A high-rise building intelligent fire-fighting system and method with built-in autonomous mobile fire-fighting body

Abstract

The application relates to a high-rise building intelligent fire-fighting system and method with built-in autonomous mobile fire-fighting bodies. The system comprises: a three-dimensional fire-fighting track network module, which is pre-installed in the building and provides a moving channel, energy and fire extinguishing agent for the autonomous mobile fire-fighting bodies; a distributed sensing and fire location module, which collects fire information by using a key node sensor group; a central intelligent decision and cluster scheduling module, which receives information based on a digital twin model, generates a fire-fighting scheme containing scheduling instructions through multi-agent collaborative path planning, and generates a fire-fighting scheme containing scheduling instructions; an autonomous mobile fire-fighting body execution module, which contains at least one autonomous mobile fire-fighting body, and the autonomous mobile fire-fighting body moves on the track network and extinguishes fire according to the scheme; and a man-machine interaction and collaborative command module, which displays fire scene information to fire-fighting personnel through a control center and receives manual instructions. The modules are cooperatively connected to build an intelligent fire-fighting system, solve problems such as traditional fire-fighting response lag, high risk of internal attack, and inaccurate delivery of fire extinguishing agent, and realize the transformation from passive response to active intelligence in fire prevention and control.

Description

Technical Field

[0001] This invention relates to the field of smart fire protection and building safety technology, and in particular to a smart fire protection system and method for high-rise buildings with built-in autonomous mobile fire-fighting bodies. Background Technology

[0002] Existing technological systems generally face a dual dilemma in responding to high-rise building fires: limited external reach and delayed internal response. On the one hand, external rescue efforts are constrained by equipment operating height, site conditions, and building barriers, making it difficult to quickly reach the fire at the top of the building. On the other hand, internal fire protection systems are mostly passively triggered, lacking intelligent sensing and proactive intervention capabilities in the early stages of a fire, leading to missed critical rescue opportunities. Current mainstream solutions and their limitations are as follows:

[0003] (I) Traditional Fire Hydrants and Manual Operation: Although fire hydrants are widely installed inside buildings, they are entirely passive facilities, relying on personnel to detect the fire, arrive at the scene, and perform professional operation. Under complex conditions such as panic, dense smoke, and power outages caused by a fire, non-professionals face difficulties in operation, and firefighters' internal movement is time-consuming, easily missing the golden window for extinguishing initial fires. This system inherently lacks the function of "early warning and immediate response."

[0004] (ii) External aerial rescue equipment: Aerial rescue equipment, represented by ladder trucks and high-pressure water cannons, is the main external rescue method for fire brigades. Its operating height is limited by multiple factors such as equipment limits, site conditions, and weather conditions, and it often cannot effectively cover buildings over 100 meters high.

[0005] (iii) Firefighting drones: These are mostly used for external fire reconnaissance and small-scale firefighting assistance. They suffer from limited payload, insufficient endurance, and weak anti-interference capabilities. Furthermore, they cannot penetrate the building's external structure to enter the interior to perform missions. Currently, they can only serve as a supplementary means and are unable to undertake core firefighting functions.

[0006] (iv) Fixed fire protection systems in buildings: such as automatic sprinkler systems, which activate to spray water after the temperature reaches a set threshold via ceiling sprinklers. They have low response sensitivity, fixed coverage area, and cannot dynamically adjust the spray direction, making it difficult to achieve precise suppression of fire sources.

[0007] In summary, the existing fire protection technology system still suffers from key shortcomings in high-rise building fire scenarios, including delayed response, insufficient coverage, and low flexibility. Therefore, there is a need to develop an intelligent fire protection method that can be embedded within buildings, possesses autonomous sensing and mobile execution capabilities, and can achieve precise fire suppression in the early stages of a fire. Summary of the Invention

[0008] In view of the shortcomings of the prior art, the purpose of this invention is to provide an intelligent fire protection system and method for high-rise buildings with built-in autonomous mobile fire-fighting bodies, so as to solve one or more problems in the prior art.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows:

[0010] A high-rise building intelligent fire protection system with built-in autonomous mobile fire-fighting unit, the system includes:

[0011] The three-dimensional fire-fighting track network module, which is pre-installed inside the building, provides a three-dimensional movement channel and energy and extinguishing agent delivery for autonomous mobile fire-fighting bodies.

[0012] The distributed sensing and fire location module includes a sensor group deployed at key nodes of the three-dimensional fire track network to collect fire information;

[0013] The central intelligent decision-making and cluster dispatch module receives the fire information based on a digital twin model and generates a fire-fighting plan that includes dispatch instructions for autonomous mobile fire-fighting units based on multi-agent collaborative path planning.

[0014] The autonomous mobile firefighting unit execution module includes at least one autonomous mobile firefighting unit, which can move autonomously on the three-dimensional firefighting track network and perform firefighting tasks according to the firefighting plan;

[0015] The human-computer interaction and collaborative command module is configured to display fire scene information to firefighters through the control center and receive manual commands.

