a method for fire situation awareness and autonomous escape path generation for forest fire trucks, aiming to solve the problems of inaccurate and real-time fire situation awareness, single dimension of escape path selection, lack of dynamic combination of fire situation and terrain to adjust the path, and lack of thermal radiation prediction optimization in the existing technology, which leads to low safety factor and poor efficiency of forest fire trucks in escaping, making it difficult to cope with complex fire environments.

[0006] In one aspect, this invention proposes a method for fire situation awareness and autonomous breakout path generation for forest fire trucks, comprising: Based on satellite remote sensing imagery, the forest area where the fire occurred is divided into several positioning spaces. The positioning spaces representing the current location of the fire trucks are obtained in real time using satellite remote sensing technology. At the same time, the maximum slope, initial thermal radiation intensity, and obstacle resistance value in each positioning space are extracted using satellite remote sensing technology. Based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle resistance value, an escape path space is selected within the positioning space surrounding the fire truck positioning space. The current escape route space is used as the fire truck positioning space, and the operation of filtering the escape route space is repeated until the filtered escape route space is in a safe area. The fire truck positioning space and the escape route space constitute the initial escape route. Based on the average speed of the fire truck, the travel time required for the fire truck to travel from its initial position to each of the escape path spaces is calculated, and the predicted thermal radiation intensity of the escape path space after the travel time is predicted is calculated. Based on the predicted thermal radiation intensity, the escape path spaces are re-selected, and the final escape path is generated.

[0007] Furthermore, when selecting an escape route space within the positioning space surrounding the fire truck positioning space based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle resistance value, the process includes: A first positioning space around the fire truck positioning space is selected, and the first positioning spaces form a first positioning space set. Within the first set of positioning spaces, a second positioning space with a maximum slope less than the fire truck climbing threshold is selected, and the second positioning spaces form a second set of positioning spaces. Within the second set of positioning spaces, a third set of positioning spaces is selected where the obstacle value is less than or equal to the threshold of the fire truck's obstacle removal capability.

[0008] Furthermore, when selecting an escape route space within the positioning space surrounding the fire truck positioning space based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle resistance value, the method further includes: Within the third set of positioning spaces, a fourth set of positioning spaces is selected where the initial thermal radiation intensity is less than the radiation intensity threshold. The fourth set of positioning spaces constitutes the fourth set of positioning spaces. Within the fourth set of positioning spaces, a fifth positioning space with the smallest initial radiation intensity is selected, and the fifth positioning space constitutes a fifth set of positioning spaces. The direction of fire spread is obtained based on satellite remote sensing technology, and the escape path space is selected from the fifth positioning space set based on the direction of fire spread.

[0009] Furthermore, when selecting the escape route space from the fifth location space set based on the direction of fire spread, the process includes: When the fifth positioning space exists in the reverse area, vertical area, and forward area of ​​the fire spread direction, the fifth positioning space located in the reverse area of ​​the fire spread direction and having the shortest straight-line distance from the fire truck positioning space is selected as the escape path space. When the fifth positioning space exists in both the vertical and horizontal regions along the fire spread direction, the fifth positioning space located in the vertical region along the fire spread direction and having the shortest straight-line distance from the fire truck positioning space is selected as the escape route space.

[0010] Furthermore, when selecting the escape route space from the fifth location space set based on the direction of fire spread, the method also includes: When the fifth positioning space exists in the forward region of the fire spread direction, the fifth positioning space with the largest angle to the fire spread direction is selected as the breakout path space.

[0011] Furthermore, when obtaining the direction of fire spread based on satellite remote sensing technology, this includes: The fire line boundary is identified and extracted from the images of the fire area over a predetermined number of consecutive time periods using satellite remote sensing technology. Several sets of feature points are selected on the fire line boundary between adjacent time periods, and the displacement vector of each set of feature points is calculated. Cluster analysis is performed on all displacement vectors, and the direction of the vector with the highest frequency is taken as the dominant displacement direction; Based on the maximum slope of the positioning space where the fire line boundary is located, the dominant displacement direction is adjusted, and the adjusted dominant displacement direction is the direction of fire spread.

[0012] Furthermore, when adjusting the dominant displacement direction based on the maximum slope of the positioning space where the fire line boundary is located, the following steps are included: When the terrain of the positioning space where the fire line boundary is located is uphill, the dominant displacement direction is adjusted upwards by a first angle. When the terrain of the positioning space where the fire line boundary is located is downhill, the dominant displacement direction is adjusted to a second angle downwards towards the slope. When the terrain of the positioning space where the fire line boundary is located is a plain, the adjustment amount of the dominant displacement direction is 0. The first angle is equal to the second angle, which is equal to the maximum slope multiplied by the adjustment ratio.

