Space-air-ground cooperative intelligent forest and grassland fire-fighting robot cluster

Abstract

The present application relates to the field of fire fighting technology, in particular to a space-air-ground cooperative intelligent forest and grassland fire fighting robot cluster, which comprises a space-air-ground integrated monitoring unit, a short-range fire extinguishing control unit, a decision-making generation unit, a remote fire extinguishing control unit and an aftereffect evaluation unit. The present application eliminates the monitoring blind area through the mutual complementation of the three monitoring means in the time-space dimension, realizes the full autonomous flight of the fire extinguishing unmanned aerial vehicle in the initial stage and the execution of small-area fire extinguishing by remote bomb launching, and the whole process does not need to enter the dangerous area of the fire field, and the high-cost cluster rocket fire extinguishing system remotely launches the fire extinguishing bomb to execute the out-of-control large-area fire extinguishing, which is conducive to the optimal allocation of tactical resources, avoids excessive damage, makes the combat economy of the whole system reach the best, and on the other hand, adapts to different complex terrains and extreme fire conditions.

Description

Technical Field

[0001] This invention relates to the field of fire protection technology, and more specifically, to a cluster of intelligent forest and grassland fire-fighting robots that integrates air, space, and ground systems. Background Technology

[0002] Forest and grassland fires often occur in mountainous areas, steep slopes, ravines, and densely forested areas. The terrain is rugged, the vegetation is dense, the fire site has a lot of smoke and extremely high temperatures, making it difficult for traditional manual firefighting and vehicle-mounted large firefighting equipment to enter the core fire area, resulting in low firefighting efficiency and extremely high risk of injury or death to firefighters. Existing technologies include some tracked robots used for fire rescue, such as the tracked firefighting robot disclosed in Chinese patent CN205287339U. This robot uses flame sensors to identify flames and ultrasonic ranging sensors to detect distances. A microcontroller controls the tracks to move to the flame location and extinguish the fire. While this solution possesses basic autonomous firefighting capabilities, it is only suitable for indoor or simple terrain scenarios. It cannot cope with complex terrains such as steep slopes and ravines in forests, grasslands, and mountains. Furthermore, it lacks aerial collaborative reconnaissance and global situational awareness capabilities, and its firefighting distance is short and its coverage area is small, making it difficult to deal with large-scale, high-intensity forest fires. Chinese patent CN220443080U discloses a tracked firefighting robot with an intelligent spraying device. This robot uses an electric telescopic rod to adjust the distance between its chassis and the ground to cross obstacles and uses an infrared camera to monitor temperature changes. However, this solution is still a single ground-based operation device and does not have the ability to conduct global aerial fire reconnaissance. It cannot achieve precise fire point location and coordinated air-ground firefighting. Furthermore, its firefighting operation mode is limited, mostly consisting of single water mist spraying or dry powder spraying, and it cannot adaptively switch firefighting media according to the intensity and type of fire. In view of this, we propose an air-ground coordinated intelligent forest and grassland firefighting robot cluster. Summary of the Invention

[0003] The purpose of this invention is to provide an air-ground collaborative intelligent forest and grassland firefighting robot cluster to solve the problems mentioned in the background art, such as the poor mobility of existing equipment in complex terrain conditions such as forests, grasslands, mountains, steep slopes, and under forests, which prevents it from reaching the core fire site; the lack of aerial global reconnaissance and situational awareness capabilities of a single ground robot, which leads to inaccurate fire point location and lack of global basis for path planning; and the problem that existing firefighting robots have limited firefighting methods and cannot adaptively switch firefighting modes according to the type and intensity of the fire.

[0004] To achieve the above objectives, the present invention provides an air-ground integrated intelligent forest and grassland firefighting robot cluster, including an air-ground integrated monitoring unit, a short-range fire extinguishing control unit, a decision generation unit, a long-range fire extinguishing control unit, and a post-effect evaluation unit; The integrated air-space-ground monitoring unit is used to monitor the fire situation in the target area through the collaboration of fixed monitoring nodes, mobile monitoring platforms and space-based monitoring. After confirming the fire, it issues an early warning signal, locates the fire location information, and continuously collects fire scene perception data. The short-range fire extinguishing control unit is used to activate the intelligent command system after receiving the early warning signal, automatically plan the flight route of the fire extinguishing drone to the fire point, execute the fire extinguishing bomb drop command on the fire extinguishing drone to the fire point, carry out fire extinguishing operations, and obtain fire extinguishing effect information. The decision generation unit is used to receive fire extinguishing effect information and, when it is determined that the fire is not under control, generate a decision instruction containing global situation information and fire extinguishing strategy. The remote fire suppression control unit is used to receive the decision command and drive the cluster rocket fire suppression system to launch multiple fire extinguishing projectiles based on the fire location information to carry out remote fire suppression operations. The aftereffect assessment unit is used to detect fire reignition after the fire extinguishing operation is completed, and decide whether to trigger supplementary fire extinguishing based on the detection results. If so, the short-range fire extinguishing control unit is immediately driven to perform secondary fire extinguishing to completely eliminate hidden dangers and generate fire extinguishing action records.

[0005] As a further improvement to this technical solution, the integrated air-space-ground monitoring unit includes a fixed monitoring subunit, an aerial patrol subunit, and a space-based monitoring subunit: The fixed monitoring subunit is a panoramic photoelectric alarm system deployed at fixed nodes in the forest area. It is used to continuously and automatically scan and detect the surrounding area using infrared thermal imaging detection components, identify infrared data, output fire perception data after detecting a heat source, generate early warning signals and fire location information, and use visible light high-definition camera components to capture images in the direction of the target to generate visible light images of the heat source area. The aerial patrol subunit consists of at least one reconnaissance drone, which is equipped with an electro-optical pod to patrol the target area according to a preset route, acquire fire scene perception data including high-definition images and thermal images, identify targets that match the characteristics of a fire, output early warning signals and fire location information, and continuously perform fire scene video acquisition tasks over the fire scene until the firefighting operation ends or the reconnaissance drone reaches its limit of endurance and needs to return to its base for battery replacement. After receiving the early warning signal from the fixed monitoring subunit, it flies to the fire location to conduct close-range confirmation and high-definition image acquisition. The space-based monitoring subunit is used to access space-based remote sensing satellite data to obtain macroscopic fire field perception data of the target area, identify targets that meet the characteristics of the fire, and output early warning signals and fire location information.