[0016] Furthermore, the three-dimensional fire-fighting track network module includes:

[0017] A dedicated fire shaft pre-installed within the building serves as a main vertical transport route;

[0018] A horizontal lightweight track network installed in the ceilings or inside the walls of public areas on each floor;

[0019] Water supply mains, power supply lines, and data communication bus are integrated into the dedicated fire shaft and horizontal light rail network;

[0020] The fast-connection base stations, located at key nodes, are configured to have physical locking, automatic water and charging, and data interaction. Key nodes include stairwells on each floor and outside equipment rooms.

[0021] Furthermore, the distributed sensing and fire location module constructs a distributed sensing network based on the fusion of signals from the building's built-in fire alarm system and sensor group information. The sensor group includes smoke and temperature detectors, thermal imaging cameras, multispectral flame detectors, and gas composition sensors.

[0022] Furthermore, the firefighting plan is designed based on fire information and the real-time status of autonomous mobile firefighting units. The plan includes the dispatch of autonomous mobile firefighting units, route selection, task allocation, and dynamic adjustment of building equipment.

[0023] Furthermore, the autonomous mobile fire-fighting system structure includes:

[0024] The mobile chassis is configured to move along three-dimensional fire-fighting tracks and connect to base stations;

[0025] The task module interface is configured to mount different task modules.

[0026] The breaching and entry module is configured to enable the removal of obstacles in a fire and rapid entry into building structures.

[0027] The extinguishing agent spraying module is configured to achieve precise spraying of different types of extinguishing agents;

[0028] The reconnaissance relay module is configured to achieve real-time image acquisition, environmental parameter monitoring, and communication signal relay of the fire scene environment;

[0029] The local controller is configured to accept commands from the control center and has edge computing capabilities.

[0030] Furthermore, the execution steps of the autonomous mobile fire-fighting unit in carrying out fire-fighting tasks are as follows:

[0031] Deployment and Autonomous Movement: After receiving instructions from the control center, the autonomous mobile firefighting unit activates its chassis to move along a pre-set three-dimensional firefighting track network;

[0032] Obstacle removal and entry: In confined space scenarios, the obstacle removal module is used to perform the removal of obstacles.

[0033] Fire reconnaissance and environmental awareness: Fire information is collected through the reconnaissance relay module, and it interacts with the human-machine interaction and collaborative command module as a communication relay node;

[0034] Precise fire suppression and sustained fire control: The fire extinguishing agent spraying model selects the appropriate fire extinguishing method based on the fire situation information until the fire is extinguished;

[0035] Mission Completion and Autonomous Return: After the fire is extinguished, the autonomous mobile firefighting unit returns to the nearest or designated base station to resupply and resume standby status.

[0036] Furthermore, the human-computer interaction and collaborative command module receives manual instructions in the following modes:

[0037] Supervision mode: Firefighting tasks are deployed based on the fire development simulated by the digital twin model;

[0038] Remote control mode: In complex situations, it allows direct control of one or more autonomous mobile fire-fighting units for operation;

[0039] Collaborative mode: The system plans a safe path and adopts a collaborative approach where robots perform tasks first, followed by personnel.

[0040] A fire-fighting method, applied to the aforementioned intelligent fire-fighting system for high-rise buildings with a built-in autonomous mobile fire-fighting unit, the method comprising the following steps:

[0041] S1. After a fire occurs, the distributed sensing and fire location module detects and analyzes the fire situation and collects fire information based on the sensor network of the sensor group.

[0042] S2, the central intelligent decision-making and cluster dispatch module generates fire-fighting plans based on the digital twin model and multi-agent collaborative path planning according to fire information and dispatches the decision to the autonomous mobile fire-fighting body execution module;

[0043] S3, the autonomous mobile fire-fighting unit execution module dispatches the fire-fighting unit to move along the track network and perform fire-fighting tasks based on the received fire information;

[0044] S4. During the firefighting process, the human-machine interaction and collaborative command module conducts human-machine collaborative command based on scheduling decisions, realizing status monitoring and dynamic adjustment until the fire is brought under control.

[0045] Furthermore, the multi-agent cooperative path planning includes the following steps:

[0046] Task allocation: Based on fire information and fire protection system status processing, including task decomposition and classification, as well as matching and allocation;

[0047] Global path: Based on task allocation results, 3D fire track network topology, and real-time fire dynamic weight processing, including track network modeling and single-machine path generation;

[0048] Local adjustments: By setting the local controller of the fire protection system to enable autonomous decision-making in multiple scenarios, including when encountering temporary obstacles, when multiple machines converge, and when the fire situation changes suddenly.

[0049] Furthermore, the single-machine path generation is based on The algorithm evaluates the priority of each node using a cost function to obtain the optimal path, as shown in the following formula:

[0050]

[0051] In the formula: The actual cost represents the distance from the starting point to the current node. The travel time already consumed; As a heuristic function, it estimates the value starting from the current node. The time still needed to reach the destination; This is a dynamic penalty term, reflecting the process from the starting point to the node. The additional risk costs incurred due to passing through areas with high temperatures or dense smoke along the route.

[0052] Compared with the prior art, the beneficial technical effects of the present invention are as follows:

[0053] (i) The system of the present invention is pre-installed with a three-dimensional fire-fighting track network module. With a dedicated fire-fighting shaft as the vertical main road and the horizontal light tracks on each floor as the branch network, it constructs a full-area mobile channel for autonomous mobile fire-fighting bodies that is not limited by building height and external site conditions, effectively solving the problem that traditional external rescue equipment cannot form effective coverage for high-rise buildings above 100 meters.