[0013] Furthermore, when the fire truck positioning space and the escape path space constitute the initial escape path, it includes: When the escape path space is selected, the escape path space is sequentially connected with the historical fire truck positioning space and the historical escape path space in chronological order to form a continuous path node sequence. When the last path node is a safe area, the path node sequence is the initial escape path.

[0014] Furthermore, the predicted thermal radiation intensity is equal to the initial thermal radiation intensity multiplied by the cumulative change factor of thermal radiation intensity; The cumulative change factor of thermal radiation intensity is equal to the travel time multiplied by the thermal radiation intensity change coefficient plus 1.

[0015] Furthermore, when re-filtering the escape path space based on the predicted thermal radiation intensity and generating the final escape path, the process includes: When the predicted thermal radiation intensity is less than the radiation intensity threshold, the current escape path space remains unchanged. When the predicted thermal radiation intensity is greater than or equal to the radiation intensity threshold, the fourth positioning space is re-selected, and the breakout path space is re-selected based on the re-selected fourth positioning space. The breakout path space corresponding to the initial breakout path is replaced with the re-selected breakout path space to obtain the final breakout path.

[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: The method for fire situation perception and autonomous escape path generation for forest fire trucks uses satellite remote sensing technology to perceive the fire situation in real time, divides the forest area into positioning spaces and extracts key parameters such as maximum slope and thermal radiation, and combines fire truck performance thresholds to screen escape path spaces from multiple dimensions. At the same time, it considers the direction of fire spread and the influence of terrain to make dynamic adjustments, and optimizes the path by predicting the thermal radiation intensity. This achieves the accuracy of fire situation perception, the dynamic adaptability of escape path generation and autonomous decision-making, effectively avoids complex terrain, obstacles and fire risks, ensures the safe and efficient autonomous escape of forest fire trucks in fire environments, and improves the survival protection capabilities of vehicles and personnel in fire rescue. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1This is a flowchart of a method for fire situation awareness and autonomous breakout path generation for forest fire trucks, provided in an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the invention to those skilled in the art. It should be noted that, without conflict, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] Reference Figure 1 As shown in some embodiments of this application, a method for fire situation awareness and autonomous breakout path generation for forest fire trucks includes: Based on satellite remote sensing imagery, the forest area where the fire occurred is divided into several positioning spaces. The positioning spaces representing the current location of the fire trucks are obtained in real time using satellite remote sensing technology. At the same time, the maximum slope, initial thermal radiation intensity, and obstacle resistance value in each positioning space are extracted using satellite remote sensing technology. Based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle resistance value, an escape path space is selected within the positioning space surrounding the fire truck positioning space. The current escape route space is used as the fire truck positioning space, and the operation of filtering the escape route space is repeated until the filtered escape route space is in a safe area. The fire truck positioning space and the escape route space constitute the initial escape route. Based on the average speed of the fire truck, the travel time required for the fire truck to travel from its initial position to each of the escape path spaces is calculated, and the predicted thermal radiation intensity of the escape path space after the travel time is predicted is calculated. Based on the predicted thermal radiation intensity, the escape path spaces are re-selected, and the final escape path is generated.

[0020] Specifically, based on the spatial resolution of satellite remote sensing imagery and the continuity of land features in the fire area, the forest area where the fire occurred is divided into several clearly defined and non-overlapping positioning spaces according to a unified spatial scale, ensuring that the geographical scope of each positioning space matches the effective observation unit of the satellite remote sensing imagery. A digital elevation model (DEM) of the fire area is obtained by inverting satellite remote sensing imagery, and the maximum slope is calculated by the slope extreme value in each positioning space. The thermal radiation energy distribution in the positioning space is inverted using satellite remote sensing thermal infrared band data, and the initial thermal radiation intensity is extracted. The type and distribution characteristics of obstacles in the positioning space are identified by satellite remote sensing multispectral bands, and then the obstacle resistance value is quantified. There are no cliffs or steep walls in the divided positioning spaces, and each positioning space has a path that fire trucks can travel through.

[0021] Understandably, by integrating satellite remote sensing technology to achieve real-time fire situation awareness, spatial segmentation, key parameter extraction, and dynamic path generation and optimization, a complete process from situation awareness to autonomous breakout path generation has been constructed. This provides a basic framework for subsequent steps, ensuring that the method can systematically solve the problem of forest fire trucks breaking out in complex fire environments, achieving full-chain optimization from the initial path to the final path, and guaranteeing the autonomy and safety of the breakout process.

[0022] In some embodiments of this application, when selecting an escape route space within the positioning space surrounding the fire truck positioning space based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle resistance value, the process includes: A first positioning space around the fire truck positioning space is selected, and the first positioning spaces form a first positioning space set. Within the first set of positioning spaces, a second positioning space with a maximum slope less than the fire truck climbing threshold is selected, and the second positioning spaces form a second set of positioning spaces. Within the second set of positioning spaces, a third set of positioning spaces is selected where the obstacle value is less than or equal to the threshold of the fire truck's obstacle removal capability.