[0006] As a further improvement to this technical solution, the integrated air-space-ground monitoring unit also includes a cross-verification module, which is used to receive fire scene perception data output by the fixed monitoring subunit, the aerial patrol subunit, and the space-based monitoring subunit. If a certain fire scene perception data is continuously output inconsistent with the other two fire scene perception data, the fire scene perception data is automatically marked as abnormal, an equipment status alarm is generated, and pushed to the maintenance personnel.

[0007] As a further improvement to this technical solution, the fixed monitoring subunit, the aerial patrol subunit, and the space-based monitoring subunit are connected to the command center for communication. When outputting early warning signals, they report fire scene perception data, fire location information, and fire video to the command center within seconds through the forest and grassland emergency communication network. The command center includes the 119 command center and the emergency command center.

[0008] As a further improvement to this technical solution, the short-range fire suppression control unit includes a flight path planning subunit, a bomb release control subunit, and an effect recognition subunit: The flight path planning subunit is configured to acquire the fire location information contained in the early warning signal, and automatically plan the optimal flight path of the fire-fighting drone from its current location to the fire location based on GIS geographic information data and digital elevation model. The optimal flight path includes obstacle avoidance path and no-fly zone avoidance path. The bombing control subunit is used to acquire real-time images of the fire scene through the optoelectronic pod carried on the fire-fighting drone after the fire-fighting drone arrives over the fire scene, identify the core area of ​​the fire scene, and control the timing and location of the fire-fighting bombs. The effect recognition subunit is used to identify the fire extinguishing effect based on the fire scene image transmitted back by the photoelectric pod after the fire extinguishing bomb is dropped, and generate fire extinguishing effect information. If the highest temperature in the core area of ​​the fire drops below the preset fire extinguishing temperature threshold after the bomb is dropped, and no continuous open flame is observed in the visible light image, and there are no thermal anomalies exceeding the temperature threshold in the infrared thermal image, then the fire extinguishing is successful and the fire has been effectively extinguished without further action. Otherwise, the fire is not under control.

[0009] As a further improvement to this technical solution, the fire-fighting drone is a rotary-wing drone with vertical take-off and landing capabilities. The rotary-wing drone carries 6 to 8 fire-fighting bombs, with each fire-fighting bomb having a charge of 50 kg. The operating radius of the rotary-wing drone is not less than 50 km, and the area controlled by a single bomb drop is not less than 400 m².

[0010] As a further improvement to this technical solution, the decision generation unit includes a communication relay subunit, a situation fusion subunit, and an intelligent decision-making subunit: The communication relay subunit is used to control the intelligent command system to arrive at the periphery of the fire and carry out communication relay when it is determined that the fire is not effectively controlled. This enables the fire-fighting equipment around the fire to be interconnected through the self-organizing network communication protocol. The fire-fighting equipment around the fire includes a modular command vehicle, various fire-fighting drones, cluster rocket fire-fighting vehicles, reconnaissance drones, fixed monitoring nodes, and handheld terminals for ground firefighters. The situation fusion subunit is used to receive the fire scene perception data continuously collected by the air-space-ground integrated monitoring unit, including fire video, infrared thermal image data and fire location information, and to perform spatiotemporal registration and fusion processing on the fire scene perception data to generate a panoramic fire scene situation map under a unified spatiotemporal reference. The intelligent decision-making subunit is used to call the forest database and expert database, analyze and process the panoramic fire situation map based on machine learning algorithms, extract the fire boundary contour from the fire video through image segmentation algorithm, extract the temperature anomaly area and temperature distribution cloud map from the infrared thermal image data through temperature threshold method, calculate the three-dimensional coordinates of the fire point from the fire location information, calculate the fire area change rate from the UAV position change data, and generate a fire extinguishing plan as the decision instruction. The fire extinguishing plan includes launch position planning and safe firing range planning.

[0011] As a further improvement to this technical solution, the intelligent command system is a modular command vehicle, which is equipped with a command and management seat, a collaborative control seat, a drone operation seat, and an observation and command seat. The modular command vehicle has a seating capacity of 9 people.

[0012] As a further improvement to this technical solution, the cluster rocket fire extinguishing system is mounted on a high-mobility off-road chassis to form a cluster rocket fire extinguishing vehicle, equipped with a launcher for 40 fire extinguishing bombs; The remote fire extinguishing control unit automatically calculates the firing azimuth and elevation angles based on the launch position planning and safe firing range planning in the decision command, and controls the azimuth and elevation angles of the launcher through the hydraulic elevation mechanism and the hydraulic rotation mechanism to drive the fire extinguishing shells to fire simultaneously. The fire extinguishing agent coverage and control area of ​​a single salvo is not less than 10,000 m².