[0054] (ii) The present invention deploys a distributed sensing and fire location module, integrating smoke and temperature detectors, thermal imaging cameras, multispectral flame detectors and gas composition sensors at key nodes of the track network. Through multi-source information fusion, it achieves ultra-early sensing and three-dimensional precise location of fire, overcoming the passive response mode of traditional fire hydrants that require personnel to arrive on site for operation and automatic sprinkler systems that have low response sensitivity and fixed coverage.

[0055] (III) The system design of this invention includes a central intelligent decision-making and cluster scheduling module. Based on a digital twin model and multi-agent collaborative path planning, it automatically generates a fire-fighting plan that includes dispatch, path, task allocation and building equipment linkage according to fire information and real-time status of fire-fighting equipment. This breaks through the technical limitations of traditional fixed fire-fighting systems, which cannot dynamically adjust the spray direction and are difficult to achieve precise fire suppression.

[0056] (iv) The system of the present invention is equipped with an autonomous mobile fire-fighting execution module, which integrates modular functions such as mobile chassis, demolition and entry, fire extinguishing agent spraying and reconnaissance relay. It can quickly move to the fire scene along the track network to perform composite tasks such as obstacle removal, precise fire extinguishing, environmental reconnaissance and communication relay. It effectively solves the application pain points of traditional fire-fighting drones, such as limited payload, insufficient endurance and inability to enter the interior of buildings to perform core fire-fighting functions.

[0057] (v) The system of the present invention is equipped with a human-computer interaction and collaborative command module. It achieves efficient cooperation between command personnel and robots through three modes: supervision, remote control and collaboration. The system continuously records human intervention data and optimizes subsequent decision-making models through reinforcement learning, forming a tiered combat pattern of "robot attack and personnel clean up the aftermath", which significantly improves the proactive intervention capability and overall prevention and control level in the initial stage of high-rise building fires. Attached Figure Description

[0058] Figure 1The diagram shows a system architecture schematic of an intelligent fire protection system and method for high-rise buildings with built-in autonomous mobile fire-fighting bodies, according to an embodiment of the present invention.

[0059] Figure 2 The diagram illustrates a process flow of an intelligent fire protection system and method for high-rise buildings with a built-in autonomous mobile fire-fighting body, according to an embodiment of the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a further detailed explanation of the intelligent fire protection system and method for high-rise buildings with a built-in autonomous mobile fire-fighting body. The advantages and features of this invention will become clearer from the following description. It should be noted that the accompanying drawings are in a very simplified form and use non-precise proportions, used only to facilitate and clearly illustrate the purpose of the embodiments of this invention. Please refer to the accompanying drawings to make the objectives, features, and advantages of this invention more apparent and understandable. It should be understood that the structures, proportions, sizes, etc., depicted in the accompanying drawings are only for illustrative purposes to aid those skilled in the art and are not intended to limit the implementation conditions of this invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in proportions, or adjustments to the size, without affecting the effects and objectives achieved by this invention, should still fall within the scope of the technical content disclosed in this invention.

[0061] Please refer to the following: Figure 1 and Figure 2 The intelligent fire protection system for high-rise buildings with built-in autonomous mobile fire-fighting units in this embodiment includes:

[0062] The 3D fire-fighting track network module, pre-installed inside the building, provides a three-dimensional movement channel and energy and extinguishing agent delivery for autonomous mobile fire-fighting units. As the physical framework of the system, its core function is to construct a stable and reliable energy and extinguishing agent delivery channel. This module ensures that the autonomous mobile fire-fighting units can achieve rapid, reliable mobility and continuous operation throughout the entire building's three-dimensional space. Specifically, the 3D fire-fighting track network module includes:

[0063] a. A dedicated fire shaft pre-installed within the building serves as the main vertical access route. Preferably, the shaft's cross-sectional dimensions are determined based on the building height and fire protection system specifications. An internal double-track structure supports two-way passage, and the shaft walls are encased in high-fire-resistance fire-resistant materials and reliably connected to the main building structure.

[0064] b. A horizontal lightweight track network installed within the ceilings or walls of public areas on each floor. Preferably, the horizontal tracks adopt a modular design, with pre-installed quick-access interfaces on the track sides for local expansion or maintenance and replacement. The track layout covers key areas such as corridors, anterooms, and the perimeter of equipment rooms, forming a horizontal mobile network that connects to dedicated fire shafts.

[0065] c. Water supply mains, power supply lines, and data communication buses integrated into dedicated fire shafts and horizontal light rail networks. Preferably, the water supply mains are arranged in a ring, with the pipe diameter determined based on the building's fire water demand. Pressure reducing and stabilizing devices are installed at the ends of the mains to accommodate the pressure requirements of different floors. The power supply lines adopt a dual-circuit design to ensure continuous power supply to the fire protection system. The data communication bus uses a redundant industrial Ethernet design, supporting real-time bidirectional transmission of fire data, control commands, and video streams.