[0023] Specifically, the maximum gradient refers to the largest gradient value among all geographical points within a single location space. It quantifies the steepness of the terrain in that location space and reflects the difficulty of climbing slopes faced by fire trucks when driving within it. The fire truck climbing threshold is determined based on the performance of the fire truck's power system, transmission specifications, tire grip parameters, and vehicle stability design standards. It represents the maximum gradient limit that allows the fire truck to drive safely and stably and is a key indicator for determining whether a location space has the basic conditions for passage. The obstacle resistance value comprehensively considers the types of obstacles within the location space (such as trees, rocks, gullies, fallen logs, etc.). Dimensions (such as tree diameter at breast height, rock diameter, gully width and depth), distribution density, and distribution pattern (concentrated or dispersed distribution) are parameters that quantify the degree to which obstacles hinder the passage of fire trucks. Higher values ​​indicate greater obstruction. The fire truck obstacle clearing capability threshold is the maximum upper limit of the degree of obstruction that a fire truck can effectively handle, determined based on the performance parameters of the obstacle clearing devices (such as hydraulic obstacle clearing shovels and obstacle breaking hammers) equipped on the fire truck, including structural strength, operating power, and operating range, combined with actual obstacle clearing operation safety standards. It is used to determine whether obstacles in the location space can be cleared by the fire truck's own obstacle clearing capability.

[0024] Specifically, the maximum slope is determined by acquiring high-resolution digital elevation model (DEM) data covering the fire area using satellite remote sensing technology. The slope is calculated point-by-point for each DEM grid cell corresponding to a single location space (slope = arctan(vertical height difference / horizontal distance)). The maximum value is then extracted from all calculation results. The fire truck climbing threshold is determined by consulting the fire truck's manufacturer's technical specifications to obtain design climbing parameters. Combined with on-site climbing test data for different terrains (soil, gravel, grassland, etc.), outliers in extreme environments are removed, and the maximum slope value that has been consistently passed through multiple tests is used. The obstacle resistance value is determined using satellite remote sensing multispectral imagery. The ground feature recognition algorithm distinguishes the types of obstacles in the location space, extracts the size parameters of obstacles using image pixel resolution, calculates the distribution density of obstacles through a density statistical model, and then quantifies the results using a weighted summation formula based on the preset obstacle weight coefficients for different types, sizes, and densities of obstacles. The fire truck obstacle clearing capability threshold is determined by conducting special performance tests on the obstacle clearing devices equipped on fire trucks, recording their handling effects on different types and sizes of obstacles (such as whether they can be cleared, clearing efficiency, and road traffic conditions after the operation), setting redundancy coefficients in conjunction with fire operation safety regulations, and finally determining the upper limit of the obstacle value that can be effectively handled.

[0025] Specifically, the maximum slope is determined by acquiring high-resolution digital elevation model (DEM) data covering the fire area using satellite remote sensing technology. The slope is calculated point-by-point for each DEM grid cell corresponding to a single location space (slope = arctan(vertical height difference / horizontal distance)). The maximum value is then extracted from all calculation results. The fire truck climbing threshold is determined by consulting the fire truck's manufacturer's technical specifications to obtain design climbing parameters. Combined with on-site climbing test data for different terrains (soil, gravel, grassland, etc.), outliers in extreme environments are removed, and the maximum slope value that has been consistently passed through multiple tests is used. The obstacle resistance value is determined using satellite remote sensing multispectral imagery. The ground feature recognition algorithm distinguishes the types of obstacles in the location space, extracts the size parameters of obstacles using image pixel resolution, calculates the distribution density of obstacles through a density statistical model, and then quantifies the results using a weighted summation formula based on the preset obstacle weight coefficients for different types, sizes, and densities of obstacles. The fire truck obstacle clearing capability threshold is determined by conducting special performance tests on the obstacle clearing devices equipped on fire trucks, recording their handling effects on different types and sizes of obstacles (such as whether they can be cleared, clearing efficiency, and road traffic conditions after the operation), setting redundancy coefficients in conjunction with fire operation safety regulations, and finally determining the upper limit of the obstacle value that can be effectively handled.

[0026] Understandably, by screening the first to third positioning spaces in stages, the focus is first placed on the area around the fire truck's positioning space, and then progressive screening is carried out based on the maximum slope and the fire truck's climbing threshold, the obstacle resistance value and the fire truck's obstacle clearing capability threshold. This effectively narrows down the range of path selection, ensuring that the initially screened spaces are all within the fire truck's passage capability. Impassable areas are excluded from the perspective of terrain and obstacles, laying the foundation for more accurate path selection in the future and improving the basic feasibility of the path.