[0013] As a further improvement to this technical solution, the aftereffect management unit includes a reignition detection subunit, a report generation subunit, and a data archiving subunit: The reignition detection subunit is used to control the integrated air-ground monitoring unit to perform a full-coverage thermal imaging scan of the fire area after the fire extinguishing operation is completed, extract the temperature distribution data of the fire area, determine the presence of reignition when an abnormal heat source point exceeding the preset temperature threshold is detected, and immediately trigger the near-field response unit to perform supplementary fire extinguishing. After confirming that there is no risk of reignition in the fire area, the burned area and vegetation damage classification statistics are calculated based on the fire perception data collected by the integrated air-ground monitoring unit. The report generation subunit is used to store fire scene perception data, calculate the burned area and vegetation damage classification statistics, and summarize all fire scene perception data and work records continuously collected and stored throughout the entire firefighting operation to generate a firefighting operation record. All fire scene perception data includes early warning information at the moment of fire discovery, visible light video and infrared thermal image video recorded by reconnaissance drones throughout the process, and scanning detection records of fixed monitoring nodes to generate a structured firefighting operation report. At the same time, the work record includes the flight path planning log of the short-range firefighting control unit, the bomb drop time and the number of fire extinguishing bombs used, as well as the time when the cluster rocket fire extinguishing vehicle enters the battle position, the launch time of each rocket, the time of salvo completion, and ammunition resupply records.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: In this cluster of intelligent forest and grassland firefighting robots, a three-dimensional monitoring network is formed through integrated air-space-ground monitoring units, consisting of fixed monitoring, aerial mobile patrols, and space-based surveys. Fixed monitoring nodes ensure uninterrupted monitoring of key areas, mobile monitoring platforms respond to alarms for rapid close-range confirmation and detailed investigation, and space-based monitoring provides extensive background screening and cross-verification. These three monitoring methods complement each other in the spatiotemporal dimensions, eliminating monitoring blind spots. The firefighting drones are used to execute missions with fully autonomous flight and remote bombing intelligent command systems, without needing to enter dangerous fire zones. The unmanned operation mode eliminates the inherent personnel safety risks in traditional firefighting operations, such as firefighters being trapped by fire, inhaling dense smoke, and suffering high-temperature burns, under complex terrain and extreme fire conditions. On the other hand, it is beneficial to quickly reach the fire area in the early stages of a fire to carry out precise firefighting strikes, achieving the integrated reconnaissance and strike objectives of detection and suppression, and fire control. When a fire is determined to be out of control, the cluster rocket fire extinguishing system is driven to launch multiple fire extinguishing projectiles based on the fire location information to carry out remote fire extinguishing operations, forming area coverage suppression. This allows fire extinguishing drones to perform fully autonomous flight and remote bombing for small-area fire extinguishing in the initial stage, while the high-cost cluster rocket fire extinguishing system can remotely launch fire extinguishing projectiles to extinguish large-area fires that are out of control. On the one hand, this is conducive to the optimal allocation of tactical resources, avoiding excessive casualties and maximizing the operational economy of the entire system. On the other hand, it is adaptable to different complex terrains and extreme fire conditions. Attached Figure Description

[0015] Figure 1 This is a block diagram illustrating the overall structural principle of the present invention; Figure 2 This is a schematic diagram illustrating the overall structure of the present invention broken down. Figure 3 This is a schematic diagram of the structure of the cluster rocket fire engine of the present invention; Figure 4 This is a schematic diagram of the operational supply system of the present invention. Detailed Implementation

[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1 Please refer to the full text. Figures 1-4 As shown, this embodiment provides a space-air-ground collaborative intelligent forest and grassland firefighting robot cluster, specifically as follows: Figures 1-2 As shown, it includes an integrated air-ground monitoring unit, a short-range fire suppression control unit, a decision generation unit, a long-range fire suppression control unit, and a post-effect assessment unit; Perception Layer: The integrated air-space-ground monitoring unit is used to monitor fires in target areas through the collaboration of fixed monitoring nodes, mobile monitoring platforms, and space-based monitoring. After confirming a fire, it issues an early warning signal, locates the fire's position, and continuously collects fire perception data. This facilitates the formation of a three-tiered three-dimensional monitoring network consisting of "fixed monitoring + aerial mobile patrols + space-based surveys." Fixed monitoring nodes enable uninterrupted monitoring of key areas, mobile monitoring platforms respond to alarms for rapid close-range confirmation and detailed investigation, and space-based monitoring provides extensive background screening and cross-verification. The three monitoring methods complement each other in the spatiotemporal dimensions, eliminating monitoring blind spots. Furthermore, the integrated air-space-ground monitoring unit includes fixed monitoring subunits, aerial patrol subunits, and space-based monitoring subunits: The fixed monitoring subunit is a panoramic photoelectric alarm system deployed at fixed nodes in the forest area (including but not limited to lookout towers, mountain tops, communication towers and other locations with good visibility). It is used to continuously and automatically scan and detect the surrounding area using infrared thermal imaging detection components, identify infrared data, output fire perception data after detecting a heat source, generate early warning signals and fire location information, and use visible light high-definition camera components to capture images in the direction of the target to generate visible light images of the heat source area. Specifically, the Zhoushi Optoelectronic Alarm System includes an infrared thermal imaging detection component, a visible light high-definition camera component, an embedded intelligent processing unit, a solar power supply system and battery pack, and a wireless communication module. The infrared thermal imaging detection component uses an uncooled vanadium oxide (VOx) focal plane detector or a cooled mercury cadmium telluride (MCT) focal plane detector, with a field of view of 360° × (-15°~+60°). It uses a motor-driven pan-tilt unit to achieve continuous horizontal rotation scanning and step-by-step tilt scanning to identify heat source targets. The visible light high-definition camera component uses a high-definition camera with an optical zoom lens of 30x or higher. After the infrared thermal imaging detects a suspected heat source, it automatically zooms to align with the target direction and acquires a high-definition visible light image of the target area. The embedded intelligent processing unit is deployed on a fixed monitoring section. The system operates locally and runs a fire detection algorithm to perform background modeling and moving target extraction on infrared thermal image sequences. Based on temperature thresholding, morphological feature analysis, and temporal change detection, it identifies heat source targets that match fire characteristics from complex backgrounds (such as solar warming and geothermal anomalies). After detecting a fire, it calculates the target's azimuth, elevation, and distance (based on a laser rangefinder or line-of-sight calculation from a known terrain database) and converts them into latitude and longitude coordinates as fire location information. A solar power system and battery pack are used to ensure continuous and uninterrupted power supply to fixed monitoring nodes in remote forest areas without mains power. The battery pack capacity is no less than the power required for 72 hours of continuous operation. The wireless communication module supports three communication methods: 4G / 5G public network, self-organizing network, and Beidou short message, ensuring reliable uploading of monitoring data. The aerial patrol subunit consists of at least one reconnaissance drone, equipped with an electro-optical pod. This drone patrols the target area along a pre-set route, acquiring fire-sensing data including high-definition images and thermal images. After identifying targets matching fire characteristics, it outputs warning signals and fire location information. Simultaneously, it continuously performs fire-field video acquisition tasks over the fire until the firefighting operation ends or the drone reaches its maximum endurance and needs to return for battery recharging. Upon receiving a warning signal from the fixed monitoring subunit, it flies to the fire location to conduct close-range confirmation and high-definition image acquisition. The reconnaissance drone employs a multi-rotor configuration (quadcopter or hexcopter), possessing vertical takeoff and landing (VTOL) and hovering capabilities. It does not require a dedicated takeoff and landing site (such as a runway) and can take off and land in small areas such as forest clearings, roads, and helipads. It also possesses low-speed flight and stationary hovering capabilities. Suitable for precise low-altitude observation and continuous monitoring of target areas; flexible in operation, adaptable to flight operations in complex terrain conditions (valleys, steep slopes, and over dense forests). The reconnaissance drone is in standby mode, parked in drone nests or fixed parking points in forest areas. The nest has automatic charging (the drone automatically replenishes its power through contact or wireless charging devices after returning to the nest, ensuring it is fully charged for the next sortie), environmental temperature and humidity control (maintaining the temperature and humidity inside the nest within a suitable range to protect battery performance and electronic equipment reliability), and remote wake-up function (automatically starting the drone system self-check and take-off procedures by receiving remote control commands, without the need for on-site personnel operation). The nest maintains a connection with the command system through self-organizing network communication or public network communication, continuously reporting the drone's status (battery level, self-check status, location, etc.). Upon receiving an early warning signal from a fixed monitoring subunit, the system automatically generates a flight path based on a high-precision digital elevation model, using the fire location coordinates and the current location coordinates of the UAV. The flight path planning algorithm must meet the following constraints: avoid known obstacles (including mountains, high-voltage power line towers, communication towers, buildings, etc., with the obstacle database pre-installed in the onboard computer); avoid no-fly zones (such as airport airspace protection zones, military restricted zones, etc.); prioritize routes that shorten flight time (reaching the fire area as quickly as possible); and ensure sufficient power for round-trip flights and loitering operations within the range constraint (if the remaining power is insufficient to complete the round-trip flights and the specified loitering operations, the system will automatically select a mid-flight power-swapping option or dispatch a backup UAV). The space-based monitoring subunit is used to access space-based remote sensing satellite data to obtain macroscopic fire field perception data of the target area. After identifying targets that meet the characteristics of the fire, it outputs early warning signals and fire location information. Specifically, the monitoring data of space-based remote sensing satellite is not limited by ground terrain and forest area transportation conditions. It can cover vast forest areas that cannot be covered by fixed monitoring nodes in a very short time and obtain thermal anomaly background information of the entire forest area.