[0066] d. Rapid connection base stations, located at key nodes, are configured with physical locking, automatic water and charging, and data exchange functions. As energy supply stations and mission starting points for the fire-fighting system, the deployment locations of these base stations are optimized to prioritize coverage of stairwells on each floor, the exterior of equipment rooms, and key fire spread paths. Preferably, the base stations are equipped with standardized mechanical interfaces to enable rapid connection between the fire-fighting system and water supply mains and power lines. The base stations have built-in intelligent locking mechanisms to physically secure the fire-fighting system when not in mission mode, preventing unauthorized movement or accidental slippage. The base stations are also equipped with level sensors and power monitoring modules to automatically trigger the water and charging process, ensuring the fire-fighting system is always in full readiness.

[0067] Distributed Sensing and Fire Location Module: This module includes sensor arrays deployed at key nodes of the three-dimensional fire track network to collect fire information. This module constitutes the system's "neural network," constructing a distributed sensing network based on the fusion of signals from the building's built-in fire alarm system and sensor array information. The sensor array includes smoke and heat detectors, thermal imaging cameras, multispectral flame detectors, and gas composition sensors. Simultaneously, it integrates signals from the building's own fire alarm system to build a distributed sensing network. Through sensor array information fusion, the system can achieve ultra-early detection of fires and preliminary judgment of precise location, providing the decision-making module with real-time, multi-dimensional battlefield situation information. Preferably, smoke and heat detectors are deployed at track network nodes and conventional fire-fighting points in the building to detect abnormal changes in smoke concentration and ambient temperature. Thermal imaging cameras are installed at track bends and shaft entrances / exits, using infrared thermal imaging technology to penetrate dense smoke and obtain the temperature distribution and spread trend of the fire source. Multispectral flame detectors are deployed in key protection areas, using specific spectral characteristics to identify open flames and eliminate interfering light sources. Gas composition sensors monitor the concentration of characteristic fire gases such as carbon monoxide and volatile organic compounds, assisting in judging the nature of combustibles and the stage of fire development. Data collected by each sensor is aggregated in real time via a data communication bus, and after preliminary fusion processing by edge computing nodes, it is uploaded to the central intelligent decision-making and cluster scheduling module.

[0068] The Central Intelligent Decision-Making and Cluster Scheduling Module receives fire information based on a publicly known digital twin model and generates a firefighting plan containing dispatch instructions for autonomous mobile firefighters based on multi-agent collaborative path planning. This module is the "brain" of the system. Running on a building model built using publicly known digital twin technology, it first simulates the fire's development within the digital twin after receiving data from the distributed sensing and fire location module. Subsequently, based on the fire's location, type, intensity, and the real-time status (location, agent type, and power) of each main mobile firefighter, it calculates the optimal firefighting plan through multi-agent collaborative path planning. The firefighting plan is designed based on fire information and the real-time status of the autonomous mobile firefighters. The plan includes the dispatch of autonomous mobile firefighters (which autonomous mobile firefighters to dispatch), path selection (choosing the optimal path), task allocation (e.g., No. 1 breaches the door, No. 2 sprays extinguishing agent, and No. 3 is responsible for cooling and isolation), and dynamic adjustments to building equipment (e.g., closing ventilation, opening smoke exhaust windows, and forcing elevators to land) to create favorable conditions.

[0069] Autonomous Mobile Firefighting Unit Execution Module: This module includes at least one autonomous mobile firefighting unit capable of moving autonomously on a three-dimensional firefighting track network and performing firefighting tasks according to the firefighting plan. The autonomous mobile firefighting unit structure includes:

[0070] The mobile chassis is configured to move along the three-dimensional fire-fighting track and connect to the base station. Preferably, the mobile chassis adopts a wheel-rail hybrid drive structure, equipped with a permanent magnet synchronous motor and a high-precision encoder to achieve precise positioning and smooth operation within the track network. The chassis is designed with an active guidance mechanism that can adapt to changes in track curvature and achieve smooth switching between vertical and horizontal sections of the shaft. The chassis bottom integrates an electromagnetic braking device and a mechanical locking mechanism for dual safety protection, ensuring reliable stopping in the event of a sudden power outage or emergency. Guide wheel sets and shock-absorbing suspension systems are configured on both sides of the chassis to reduce vibration and noise during high-speed movement and extend the service life of the equipment.

[0071] The mission module interface is configured to mount different mission modules. The interface employs a standardized mechanical connection and electrical docking design, supporting rapid replacement of breaching and entry modules, fire extinguishing agent spraying modules, and reconnaissance relay modules. The mechanical interface ensures the rigidity and impact resistance of the module connections. The electrical interface integrates power supply, control signals, and data communication into a single connector, enabling plug-and-play functionality and automatic module identification.

[0072] The breaching and entry module is configured to quickly breach obstacles in a fire and enter building structures. For confined spaces or obstacle-blocked scenarios, the module is equipped with a multi-degree-of-freedom robotic arm, the end of which can be quickly fitted with tools such as hydraulic shears, impact drills, or thermal cutting torches. Preferably, the module has a built-in force feedback sensor to monitor breaching resistance in real time and adjust operating parameters to avoid excessive damage to the building's load-bearing structure. After breaching, the module can deploy foldable support legs to provide a stable platform for subsequent firefighting operations.