[0027] In some embodiments of this application, when selecting an escape route space within the positioning space surrounding the fire truck positioning space based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle resistance value, the method further includes: Within the third set of positioning spaces, a fourth set of positioning spaces is selected where the initial thermal radiation intensity is less than the radiation intensity threshold. The fourth set of positioning spaces constitutes the fourth set of positioning spaces. Within the fourth set of positioning spaces, a fifth positioning space with the smallest initial radiation intensity is selected, and the fifth positioning space constitutes a fifth set of positioning spaces. The direction of fire spread is obtained based on satellite remote sensing technology, and the escape path space is selected from the fifth positioning space set based on the direction of fire spread.

[0028] Specifically, the initial thermal radiation intensity refers to the thermal radiation energy density per unit area monitored by satellite remote sensing in the early stages of a fire in a single location space (before it is significantly affected by the spread of the fire). It can intuitively reflect the degree of thermal impact of the fire on the current location space. Its value is directly related to the type of combustibles in the location space, the degree of combustion, the stage of fire development, and the heat accumulation in the space. The radiation intensity threshold is a safe upper limit of thermal radiation intensity determined comprehensively based on the heat resistance limit of the fire truck body material, the operating temperature tolerance range of the on-board equipment, the thermal safety tolerance standards of firefighters, and the safety specifications of the forest fire rescue industry. It is the core thermal environment indicator for judging whether the location space meets the safe conditions for fire truck passage. Exceeding this threshold may cause thermal damage to vehicles, equipment, and personnel.

[0029] Specifically, the initial thermal radiation intensity is obtained by acquiring raw thermal radiation data of the positioning space through the thermal infrared band sensor of the satellite remote sensing system. This data is then combined with an atmospheric correction model to eliminate interference factors such as atmospheric attenuation and cloud cover. The thermal radiation energy density distribution within the positioning space is then calculated by inverting the radiative transfer equation, and finally, the average thermal radiation intensity within this space is extracted as the initial thermal radiation intensity. The radiation intensity threshold is initially determined by consulting national standards for forest fire rescue safety, factory heat resistance performance parameters of fire trucks and on-board equipment, and human thermophysiological tolerance test data. The threshold range is then initially determined by simulating vehicle operation tests and personnel exposure experiments under different thermal radiation environments, combined with thermal radiation safety data from actual fire rescue cases, to verify and calibrate the data. Finally, a fixed threshold or a dynamically adjusted threshold range is determined.

[0030] Specifically, the initial thermal radiation intensity is measured in kW / m². 2 The numerical range is divided into 0-10kW / m 2 (Low strength), 10-30kW / m 2 (Medium strength), 30kW / m 2 The radiation intensity threshold for the above (high intensity) is set at 30 kW / m². 2 In the third set of selected locations, the initial thermal radiation intensity of each location was compared with 30kW / m². 2 Based on the magnitude relationship, initial thermal radiation intensity <30kW / m 2 The positioning spaces are then combined to form a fourth positioning space set. The initial thermal radiation intensities of all positioning spaces within this fourth positioning space set are then numerically sorted, and the space with the lowest initial thermal radiation intensity (assumed to be 0-5 kW / m²) is extracted. 2The fifth positioning space set is composed of several positioning spaces in the interval. The direction of fire spread (such as northwest) is obtained by satellite remote sensing technology. Then, based on the direction, the reverse region (southeast), vertical region (northeast, southwest), and forward region (northwest) of each positioning space in the fifth positioning space set are determined. Finally, the escape path space is selected according to the rule of "prioritizing the reverse region with the shortest distance, secondly selecting the vertical region with the shortest distance, and selecting the region with the largest angle when only forward region is selected".

[0031] Understandably, based on the third positioning space, the initial thermal radiation intensity and radiation intensity threshold are further introduced to screen the fourth positioning space, and then the principle of minimizing the initial radiation intensity is combined to screen the fifth positioning space. Finally, the escape route space is determined based on the direction of fire spread. Through multi-dimensional superposition screening, the space with low thermal radiation and in line with the direction of fire avoidance is selected first, which significantly improves the safety factor of the path and makes the path selection more in line with the thermal safety requirements in the fire environment.

[0032] In some embodiments of this application, when selecting the escape path space from the fifth location space set based on the direction of fire spread, the process includes: When the fifth positioning space exists in the reverse area, vertical area, and forward area of ​​the fire spread direction, the fifth positioning space located in the reverse area of ​​the fire spread direction and having the shortest straight-line distance from the fire truck positioning space is selected as the escape path space. When the fifth positioning space exists in both the vertical and horizontal regions along the fire spread direction, the fifth positioning space located in the vertical region along the fire spread direction and having the shortest straight-line distance from the fire truck positioning space is selected as the escape route space.