[0018] Furthermore, the integrated air-space-ground monitoring unit also includes a cross-validation module, which receives fire scene perception data output from the fixed monitoring subunit, the aerial patrol subunit, and the space-based monitoring subunit. If a certain fire scene perception data (such as the photoelectric payload of a fixed monitoring node or a reconnaissance drone) continuously outputs data that is inconsistent with the other two fire scene perception data (the inconsistency frequency exceeds a preset threshold, typically with an inconsistency rate exceeding 10% in 100 valid comparisons), the fire scene perception data is automatically marked as abnormal, an equipment status alarm is generated, and pushed to maintenance personnel. Maintenance personnel can then arrange for on-site testing, sensor calibration, or component replacement of the equipment, detecting performance degradation before the equipment completely fails, avoiding missed alarms or false alarms due to long-term operation of faulty equipment, improving the long-term safety of the system and enhancing the reliability of fire confirmation.

[0019] The fixed monitoring subunit, aerial patrol subunit, and space-based monitoring subunit are connected to the command center. When outputting early warning signals, they report fire scene perception data, fire location information, and fire video to the command center within seconds through the forest and grassland emergency communication network. The command center includes the 119 command center and the emergency command center. The 119 command center and the emergency command center receive and respond synchronously, realizing the simultaneous activation of fire fighting force dispatch and cross-departmental linkage coordination, avoiding secondary delays in information forwarding.