[0073] Extinguishing Agent Spraying Module: Configured to achieve precise spraying of different types of extinguishing agents. Preferably, the module is equipped with a rotatable and pitchable fire monitor at the front end, with the jet pattern infinitely adjustable between direct current, mist, and spray patterns, and the maximum range optimized according to the building's floor height. For electrical fires or protected areas containing valuable equipment, the module can be fitted with high-pressure fine water mist nozzles, achieving efficient cooling and suffocation extinguishing through micron-level droplets while minimizing water damage. For oil or chemical fires, the module has a built-in foam proportioner that automatically mixes 3% or 6% aqueous film-forming foam concentrate. Dry powder and gaseous extinguishing agent storage tanks are manufactured to pressure vessel standards and equipped with a dual-start mechanism of electric explosion valve and solenoid valve to ensure reliable spraying in extreme environments. The spray angle and flow rate are calculated in real time by a local controller based on fire information, combined with fine-tuning of the motorized chassis position to achieve precise coverage of the fire source.

[0074] The reconnaissance relay module is configured to acquire real-time images of the fire scene, monitor environmental parameters, and relay communication signals. Preferably, the reconnaissance relay module integrates a high dynamic range panoramic camera and a long-wave infrared thermal imager, supporting simultaneous transmission of dual-channel video from visible light and thermal imaging. The environmental sensor group includes temperature, humidity, toxic gas concentration, and oxygen content monitoring units, with a data sampling frequency of no less than 10Hz to ensure real-time performance. The communication relay equipment employs Mesh self-organizing network technology to automatically construct multi-hop transmission links in the complex electromagnetic environment of the fire scene, extending the communication coverage between the control center and the frontline firefighting units.

[0075] The local controller is configured to receive commands from the control center and possess edge computing capabilities. Preferably, the local controller employs an industrial-grade embedded computing platform, equipped with a multi-core ARM processor and GPU acceleration unit, supporting the local deployment of deep learning inference models. The controller parses task commands from the control center in real time and, based on fire scene data collected by the reconnaissance relay module, performs edge computing tasks such as fire source identification, spread prediction, and spray parameter optimization. When the communication link with the control center is interrupted, the local controller can independently execute critical operations such as emergency evacuation and fire source suppression according to preset autonomous decision-making rules, ensuring the survivability and mission continuity of the firefighting equipment in extreme situations.

[0076] Furthermore, the execution steps for autonomous mobile firefighting units to carry out firefighting tasks are as follows:

[0077] Deployment and Autonomous Movement: After receiving instructions from the control center, the autonomous mobile firefighting unit activates its chassis to move along a pre-set three-dimensional firefighting track network. Based on the path planning instructions issued by the control center and the topology of the three-dimensional firefighting track network, the chassis moves along the optimal route. During movement, the autonomous mobile firefighting unit continuously transmits position, speed, power level, and equipment status data back to the control center, supporting global situational awareness monitoring.

[0078] Obstacle removal and entry: In confined space scenarios, the demolition and entry module performs obstacle removal. The module first obtains obstacle material and structural information through a reconnaissance relay module. The local controller identifies obstacle types (such as wooden doors, metal roller shutters, and plasterboard partitions) based on a pre-trained neural network model and automatically matches the optimal demolition strategy and tool combination. The robotic arm adjusts its working posture in real time based on force feedback data, strictly controlling the impact load on surrounding building structures while ensuring demolition efficiency.

[0079] Fire reconnaissance and environmental awareness: Fire information is collected through a reconnaissance relay module, which also acts as a communication relay node to interact with the human-machine interface and collaborative command module. Preferably, the reconnaissance relay module automatically performs identification processing after the firefighters arrive at the target area, a panoramic camera collects visible light images, and a thermal imager simultaneously generates a temperature field distribution map. Environmental parameter monitoring includes temperature, humidity, carbon monoxide concentration, oxygen content, and volatile organic compound concentration. When any parameter exceeds a safety threshold, an audible and visual alarm is immediately triggered, and a warning message is pushed to the control center. The communication relay equipment assesses the surrounding signal strength in real time and dynamically adjusts the transmission power and frequency hopping strategy to ensure a stable data link with the control center in environments with dense smoke, high temperatures, and electromagnetic interference. Precise fire suppression and continuous fire control: The fire extinguishing agent spraying model selects the corresponding fire extinguishing method based on fire information until the fire is extinguished. The fire extinguishing agent spraying module automatically selects the optimal fire extinguishing strategy based on multi-source fire data. For example, for fires involving solid combustibles, the fire monitor switches to a direct current jet mode to penetrate the combustion layer with high-momentum water flow for deep cooling. For fires involving liquid fuels, foam mixing and spraying are employed to form an isolation film on the burning surface, blocking oxygen supply. During the spraying process, a thermal imager tracks the temperature changes of the fire source in real time, and the autonomous mobile fire-fighting unit dynamically adjusts the jet angle and flow rate to ensure that the extinguishing agent accurately covers the core area of ​​the fire source and avoids excessive spraying that could cause secondary damage.