[0033] Specifically, the reverse zone refers to the positioning space located at an angle of 170°-190° to the direction of fire spread. This zone is completely opposite to the mainstream direction of fire spread, and the fire spreads to this zone at the slowest speed and with the lowest probability. The positioning space is also affected by the fire's heat radiation for the longest time, making it the safest zone. The vertical zone refers to the positioning space located at an angle of 80°-100° to the direction of fire spread. This zone is neither along nor against the direction of fire spread. The fire spreads to this zone at a moderate speed, and the degree of fire impact on the positioning space is between that of the reverse and forward zones, placing its safety factor at a moderate level. The forward zone refers to the positioning space located at an angle of 0°-80° to the direction of fire spread. This zone is consistent with or close to the mainstream direction of fire spread. The fire spreads to this zone at the fastest speed and is most easily covered by the fire line. The positioning space faces the highest risk of fire heat radiation and flame attack, resulting in the lowest safety factor.

[0034] Understandably, by prioritizing areas opposite to the fire's path and then selecting vertical areas, the high-risk directions for fire spread are actively avoided, making path selection more targeted and reducing the probability of being attacked by the fire.

[0035] In some embodiments of this application, when selecting the escape path space from the fifth location space set based on the fire spread direction, the method further includes: When the fifth positioning space exists in the forward region of the fire spread direction, the fifth positioning space with the largest angle to the fire spread direction is selected as the breakout path space.

[0036] Understandably, choosing the fifth positioning space with the largest angle to the direction of fire spread covers all possible regional distribution scenarios. Even in the extreme case where only forward areas are available, the direct impact of the fire can be reduced by maximizing the angle, avoiding blind spots in path selection caused by incomplete scenario coverage, and enhancing the comprehensiveness and adaptability of the method.

[0037] In some embodiments of this application, when obtaining the direction of fire spread based on satellite remote sensing technology, the following are included: The fire line boundary is identified and extracted from the images of the fire area over a predetermined number of consecutive time periods using satellite remote sensing technology. Several sets of feature points are selected on the fire line boundary between adjacent time periods, and the displacement vector of each set of feature points is calculated. Cluster analysis is performed on all displacement vectors, and the direction of the vector with the highest frequency is taken as the dominant displacement direction; Based on the maximum slope of the positioning space where the fire line boundary is located, the dominant displacement direction is adjusted, and the adjusted dominant displacement direction is the direction of fire spread.

[0038] Specifically, the preset quantity refers to the number of time periods for which continuous remote sensing images need to be acquired in advance to analyze the fire spread trend. Its value is determined according to the fire type (such as surface fire, crown fire), satellite remote sensing temporal resolution, and fire change rate to ensure effective capture of the dynamic change pattern of the fire line. Feature points refer to points with significant recognizability and stability selected on the fire line boundary. They are usually inflection points, endpoints, or points with large curvature changes on the boundary, used to accurately track the position movement of the fire line boundary at different time periods. Displacement vector refers to the change in position of the same feature point in remote sensing images of two adjacent time periods. It includes two elements: the direction of movement and the distance of movement, which intuitively reflects the spread trend of the fire line segment represented by the feature point. The fire line boundary refers to the dividing line between the flame front and the unburned area in a forest fire. It is the boundary outline between the burning and unburned areas of the fire, and its shape and position change in real time with the spread of the fire.

[0039] Specifically, the preset number is determined by analyzing the shortest continuous observation period required for significant changes in the fire line in historical fire cases, combined with the revisit cycle of the satellite used (e.g., 30 minutes / time) and the current fire spread speed (e.g., slow ≤5m / min, fast ≥15m / min), and calculated accordingly. For example, 5 to 8 time periods are set for rapidly spreading fires. Feature points are obtained by automatically identifying and filtering points with response values ​​higher than the preset threshold on the extracted fire line boundary vector data using corner detection algorithms (e.g., Harris algorithm), and then manually verifying and removing false detection points. Displacement vectors are obtained by spatially registering remote sensing images of adjacent time periods, determining the pixel coordinates of the same feature point in the preceding and following time periods, calculating the coordinate difference to obtain the displacement components in the x and y directions, and then obtaining the direction and distance through vector synthesis. The fire line boundary is obtained by threshold segmentation of satellite remote sensing thermal infrared band data (e.g., setting 300K as the flame temperature threshold), combined with the vegetation burning index (e.g., NBR) in the near-infrared band, using edge detection algorithms (e.g., Canny algorithm) to extract the boundary line between the flame and the unburned area, and then removing noise through morphological processing.