[0020] Secondly, the rapid response layer: The short-range fire extinguishing control unit is used to activate the intelligent command system after receiving the early warning signal. It automatically plans the flight path of the fire extinguishing drone to the fire point, executes the command to drop fire extinguishing bombs on the drone to the fire point, carries out fire extinguishing operations, and obtains information on the fire extinguishing effect. On the one hand, the fully autonomous flight and remote bomb dropping operation of the fire extinguishing drone allows the operator to carry out the task within the intelligent command system at a distance of 50km (within the operation radius) from the fire site. The operator does not need to enter the dangerous area of ​​the fire site. The unmanned operation mode eliminates the inherent personnel safety risks in traditional fire extinguishing operations, such as firefighters being trapped by fire, inhaling thick smoke, and suffering high-temperature burns, in complex terrain (steep slopes, canyons, dense forests) and extreme fire site conditions. On the other hand, it is conducive to the rapid arrival of fire extinguishing drones in the airspace above the fire site in the early stage of the fire (when the burned area is small and the fire has not spread on a large scale) to carry out precise fire extinguishing strikes, achieving the integrated goal of detection and suppression and fire control. Furthermore, the short-range fire suppression control unit includes a flight path planning subunit, a bomb release control subunit, and an effect recognition subunit: The flight path planning subunit is configured to acquire fire location information contained in the early warning signal. Based on GIS geographic information data and digital elevation model, it automatically plans the optimal flight path for the firefighting drone from its current location to the fire location. The optimal flight path includes obstacle avoidance paths and no-fly zone avoidance paths, ensuring that the drone safely reaches the fire scene in the shortest possible time. The specific working principle is as follows: Using the drone's origin and the fire location as diagonal points, a three-dimensional planning space is formed by expanding outwards with a certain margin. This planning space is discretized into a three-dimensional grid (typical horizontal grid size is 100m × 100m, and typical vertical grid size is 50m). For each grid cell, the digital elevation model is used to mark whether the cell is occupied by terrain (impassable), whether it belongs to an obstacle-affected area (with a safety margin around the obstacle, entry is prohibited), whether it belongs to a no-fly zone (entry is prohibited), and whether it is a passable airspace. The flight path planning must meet strict safety constraints, including a minimum ground clearance of no less than 50m (safety margin, to prevent the drone from crashing into mountains due to errors in barometric altitude measurement and terrain data); a horizontal distance of no less than 100m from obstacles (safe detour distance); no entry into no-fly zones; and the total length of the flight path must not exceed the drone's maximum range under the current payload and weather conditions (to ensure sufficient battery power for flight and return). The three-dimensional A search algorithm is used to search for the optimal path in the discretized planning space. The search process continuously expands the node with the minimum cost function until the destination node is selected from the open list for expansion. At this point, the path from the starting point to the destination is the optimal path. The searched path is a polyline path (composed of multiple grid center points connected sequentially), which has sharp turns and is unsuitable for flight. The route planning sub-unit performs B-spline curve smoothing on the polyline path to generate a smooth three-dimensional route with continuous curvature. The smoothed route consists of a series of waypoints (each waypoint includes longitude, latitude, altitude and expected arrival time), which are then imported into the firefighting drone flight control computer for execution. Note: During the flight mission of the firefighting drone, the route planning subunit remains online and continues to identify the real-time position of the firefighting drone through GIS geographic information data. It continuously receives the real-time position and meteorological data transmitted back by the firefighting drone. If a sudden obstacle appears on the preset route during the flight (such as a drone that is temporarily launched, a sudden area of ​​severe convective weather, etc.) or the flight path deviates from the preset route by more than the safety threshold, the route planning subunit will re-execute the above planning process at the current position, generate a new alternative route and upload it to the drone in flight in real time to achieve dynamic obstacle avoidance and heading correction. The bombing control subunit is used to hover over the fire after the fire-fighting drone arrives at the fire site (the hovering altitude is the safe operating altitude, usually 150-300m, which can be adjusted according to the actual heat radiation intensity and wind speed at the fire site). It acquires real-time images of the fire site through the electro-optical pod mounted on the fire-fighting drone, identifies the core area of ​​the fire, and controls the timing and location of the fire-extinguishing bombs. This achieves precise aiming at the core area of ​​the fire, accurately delivering the effective extinguishing agent to the target area, avoiding the drawback of large amounts of extinguishing agent wasted in traditional fire-fighting methods. Specifically: In search mode, the electro-optical pod performs a small-range scan centered on the fire location coordinates (scanning radius 200~500m). Through the coordinated work of the infrared thermal imaging camera and the visible light camera inside the electro-optical pod, the airborne AI target detection algorithm identifies the coordinates of target points in the fire scene, including the highest temperature point (core area of ​​the fire) and the front line of the fire (the front of the direction of fire spread), as the core area of ​​the fire scene. Based on the target point coordinates, the current coordinates of the drone, and real-time wind speed and direction data (provided by the wind speed sensor on the drone), the lead time and drift correction of the fire extinguishing bomb are calculated. The fire extinguishing bomb has a free flight phase from its release to its impact with the target (affected by gravity and wind resistance). The fire extinguishing drone needs to release the bomb a certain distance upwind of the target point in advance, so that the fire extinguishing bomb can drift to the target point under the action of wind. The timing and location of the bomb release are then sent to control the fire extinguishing drone. The effect recognition subunit is used to identify the fire extinguishing effect based on the fire scene images transmitted back by the photoelectric pod after the fire extinguishing bomb is dropped, and to generate fire extinguishing effect information. If the highest temperature in the core area of ​​the fire drops below the preset fire extinguishing temperature threshold after the bomb is dropped (typically, it drops to within 20°C of the ambient temperature), and no continuous open flames are observed in the visible light image, and there are no thermal anomalies exceeding the temperature threshold in the infrared thermal image, then the fire extinguishing is successful and the fire has been effectively extinguished without further action. Conversely, if the fire is not controlled, the fire is partially suppressed, but there are still residual fire points that need to be dealt with. This avoids the situation of unknown fire extinguishing effect in traditional operations, facilitates further processing, and improves the quality of subsequent fire extinguishing.

[0021] Specifically, the firefighting drone is a rotary-wing drone with vertical takeoff and landing capabilities. It requires no dedicated takeoff and landing facilities such as runways or catapults, and can complete takeoff and landing operations on small, flat areas such as temporary helipads in forest areas, road surfaces, and firebreaks. This significantly reduces the firefighting system's dependence on infrastructure, enabling rapid deployment to remote forest areas. The hovering capability of the rotary-wing configuration allows the drone to maintain a fixed position over the fire, providing a stable platform for precise targeting of the electro-optical pod and accurate delivery of fire extinguishing bombs. The rotary-wing drone carries 6 to 8 fire extinguishing bombs, each with a charge of approximately 50 kg. Therefore, it can be seen that the firefighting drone is powered by a high-energy-density lithium battery pack, and the motor drive system provides a sufficient thrust-to-weight ratio (not less than 1.8). With the takeoff and landing point as the center, and fully loaded with fire extinguishing bombs, the firefighting drone can fly a distance of 50 km to reach the fire, complete the bombing operation, and return safely. The operating radius of the rotary-wing drone is not less than 50 km, and the single bombing coverage area is [not specified]. The area should be no less than 400 m². The type of fire extinguishing bomb can be flexibly configured according to the forest vegetation characteristics and fire type. Optional types include dry powder fire extinguishing bombs (suitable for surface fires and initial fires, extinguishing fires by isolating oxygen through dry powder coverage), water-based fire extinguishing bombs (suitable for crown fires and high-temperature fires, achieving rapid cooling through water vaporization and heat absorption), and gel fire extinguishing bombs (suitable for steep slopes, with good gel adhesion, minimal leakage, and long-lasting effect). The extinguishing agent filled in the fire extinguishing bomb will extinguish after the bomb hits the target area. The extinguishing agent is dispersed or released through a fuse to form a covering layer in the air above the fire. The containment area of ​​a single bomb drop (one or more fire extinguishing bombs dropped simultaneously or sequentially) is not less than 400 m². Under standard meteorological conditions (wind speed ≤ 3 m / s), the lower limit of the fire area that a single fire-fighting drone bomb drop operation can effectively cover and extinguish is calculated based on the effective radius of the fire extinguishing bomb and the fire extinguishing agent diffusion model, taking into account the dispersion distribution of the fire extinguishing agent in the air above the fire and the concentration decay characteristics of the fire extinguishing agent.