[0080] Mission Completion and Autonomous Return: After extinguishing the fire, the autonomous mobile firefighting unit returns to the nearest or designated base station for resupply and resumes standby status. Once it is confirmed that there is no risk of reignition, the firefighting unit autonomously returns to the nearest base station along a safe path, maintaining the alert status of the reconnaissance relay module en route to guard against sudden changes in the fire situation. Upon arrival at the base station, the autonomous firefighting unit automatically docks with the base station, triggering intelligent locking, water and charging, and data synchronization processes. After mission completion, the firefighting unit uploads sensor data, operation logs, and equipment status information from the entire mission to the central intelligent decision-making and cluster scheduling module, supporting iterative optimization of the digital twin model and post-battle analysis. After resupply, the firefighting unit returns to full combat readiness, awaiting the next mission dispatch.

[0081] The Human-Computer Interaction and Collaborative Command Module is configured to display fire scene information to firefighters through the control center and receive manual commands. Serving as a crucial bridge between the system and frontline commanders, its design fully considers the complex information and urgent decision-making characteristics of high-rise building fires. Deployed in the fire control center and mobile command terminals, this module employs a multi-screen interactive display architecture. The main screen presents a 3D fire scene based on a digital twin model, including fire spread simulation, real-time location of fire-fighting equipment, building equipment status, and personnel evacuation routes. Auxiliary screens display high-definition video streams, environmental parameter curves, and task execution progress transmitted from their respective mobile fire-fighting units. Commanders can quickly switch perspectives and zoom to view detailed information in specific areas via touch interface or voice commands.

[0082] The human-computer interaction and collaborative command module receives manual instructions in the following modes:

[0083] Supervision Mode: Firefighting tasks are deployed based on the fire development simulated by a digital twin model. Specifically, commanders review and adjust the system-generated firefighting plans based on the fire development trends presented in the digital twin model. The system pushes multiple alternative plans in real time, each marked with key indicators such as estimated firefighting time, extinguishing agent consumption, and fire protection equipment damage risk. Commanders can adjust the priority of fire protection equipment deployment or modify the task sequence by dragging and dropping. Once confirmed, the command is issued immediately for execution. This mode is suitable for routine scenarios where fire development is relatively controllable and the system's autonomous decision-making reliability is high. It leverages the rapid computing advantages of artificial intelligence while retaining the final review authority of human decision-makers.

[0084] Remote Control Mode: Allows direct control of one or more autonomous mobile fire suppression systems in complex situations. Specifically, the remote control mode is designed for extremely complex fire environments. When the system's autonomous decision-making deviates due to missing information or sudden changes in the situation, the commander can switch to full manual control with a single button. In this mode, the commander can precisely control the movement speed, spray angle, and flow rate of a single fire suppression system, or simultaneously coordinate multiple fire suppression systems to form tactical formations such as cross-jet or tiered advances. The system displays real-time safety boundary indicators such as the building structure load limit, remaining battery power of the fire suppression system, and extinguishing agent reserves to prevent manual operation from exceeding the equipment's capabilities.

[0085] Collaborative Mode: The system plans safe entry routes and employs a collaborative approach, with robots leading the way and personnel following. Specifically, by integrating the advantages of autonomous robot operation and on-site personnel response, the system first plans a relatively safe entry route based on real-time fire data, marking potential structural risk areas and toxic gas concentration distributions. Autonomous mobile firefighting units advance along the route as advance units, performing tasks such as fire suppression, passage opening, and situational reconnaissance. The environmental data they transmit continuously updates the safety route assessment. Firefighters in the rear standby area receive real-time situational briefings from the system. Once the robots confirm that the fire in a localized area is under control, the structure is stable, and the air quality meets standards, the system issues a personnel entry permit. After personnel enter, wearable devices establish a short-range data link with the firefighting units, enabling location sharing and dynamic guidance for emergency evacuation routes, forming a tiered operational structure of "robot attack and personnel cleanup."

[0086] By switching and combining the above three modes, the system continuously records the intervention operations of the command personnel and the corresponding fire handling effects. Through reinforcement learning mechanisms, it optimizes the subsequent autonomous decision-making model, forming a continuous evolutionary closed loop of complementary human and machine capabilities.

[0087] A fire protection method, applied to the aforementioned intelligent fire protection system for high-rise buildings with built-in autonomous mobile fire-fighting units, includes the following steps:

[0088] S1. After a fire occurs, the distributed sensing and fire location module detects and analyzes the fire situation based on the sensor network of the sensor group and collects fire information.

[0089] In the initial stage of a fire, sensor arrays deployed at key nodes of the 3D fire protection track network and at conventional fire-fighting points in the building immediately trigger a response. Smoke and temperature detectors detect smoke particle concentrations in the air (such as abnormal light attenuation) or ambient temperatures exceeding preset thresholds. Simultaneously, multispectral flame detectors capture the unique ultraviolet / infrared radiation spectrum of flames, eliminating interference sources such as sunlight and lighting. Gas composition sensors detect a sharp increase in the concentration of characteristic fire gases such as carbon monoxide, carbon dioxide, and hydrogen cyanide. Thermal imaging cameras penetrate the initial smoke and obtain a precise temperature distribution map of the core area of ​​the fire source. The raw data collected by each sensor is aggregated in real time to the floor edge computing node via a data communication bus, where multi-source data is cross-validated, false alarms are eliminated, and combined with spatial coordinates provided by the building information model, the fire source is precisely located in three dimensions within the digital twin model. Finally, fire information including the fire source location, predicted combustible materials, and fire intensity level is packaged and sent to the central intelligent decision-making module.