[0040] Specifically, the preset number is 5 time periods (each time period is 30 minutes apart). Remote sensing images of the fire area are acquired through satellite remote sensing for these 5 time periods. The Canny algorithm is used to extract the fire line boundary (e.g., irregular curve shape) from each image. On the fire line boundary of adjacent time periods, the Harris algorithm is used to select 20 feature points (inflection points or high curvature points). The displacement vector of each group of feature points is calculated (e.g., most vector directions are 30° north of west, average distance 80 meters / 30 minutes). K-means cluster analysis is performed on all displacement vectors. The most frequent 30° north of west is taken as the dominant displacement direction. Then, based on the maximum slope of the positioning space where the fire line boundary is located (e.g., 20° uphill), the first angle is calculated to be 4° with an adjustment ratio of 0.2. The dominant displacement direction is adjusted by 4° upwards towards the 30° north of west slope, resulting in the final fire spread direction being 34° north of west.

[0041] Specifically, when performing cluster analysis on all displacement vectors, the direction of each displacement vector is first converted into an angle value of 0°-360°, and the distance is normalized to the [0,1] interval to unify the data dimension. Then, the K-means clustering algorithm is used to determine the optimal number of clusters (usually 2-5 clusters, to adapt to the concentrated characteristics of the fire spread direction) through the elbow rule. The standardized displacement vectors are used as input data, and the mean angle and mean distance of the vectors in each cluster are used as the cluster centers. The cluster assignment of vectors is iteratively adjusted until the cluster centers are stable (the number of iterations ≥ 50 times or the change in the sum of squared errors within the cluster < 0.001). Finally, the number of vectors in each cluster is counted, and the cluster with the most vectors is selected. The direction corresponding to the cluster center is the dominant direction of all displacement vectors, thus completing the cluster analysis.

[0042] Understandably, by extracting fire line boundaries from continuous remote sensing images, calculating feature point displacement vectors, clustering analysis of the dominant direction, and combining this with terrain slope adjustments, accurate and dynamic perception of the fire spread direction was achieved. This provided a reliable basis for fire trend selection, avoided path risks caused by misjudgments of fire direction, and improved the accuracy of situational awareness.

[0043] In some embodiments of this application, adjusting the dominant displacement direction based on the maximum slope of the positioning space where the fire line boundary is located includes: When the terrain of the positioning space where the fire line boundary is located is uphill, the dominant displacement direction is adjusted upwards by a first angle. When the terrain of the positioning space where the fire line boundary is located is downhill, the dominant displacement direction is adjusted to a second angle downwards towards the slope. When the terrain of the positioning space where the fire line boundary is located is a plain, the adjustment amount of the dominant displacement direction is 0. The first angle is equal to the second angle, which is equal to the maximum slope multiplied by the adjustment ratio.

[0044] Specifically, the adjustment ratio is a pre-set coefficient used to correct the angle of the dominant displacement direction based on the maximum slope of the positioning space where the fire line boundary is located. Its value is between 0 and 1. Its core function is to quantify the degree of influence of terrain slope on the direction of fire spread. Moreover, this coefficient is positively correlated with the boosting / resisting effect of slope on fire spread. It is determined based on the physical characteristics of forest fire spread (uphill fires will shift towards the top of the slope due to the enhanced effects of heat convection and radiation, while downhill fires will shift towards the bottom of the slope due to the effects of gravity and the rolling of burning materials), the fire behavior patterns under different slope conditions, and a large amount of empirical fire data monitored by satellite remote sensing. This enables the corrected dominant displacement direction to better match the fire spread trend under the actual terrain, avoiding directional deviations caused by relying solely on displacement vectors and ignoring terrain factors.

[0045] Specifically, the adjustment ratio is determined by collecting historical forest fire data from different slope ranges (0°-5°, 5°-15°, 15°-30°, and above 30°), different combustible material types (coniferous forests, broad-leaved forests, shrubs, and herbs), and different combustion stages (initial combustion, vigorous combustion, and decay). The dominant displacement direction monitored by satellite remote sensing, the actual fire spread direction, and the maximum slope of the corresponding fire line boundary in each case are extracted. The angular deviation between the actual spread direction and the dominant displacement direction in each case is calculated. A linear regression analysis is then used to establish a functional relationship between the angular deviation and the maximum slope. The ratio of the deviation to the maximum slope is calculated as the initial adjustment ratio. Subsequently, multiple sets of field fire simulation experiments (controlling slope, combustible material type, and combustion stage as single variables) are conducted to verify the consistency between the direction corrected by the initial adjustment ratio and the actual spread direction. Abnormal data with a consistency of less than 90% are removed, and the initial adjustment ratio is calibrated. Finally, a fixed adjustment ratio (applicable to most common fire scenarios) or a dynamic adjustment ratio matrix based on slope and combustible material type is determined.