[0022] Third, the central decision-making layer: The decision generation unit is used to receive information on the fire extinguishing effect and, when it is determined that the fire is not under control, to generate decision instructions containing global situation information and fire extinguishing strategies; Furthermore, due to the fire spreading faster than expected, the fire area had already exceeded the single-use containment capacity (400m²) when the fire-fighting drone arrived. Terrain obstruction reduced the accuracy of the bombs, and the fire-extinguishing bombs on the drones failed to effectively target the core area of ​​the fire. Strong winds or rising hot air currents caused the extinguishing agent to deviate from its dispersion direction, failing to effectively cover the fire area. Alternatively, the fire area might be located on a steep slope or in a canyon, preventing the fire-fighting drone from effectively aiming at the target from a safe distance. Therefore, the decision generation unit includes a communication relay subunit, a situation fusion subunit, and an intelligent decision-making subunit. To avoid communication interruptions and ensure subsequent coordination, a communication relay subunit is used to control the intelligent command system to reach the vicinity of the fire scene and establish communication relay when the fire is determined to be out of control. This enables various firefighting equipment around the fire scene to interconnect via a self-organizing network communication protocol. These equipment include modular command vehicles, firefighting drones, cluster rocket fire trucks, reconnaissance drones, fixed monitoring nodes, and handheld terminals for ground firefighters. Specifically, the self-organizing network communication protocol adopts either the MESH protocol or the Ad hoc protocol to construct a decentralized distributed communication network. This network acts as both a communication terminal for sending and receiving information and a relay node for forwarding information from other nodes. The network topology is mesh-like, and if direct communication between any two nodes is not possible, data can be forwarded through other nodes in a multi-hop manner to reach its destination. Furthermore, the self-organizing network communication protocol avoids interference with public network signals. The situation fusion subunit is used to receive fire scene perception data continuously collected by the air-space-ground integrated monitoring unit, including fire video, infrared thermal image data and fire location information, and to perform spatiotemporal registration and fusion processing on the fire scene perception data to generate a panoramic fire scene situation map under a unified spatiotemporal reference. The intelligent decision-making subunit utilizes forest and expert databases to analyze and process panoramic fire situation maps based on machine learning algorithms. It extracts fire boundary contours (represented as polygon coordinate strings) from fire videos using image segmentation algorithms, extracts temperature anomaly areas and temperature distribution cloud maps from infrared thermal image data using temperature thresholding, calculates the three-dimensional coordinates of fire points from fire location information, calculates the rate of change of fire area (reflecting the spread speed) from UAV position change data, and generates a fire extinguishing plan as decision-making instructions. The fire extinguishing plan includes launch position planning and safe firing range planning, specifically: Launch position planning: Based on fire location information, wind direction and terrain conditions, plan the best launch / takeoff position for each cluster rocket launcher; Safe firing range planning: The safe firing range is a key constraint to ensure operational safety and avoid collateral damage. Based on the coordinates and fire location information of each cluster rocket launcher, the firing direction (the direction line from the launch point to the target point) is calculated. Then, with the firing direction as the center line, within the safe firing range sector angle (typical value ±15°) specified by the expert database, it is checked whether the sector angle area passes through densely populated areas, settlements, important facilities, protected areas, and friendly positions, etc., which are prohibited firing areas. If there are prohibited firing areas within the safe firing range, the position or firing direction is automatically adjusted until the safe firing range verification is passed. Note that if the safe firing range requirements cannot be met no matter what adjustments are made, the system will issue an alarm, and the commander will need to intervene manually to make a decision.

[0023] Specifically, the intelligent command system is a modular command vehicle. Inside the modular command vehicle are command and management seats, collaborative control seats, drone control seats, and observation and command seats. The modular command vehicle has a seating capacity of 9 people, including: The command and management position is located in the front center of the modular shelter (the best observation position facing the main display screen), and is guarded by the on-site commander (usually the highest-ranking on-site commander of the fire and rescue agency or the on-site person in charge designated by the emergency management department). Through the touch terminal, the commander can view the panoramic fire situation map generated by the situation fusion subunit in real time (including fire boundary, fire intensity distribution, spread trend prediction and current weather conditions), review the fire fighting plan generated by the intelligent decision-making subunit, comprehensively judge the rationality of the overall fire situation and tactical choices, and confirm or reject the safe firing range scheme and battle position planning recommended by the intelligent decision-making subunit. The coordination control positions are located on the left and right sides of the command and management positions (or arranged in an arc on both sides in front), and are guarded by coordinators from each combat unit. Each coordinator is responsible for the command and coordination of a combat detachment or a type of combat equipment. For example, Coordinator A is responsible for the formation management and bombing mission coordination of all firefighting drones (including the position allocation of the drone swarm, multi-squad coordinated bombing plans, etc.); Coordinator B is responsible for the command and coordination of the cluster rocket firefighting system (including the safety confirmation of the launch position, salvo timing control, ammunition supply scheduling, etc.); Coordinator C is responsible for the reconnaissance, surveillance and communication support unit (including flight path monitoring and payload control of reconnaissance drones, status monitoring of communication relay equipment, etc.); when ground firefighting forces are involved, Coordinator D is responsible for the coordination and timing arrangement of ground forces (water trucks, fire brigades) and air forces. The drone control station is located behind or to the side of the collaborative control station (in a position where the line of sight does not directly conflict with the main display screen, ensuring that the pilot is not disturbed by the screen during operation). It is staffed by a professional pilot with drone flight qualifications. Depending on the size of the fire and the number of drones deployed, for example, 2-3 pilots can be configured to work simultaneously. The main pilot is responsible for controlling the bombing operation of the fire-fighting drone, the assistant pilot is responsible for monitoring the flight of other drones and taking over in emergencies, and the backup pilots are rotated during high-intensity continuous operations to ensure that the operating accuracy does not decrease due to fatigue. The observation and command position is located at the rear of the modular facility (near the entrance / exit, facilitating personnel entry and exit without interfering with operations at the front), allowing trainees to observe and learn.

[0024] Fourth, the heavy firepower suppression layer: The remote fire suppression control unit is used to receive decision-making instructions and drive the cluster rocket fire suppression system to launch multiple fire extinguishing shells based on the fire location information to carry out remote fire suppression operations; Specifically, such as Figure 3As shown, the cluster rocket fire extinguishing system is mounted on a high-mobility off-road chassis, forming a cluster rocket fire extinguishing vehicle, which is conducive to adapting to different road conditions and avoiding delays in combat operations due to road conditions. It is equipped with a launcher for 40 fire extinguishing shells. The remote fire suppression control unit automatically calculates the firing azimuth and elevation angles based on the firing position planning and safe firing range planning in the decision command, and controls the azimuth and elevation angles of the launcher through the hydraulic elevation mechanism and the hydraulic rotation mechanism. Specifically, the launcher is connected to the chassis through the hydraulic elevation mechanism (elevation angle adjustment range 0°~+60°) and the electric / hydraulic rotation mechanism (azimuth angle adjustment range ±180°) to achieve omnidirectional coverage and precise aiming. The system drives a salvo of fire extinguishing bombs, which are loaded with highly efficient fire extinguishing agents (dry powder, ultrafine water mist, or aerosol, selected based on the type of vegetation and weather conditions at the fire site). A single salvo can cover an area of ​​no less than 10,000 m². There is zero contact between personnel and the fire extinguishing bombs during their flight. The long-range cluster rocket fire extinguishing system is launched from the ground and uses a parabolic trajectory to project the fire extinguishing bombs over a large area of ​​the fire site, forming a surface coverage suppression. The high-cost and high-intensity long-range suppression method is only used when the fire situation exceeds the capabilities of short-range response.