[0090] S2, the Central Intelligent Decision-Making and Cluster Scheduling Module, based on a digital twin model and considering fire information and multi-agent collaborative path planning, generates firefighting plans and schedules decisions to the autonomous mobile firefighting unit execution module. Upon receiving fire information, the Central Intelligent Decision-Making and Cluster Scheduling Module first synchronizes the fire data to the digital twin model. Then, the module retrieves real-time status data from all autonomous mobile firefighting units (including base station location, current battery level, extinguishing agent type and remaining quantity, and equipment health), and initiates multi-agent collaborative path planning. With the optimization objectives of minimizing total firefighting time, minimizing secondary damage, and ensuring personnel evacuation routes, it performs multi-objective optimization calculations to generate a detailed firefighting plan. After completing the decision, the module distributes the instruction set to the designated autonomous mobile firefighting units via a redundant data bus.

[0091] Furthermore, the multi-agent cooperative path planning includes the following steps:

[0092] Step 1: Task Allocation. The central intelligent decision-making and cluster scheduling module processes the fire information and the status of fire-fighting equipment. The fire information includes the location of the fire source, fire intensity, and combustion type. The status of the fire-fighting equipment includes the location, power level, extinguishing agent type, and remaining quantity of each fire-fighting unit. Details are as follows:

[0093] Task Decomposition and Priority Classification: Based on the collected fire information, the central intelligent decision-making and cluster scheduling module sets the fire fighting plan into three priorities. Level 1 is for emergency firefighting tasks, such as direct suppression of the fire source. Level 2 is for fire control tasks, such as blocking the spread of the fire. Level 3 is for auxiliary support tasks, such as reconnaissance, cooling and isolation, and communication relay.

[0094] Matching and Assignment: Prioritize assigning Level 1 and Level 2 tasks to the nearest firefighting units with matching resources. When multiple units need to coordinate for the same task, such as simultaneous demolition and firefighting, the principle of proximity-based combination will be adopted. After the critical tasks are assigned, the remaining firefighting units will stand by.

[0095] By decomposing and classifying the tasks, as well as matching and assigning them, the specific tasks performed by each fire protection unit are output, providing input conditions for subsequent path planning.

[0096] Step 2: Global Path. Based on the task allocation results, and combined with the 3D fire track network topology and real-time fire situation dynamic weight processing. Details are as follows:

[0097] 3D Track Network Modeling: Abstracting the 3D fire-fighting track system in the actual physical space of a building into a mathematically directed graph structure. , set in Represents a set of nodes. Represents the set of edges.

[0098] node These locations fall into three categories: base stations, which are the starting and supply points for firefighting operations; track intersections, where tracks from different directions converge; and mission points, which are the target locations where firefighting, demolition, or reconnaissance operations need to be carried out.

[0099] side This represents the track segment between two nodes. Each edge records the following attributes: track segment length in meters; nominal passage time, i.e., the time required for a fire truck to pass through the segment at normal speed; and current availability status, i.e., whether the track segment is closed due to a fault or fire.

[0100] Route planning is also implemented based on a dynamic fire situation weighting mechanism. In this embodiment, a passage penalty coefficient is applied to track segments located in high-temperature or dense smoke areas. This coefficient is greater than 1, increasing the equivalent length of these track segments during route calculation, thereby guiding firefighters to prioritize safer routes. The penalty coefficient is directly proportional to the measured temperature and smoke concentration; the higher the temperature and the denser the smoke, the larger the penalty coefficient.

[0101] Standalone path generation:

[0102] The system adopts The algorithm calculates the optimal path from the starting point to the target point for each fire-fighting unit. The algorithm evaluates the priority of each node through a cost function, thereby efficiently finding the optimal path. The cost function is calculated as follows:

[0103]

[0104] In the formula: The actual cost represents the distance from the starting point to the current node. The travel time already consumed is calculated by summing the nominal travel time of each track segment already traversed.

[0105] As a heuristic function, it estimates the value starting from the current node. The remaining time required to reach the target point. To simplify the calculation, this embodiment uses the Manhattan distance divided by the average speed. The Manhattan distance is the shortest path length between two points that can only move along the coordinate axis, and is suitable for scenarios with track network layouts.

[0106] This is a dynamic penalty term, reflecting the process from the starting point to the node. The additional risk costs incurred due to passing through areas of high temperature or dense smoke along the route. The calculation method is to sum the penalty coefficients of each track segment minus 1 and then multiply by the nominal travel time of that segment.

[0107] The algorithm's execution process is as follows: Starting from the starting point, it continuously examines the neighboring nodes of the current node and calculates the value of each neighboring node. Value, prioritize expansion The smallest node is used until the destination is reached. The final generated path is the optimal solution that balances travel time and security risks.

[0108] Through the above-described 3D track network modeling and single-machine path generation processes, the system generates an initial 3D path for each fire-fighting unit to determine how to reach the target area, including the following information:

[0109] A node sequence is a list of track nodes that pass through sequentially from the starting point to the target point.