[0046] Specifically, the adjustment ratio is 0.2 (fixed value). When the terrain of the positioning space where the fire line boundary is located is uphill with a maximum slope of 20°, the first angle is calculated as 20° × 0.2 = 4°, and the dominant displacement direction is adjusted upwards by 4°. When the terrain of the positioning space where the fire line boundary is located is downhill with a maximum slope of 20°, the second angle is calculated as 20° × 0.2 = 4°, and the dominant displacement direction is adjusted downwards by 4°. When the terrain of the positioning space where the fire line boundary is located is flat with a maximum slope of 0°, both the first and second angles are 0°, the adjustment amount of the dominant displacement direction is 0, and no angle adjustment is required. Finally, the direction of fire spread after correction based on terrain slope is obtained.

[0047] Understandably, by adjusting the angle rules for uphill, downhill, and plain terrain, topographical factors are incorporated into the fire direction prediction, making the corrected fire spread direction more consistent with actual fire behavior patterns (uphill fires tend to spread to the top of the slope, and downhill fires tend to spread to the bottom of the slope). This avoids directional deviations that rely solely on displacement vectors while ignoring the influence of terrain, thus enhancing the accuracy of fire direction prediction.

[0048] In some embodiments of this application, when the fire truck positioning space and the escape path space constitute an initial escape path, it includes: When the escape path space is selected, the escape path space is sequentially connected with the historical fire truck positioning space and the historical escape path space in chronological order to form a continuous path node sequence. When the last path node is a safe area, the path node sequence is the initial escape path.

[0049] Understandably, by connecting the historical positioning space and the breakout path space in chronological order to form a sequence of path nodes until the destination is a safe area, the initial path forms a continuous and traceable chain of nodes, which clarifies the endpoint conditions for path generation and ensures that the path always aims at a safe area, providing a clear basic path framework for subsequent path optimization.

[0050] In some embodiments of this application, the predicted thermal radiation intensity is equal to the initial thermal radiation intensity multiplied by the cumulative change factor of thermal radiation intensity; The cumulative change factor of thermal radiation intensity is equal to the travel time multiplied by the thermal radiation intensity change coefficient plus 1.

[0051] Understandably, by quantitatively calculating the initial thermal radiation intensity, travel time, and thermal radiation intensity variation coefficient, accurate prediction of future thermal radiation intensity is achieved. This provides a quantitative basis for judging the thermal safety of the route space when the fire truck arrives, avoids misjudgment of route safety based solely on the current thermal radiation intensity, and improves the foresight of route selection.

[0052] In some embodiments of this application, when re-screening the escape path space based on the predicted thermal radiation intensity and generating the final escape path, the process includes: When the predicted thermal radiation intensity is less than the radiation intensity threshold, the current escape path space remains unchanged. When the predicted thermal radiation intensity is greater than or equal to the radiation intensity threshold, the fourth positioning space is re-selected, and the breakout path space is re-selected based on the re-selected fourth positioning space. The breakout path space corresponding to the initial breakout path is replaced with the re-selected breakout path space to obtain the final breakout path.

[0053] Understandably, by preserving safe path space and replacing unsafe path space, dynamic optimization of the path is achieved, ensuring that each node in the final path meets thermal safety requirements when the fire truck arrives. This effectively avoids the risk of fire spreading over time, keeps the path safe and adaptable, and ultimately ensures that the forest fire truck can break through the fire autonomously and efficiently.

[0054] It should be noted that: Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known structures and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0055] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of this application and form different embodiments.

[0056] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A fire line situation awareness and autonomous breakout path generation method for a forest fire vehicle, characterized in that, The method comprises the following steps: dividing a forest area where a fire will occur into a plurality of positioning spaces based on satellite remote sensing images, acquiring a fire truck positioning space representing a current position of a fire truck in real time through satellite remote sensing technology, and extracting a maximum slope, an initial thermal radiation intensity, and an obstacle obstruction value in each positioning space through satellite remote sensing technology; screening an escape path space in the positioning spaces around the fire truck positioning space based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle obstruction value; taking the current escape path space as the fire truck positioning space, and repeating the operation of screening the escape path space until the escape path space screened is in a safe area, so as to form an initial escape path, wherein the fire truck positioning space and the escape path space constitute the initial escape path; calculating a driving time required for the fire truck to drive from an initial position to each escape path space based on an average driving speed of the fire truck, predicting a predicted thermal radiation intensity of the escape path space after the driving time, and re-screening the escape path space based on the predicted thermal radiation intensity, so as to generate a final escape path.

2. The fire line situation awareness and autonomous breakout path generation method for a forest fire vehicle of claim 1, wherein, When the escape path space is screened in the positioning spaces around the fire truck positioning space based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle obstruction value, the method comprises the following steps: screening a first positioning space around the fire truck positioning space, wherein the first positioning space constitutes a first positioning space set; screening a second positioning space with a maximum slope less than a fire truck climbing threshold value in the first positioning space set, wherein the second positioning space constitutes a second positioning space set; screening a third positioning space with an obstacle obstruction value less than or equal to a fire truck obstacle removal ability threshold value in the second positioning space set, wherein the third positioning space constitutes a third positioning space set.