[0025] It is worth noting that, for example Figure 4 As shown, both the cluster rocket fire-fighting train and the fire-fighting drone are equipped with a combat resupply system. The combat resupply system is used to load fire extinguishing shells onto the cluster rocket fire-fighting train and the fire-fighting drone via a vehicle-mounted crane on an ammunition resupply vehicle. When resupplying the cluster rocket fire-fighting train, the ammunition resupply vehicle first reverses to the side or rear of the cluster rocket fire-fighting train. The fire extinguishing shells are then pushed into the launch tube one by one and locked in place by a robotic arm or manual assistance. The circuit and fuse interface are then connected. After loading, the fire-fighting train can be put back into combat, achieving a continuous suppression capability of "resupplying after firing and firing again after resupplying". When resupplying the fire-fighting drone, after the fire-fighting drone returns and lands, ground crew (or an automatic loading mechanism) will replace the fire extinguishing shells and replace (or recharge) the batteries.

[0026] Fifth, the closed-loop protection layer: the after-effect assessment unit is used to detect fire reignition after the fire extinguishing operation is completed, and decide whether to trigger supplementary fire extinguishing based on the detection results. If so, it immediately drives the short-range fire extinguishing control unit to perform secondary fire extinguishing, completely eliminate hidden dangers, and generate fire extinguishing action records.

[0027] Furthermore, the aftereffect management unit includes a reignition detection subunit, a report generation subunit, and a data archiving subunit: The reignition detection subunit is used to control the integrated air-ground monitoring unit to perform a full-coverage thermal imaging scan of the fire area after the fire extinguishing operation is completed, extract the temperature distribution data of the fire area, and determine the presence of reignition when an abnormal heat source point exceeding the preset temperature threshold is detected. It then immediately triggers the short-range response unit to perform supplementary fire extinguishing. After confirming that there is no risk of reignition in the fire area, it calculates the burned area and vegetation damage classification statistics based on the fire perception data collected by the integrated air-ground monitoring unit, thus eliminating the risk of reignition caused by the failure to detect residual heat sources or the failure to completely eliminate them. The report generation subunit is used to store fire scene perception data, calculate the burned area and vegetation damage classification statistics, and summarize all fire scene perception data and work records continuously collected and stored throughout the entire firefighting operation to generate a firefighting operation record. All fire scene perception data includes early warning information at the moment of fire discovery, visible light video and infrared thermal image video recorded by reconnaissance drones throughout the process, and scanning detection records of fixed monitoring nodes to generate a structured firefighting operation report. At the same time, the work records include the flight path planning log of the short-range firefighting control unit, the bomb drop time and the number of fire extinguishing bombs used, as well as the time when the cluster rocket fire extinguishing vehicle entered the battle position, the launch time of each rocket, the time of salvo completion, and ammunition resupply records, so as to facilitate the retrieval or evaluation of firefighting records at any time.

[0028] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cluster of intelligent forest and grassland firefighting robots that coordinate air, space, and ground operations, characterized in that: It includes an integrated air-ground monitoring unit, a short-range fire suppression control unit, a decision generation unit, a long-range fire suppression control unit, and a post-effect assessment unit; The integrated air-space-ground monitoring unit is used to monitor the fire situation in the target area through the collaboration of fixed monitoring nodes, mobile monitoring platforms and space-based monitoring. After confirming the fire, it issues an early warning signal, locates the fire location information, and continuously collects fire scene perception data. The short-range fire extinguishing control unit is used to activate the intelligent command system after receiving the early warning signal, automatically plan the flight route of the fire extinguishing drone to the fire point, execute the fire extinguishing bomb drop command on the fire extinguishing drone to the fire point, carry out fire extinguishing operations, and obtain fire extinguishing effect information. The decision generation unit is used to receive fire extinguishing effect information and, when it is determined that the fire is not under control, generate a decision instruction containing global situation information and fire extinguishing strategy. The remote fire suppression control unit is used to receive the decision command and drive the cluster rocket fire suppression system to launch multiple fire extinguishing projectiles based on the fire location information to carry out remote fire suppression operations. The aftereffect assessment unit is used to detect fire reignition after the fire extinguishing operation is completed, and decide whether to trigger supplementary fire extinguishing based on the detection results. If so, the short-range fire extinguishing control unit is immediately driven to perform secondary fire extinguishing to completely eliminate hidden dangers and generate fire extinguishing action records.

2. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 1, characterized in that: The integrated air-space-ground monitoring unit includes a fixed monitoring subunit, an aerial patrol subunit, and a space-based monitoring subunit. The fixed monitoring subunit is a panoramic photoelectric alarm system deployed at fixed nodes in the forest area. It is used to continuously and automatically scan and detect the surrounding area using infrared thermal imaging detection components, identify infrared data, output fire perception data after detecting a heat source, generate early warning signals and fire location information, and use visible light high-definition camera components to capture images in the direction of the target to generate visible light images of the heat source area. The aerial patrol subunit consists of at least one reconnaissance drone, which is equipped with an electro-optical pod to patrol the target area according to a preset route, acquire fire scene perception data including high-definition images and thermal images, identify targets that match the characteristics of a fire, output early warning signals and fire location information, and continuously perform fire scene video acquisition tasks over the fire scene until the firefighting operation ends or the reconnaissance drone reaches its limit of endurance and needs to return to its base for battery replacement. After receiving the early warning signal from the fixed monitoring subunit, it flies to the fire location to conduct close-range confirmation and high-definition image acquisition. The space-based monitoring subunit is used to access space-based remote sensing satellite data to obtain macroscopic fire field perception data of the target area, identify targets that meet the characteristics of the fire, and output early warning signals and fire location information.

3. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 2, characterized in that: The integrated air-space-ground monitoring unit also includes a cross-verification module, which is used to receive fire scene perception data output by the fixed monitoring subunit, the aerial patrol subunit, and the space-based monitoring subunit. If a certain fire scene perception data is continuously output inconsistent with the other two fire scene perception data, the fire scene perception data will be automatically marked as abnormal, an equipment status alarm will be generated, and the alarm will be pushed to the maintenance personnel.

4. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 3, characterized in that: The fixed monitoring subunit, aerial patrol subunit, and space-based monitoring subunit are connected to the command center. When outputting early warning signals, they report fire scene perception data, fire location information, and fire video to the command center within seconds through the forest and grassland emergency communication network. The command center includes the 119 command center and the emergency command center.

5. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 4, characterized in that: The short-range fire suppression control unit includes a flight path planning subunit, a bomb release control subunit, and an effect recognition subunit. The flight path planning subunit is configured to acquire the fire location information contained in the early warning signal, and automatically plan the optimal flight path of the fire-fighting drone from its current location to the fire location based on GIS geographic information data and digital elevation model. The optimal flight path includes obstacle avoidance path and no-fly zone avoidance path. The bombing control subunit is used to acquire real-time images of the fire scene through the optoelectronic pod carried on the fire-fighting drone after the fire-fighting drone arrives over the fire scene, identify the core area of ​​the fire scene, and control the timing and location of the fire-fighting bombs. The effect recognition subunit is used to identify the fire extinguishing effect based on the fire scene image transmitted back by the photoelectric pod after the fire extinguishing bomb is dropped, and generate fire extinguishing effect information. If the highest temperature in the core area of ​​the fire drops below the preset fire extinguishing temperature threshold after the bomb is dropped, and no continuous open flame is observed in the visible light image, and there are no thermal anomalies exceeding the temperature threshold in the infrared thermal image, then the fire extinguishing is successful and the fire has been effectively extinguished without further action. Otherwise, the fire is not under control.

6. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 5, characterized in that: The firefighting drone is a rotary-wing drone with vertical take-off and landing capabilities. The rotary-wing drone carries 6 to 8 fire extinguishing bombs, with each fire extinguishing bomb having a charge of 50 kg. The operating radius of the rotary-wing drone is not less than 50 km, and the area controlled by a single bomb drop is not less than 400 m².

7. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 6, characterized in that: The decision generation unit includes a communication relay subunit, a situation fusion subunit, and an intelligent decision-making subunit: The communication relay subunit is used to control the intelligent command system to arrive at the periphery of the fire and carry out communication relay when it is determined that the fire is not effectively controlled. This enables the fire-fighting equipment around the fire to be interconnected through the self-organizing network communication protocol. The fire-fighting equipment around the fire includes a modular command vehicle, various fire-fighting drones, cluster rocket fire-fighting vehicles, reconnaissance drones, fixed monitoring nodes, and handheld terminals for ground firefighters. The situation fusion subunit is used to receive the fire scene perception data continuously collected by the air-space-ground integrated monitoring unit, including fire video, infrared thermal image data and fire location information, and to perform spatiotemporal registration and fusion processing on the fire scene perception data to generate a panoramic fire scene situation map under a unified spatiotemporal reference. The intelligent decision-making subunit is used to call the forest database and expert database, analyze and process the panoramic fire situation map based on machine learning algorithms, extract the fire boundary contour from the fire video through image segmentation algorithm, extract the temperature anomaly area and temperature distribution cloud map from the infrared thermal image data through temperature threshold method, calculate the three-dimensional coordinates of the fire point from the fire location information, calculate the fire area change rate from the UAV position change data, and generate a fire extinguishing plan as the decision instruction. The fire extinguishing plan includes launch position planning and safe firing range planning.

8. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 7, characterized in that: The intelligent command system is a modular command vehicle, which is equipped with a command and management seat, a collaborative control seat, a drone control seat, and an observation and command seat. The modular command vehicle has a seating capacity of 9 people.

9. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 8, characterized in that: The cluster rocket fire extinguishing system is mounted on a high-mobility off-road chassis, forming a cluster rocket fire extinguishing vehicle, and is equipped with a launcher for 40 fire extinguishing bombs; The remote fire extinguishing control unit automatically calculates the firing azimuth and elevation angles based on the launch position planning and safe firing range planning in the decision command, and controls the azimuth and elevation angles of the launcher through the hydraulic elevation mechanism and the hydraulic rotation mechanism to drive the fire extinguishing shells to fire simultaneously. The fire extinguishing agent coverage and control area of ​​a single salvo is not less than 10,000 m².

10. The air-space-ground collaborative intelligent forest and grassland firefighting robot cluster according to claim 9, characterized in that: The post-ignition management unit includes a reignition detection subunit, a report generation subunit, and a data archiving subunit: The reignition detection subunit is used to control the integrated air-ground monitoring unit to perform a full-coverage thermal imaging scan of the fire area after the fire extinguishing operation is completed, extract the temperature distribution data of the fire area, determine the presence of reignition when an abnormal heat source point exceeding the preset temperature threshold is detected, and immediately trigger the near-field response unit to perform supplementary fire extinguishing. After confirming that there is no risk of reignition in the fire area, the burned area and vegetation damage classification statistics are calculated based on the fire perception data collected by the integrated air-ground monitoring unit. The report generation subunit is used to store fire scene perception data, calculate the burned area and vegetation damage classification statistics, and summarize all fire scene perception data and work records continuously collected and stored throughout the entire firefighting operation to generate a firefighting operation record. All fire scene perception data includes early warning information at the moment of fire discovery, visible light video and infrared thermal image video recorded by reconnaissance drones throughout the process, and scanning detection records of fixed monitoring nodes to generate a structured firefighting operation report. At the same time, the work record includes the flight path planning log of the short-range firefighting control unit, the bomb drop time and the number of fire extinguishing bombs used, as well as the time when the cluster rocket fire extinguishing vehicle enters the battle position, the launch time of each rocket, the time of salvo completion, and ammunition resupply records.

Images