3. The fire line situation awareness and autonomous break-out path generation method for a forest fire vehicle of claim 2, wherein, When the escape path space is screened in the positioning spaces around the fire truck positioning space based on the fire truck positioning space, the maximum slope, the initial thermal radiation intensity, and the obstacle obstruction value, the method further comprises the following steps: screening a fourth positioning space with an initial thermal radiation intensity less than a radiation intensity threshold value in the third positioning space set, wherein the fourth positioning space constitutes a fourth positioning space set; screening a fifth positioning space with the minimum initial radiation intensity in the fourth positioning space set, wherein the fifth positioning space constitutes a fifth positioning space set; acquiring a fire spread direction based on satellite remote sensing technology, and screening the escape path space in the fifth positioning space set based on the fire spread direction.

4. The fire line situation awareness and autonomous break-out path generation method for a forest fire vehicle of claim 3, wherein, When the escape path space is screened in the fifth positioning space set based on the fire spread direction, the method comprises the following steps: when the fifth positioning space exists in a reverse region, a perpendicular region, and a forward region of the fire spread direction, the fifth positioning space in the reverse region of the fire spread direction and having the shortest straight-line distance from the fire truck positioning space is selected as the escape path space. When the fifth positioning space exists in the vertical area and the forward area of the fire spread direction, the fifth positioning space in the vertical area of the fire spread direction and having the shortest straight-line distance from the fire truck positioning space is selected as the breakthrough path space.

5. The fire line situation awareness and autonomous break-out path generation method for a forest fire vehicle of claim 4, wherein, When the fifth positioning space set is screened based on the fire spread direction, the method further comprises the following steps: When the fifth positioning space exists in the forward area of the fire spread direction, the fifth positioning space having the largest included angle with the fire spread direction is selected as the breakthrough path space.

6. The fire line situation awareness and autonomous break-out path generation method for a forest fire vehicle of claim 5, wherein, When the fire spread direction is obtained based on satellite remote sensing technology, the method comprises the following steps: Remote sensing images of a fire area in a continuous preset number of time periods are obtained by satellite remote sensing technology, and fire line boundaries are identified and extracted from the images of each time period; A plurality of groups of feature points are selected on the fire line boundaries of adjacent time periods, and displacement vectors of each group of feature points are calculated; All displacement vectors are subjected to cluster analysis, and the vector direction with the highest frequency of occurrence is taken as the dominant displacement direction; The dominant displacement direction is adjusted based on the maximum slope of the positioning space where the fire line boundary is located, and the adjusted dominant displacement direction is the fire spread direction.

7. The fire line situation awareness and autonomous break-out path generation method for a forest fire vehicle of claim 6, wherein, When the dominant displacement direction is adjusted based on the maximum slope of the positioning space where the fire line boundary is located, the method comprises the following steps: When the terrain of the positioning space where the fire line boundary is located is an uphill, the dominant displacement direction is adjusted by a first angle upward of the slope; When the terrain of the positioning space where the fire line boundary is located is a downhill, the dominant displacement direction is adjusted by a second angle downward of the slope; When the terrain of the positioning space where the fire line boundary is located is a plain, the adjustment amount of the dominant displacement direction is 0; The first angle is equal to the second angle, which is equal to the maximum slope multiplied by an adjustment proportion.

8. The fire line situation awareness and autonomous break-out path generation method for a forest fire vehicle of claim 7, wherein, When the fire truck positioning space and the breakthrough path space constitute an initial breakthrough path, the method comprises the following steps: When the breakthrough path space is screened, the breakthrough path space, historical fire truck positioning spaces, and historical breakthrough path spaces are sequentially connected in time order to form a continuous path node sequence, and when the last path node is a safe area, the path node sequence is the initial breakthrough path.

9. The fire line situation awareness and autonomous breakthrough path generation method for a forest fire truck according to claim 8, wherein: The predicted thermal radiation intensity is equal to the initial thermal radiation intensity multiplied by a thermal radiation intensity cumulative change factor; The thermal radiation intensity cumulative change factor is equal to the travel time multiplied by a thermal radiation intensity change coefficient plus 1.

10. The fire line situation awareness and autonomous break-out path generation method for a forest fire vehicle of claim 9, wherein, When the predicted thermal radiation intensity is less than a radiation intensity threshold, the current breakthrough path space remains unchanged. When the predicted thermal radiation intensity is greater than or equal to the radiation intensity threshold, the fourth positioning space is re-screened, the breakthrough path space is re-screened based on the re-screened fourth positioning space, the corresponding breakthrough path space in the initial breakthrough path is replaced by the re-screened breakthrough path space, and the final breakthrough path is obtained. ​

Images