[0110] Estimated transit time, which is the timestamp of arrival at each node and the total time required to complete the entire path.

[0111] Path cost is used to evaluate the overall merits of a path.

[0112] Step 3: Local Adjustments. This involves configuring the local controller of the fire protection system to enable autonomous decision-making in multiple scenarios, including situations such as encountering temporary obstacles, multiple fire engines converging, and sudden changes in fire intensity. Details are as follows:

[0113] When encountering temporary obstacles: The autonomous fire-fighting reconnaissance relay module detects that the track is blocked, such as by falling objects or closed fire doors. If a detour is possible, a partial detour is executed; if a detour is not possible, demolition is initiated or the issue is reported to the central intelligent decision-making and cluster dispatch module.

[0114] When multiple fire engines converge: The autonomous fire engine reconnaissance relay module detects that other fire engines are occupying the track ahead. Following the above-mentioned task level classification, it proceeds according to the principle of high priority first and low priority waiting.

[0115] When the fire situation changes abruptly: The autonomous fire-fighting reconnaissance relay module detects that the temperature or smoke ahead exceeds the limit. If communication is normal, it reports to the central intelligent decision-making and cluster scheduling module to wait for a new path. If communication is interrupted, it executes emergency strategies, including retreating along the original route or heading to the nearest safe point.

[0116] By setting up autonomous firefighting systems in the aforementioned special scenarios, real-time response and dynamic handling of special scenarios along the route are achieved, ensuring the autonomous adaptability and mission continuity of the firefighting system in complex fire environments.

[0117] S3, the autonomous mobile firefighting unit execution module dispatches the firefighting unit along the track network to perform firefighting tasks based on the received fire information. The firefighting unit activates its mobile chassis and departs from the base station, moving towards the target area along the preset three-dimensional firefighting track network. Upon reaching the vicinity of the fire scene, if fire doors are closed or obstacles obstruct the way, the firefighting unit activates the breaching and entry module, using a robotic arm and end effector to quickly breach (e.g., cut door locks, push aside minor obstacles) to create a passage. After entering the core fire scene, the panoramic camera and thermal imager of the reconnaissance relay module begin working, transmitting key information such as real-time video, temperature field data, and toxic gas concentrations from inside the fire scene back to the control center. Subsequently, the extinguishing agent spraying module automatically selects the extinguishing mode (e.g., direct current, spray, foam) and adjusts the spray angle and flow rate based on the locally identified fire source information, implementing precise coverage and continuous suppression of the fire source until the thermal imaging shows that the core temperature of the fire source has dropped below the safe threshold.

[0118] S4. During firefighting, the human-machine interaction and collaborative command module conducts collaborative command based on dispatch decisions, realizing status monitoring and dynamic adjustment until the fire is brought under control. Throughout the entire process of the autonomous mobile firefighting unit's mission, the human-machine interaction and collaborative command module dynamically presents all key information in real time on the control center's large screen. The main screen uses a digital twin model as its base map, dynamically updating the comprehensive situation including fire spread simulation, real-time location of each firefighting unit, remaining power / extinguishing agent dosage, and building smoke exhaust system status. The auxiliary screens display high-definition video from the first-person perspective of each firefighting unit, thermal imaging images, and environmental parameter change curves in a split-screen format. Commanders can intervene based on actual changes at the fire scene: In monitoring mode, they can confirm or reject automatically generated follow-up tasks (such as deploying additional firefighters or adjusting extinguishing agent types); in complex situations such as sudden changes in fire intensity or communication interference, they can switch to remote control mode to precisely control the movement, breaching, and spraying actions of one or more firefighters using joysticks; when personnel are required to enter the fire, a collaborative mode is activated, and the system plans a relatively safe entry path based on real-time environmental data transmitted from the firefighters, guiding firefighters along the path. Throughout the entire firefighting process, all command decisions, firefighter status, and environmental data are fully recorded until the fire is completely extinguished. Once the system confirms there is no risk of reignition, it automatically issues a return command to the firefighters, initiating a mission debriefing and post-battle assessment process.

[0119] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0120] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A high-rise building intelligent fire protection system with built-in autonomous mobile fire-fighting body, characterized in that the system... include: The three-dimensional fire-fighting track network module, which is pre-installed inside the building, provides a three-dimensional movement channel and energy and extinguishing agent delivery for autonomous mobile fire-fighting bodies. The distributed sensing and fire location module includes a sensor group deployed at key nodes of the three-dimensional fire track network to collect fire information; The central intelligent decision-making and cluster dispatch module receives the fire information based on a digital twin model and generates a fire-fighting plan that includes dispatch instructions for autonomous mobile fire-fighting units based on multi-agent collaborative path planning. The autonomous mobile firefighting unit execution module includes at least one autonomous mobile firefighting unit, which can move autonomously on the three-dimensional firefighting track network and perform firefighting tasks according to the firefighting plan; The human-computer interaction and collaborative command module is configured to display fire scene information to firefighters through the control center and receive manual commands.

2. The intelligent fire protection system for high-rise buildings with built-in autonomous mobile fire-fighting bodies as described in claim 1, characterized in that, The three-dimensional fire protection track network module includes: A dedicated fire shaft pre-installed within the building serves as a main vertical transport route; A horizontal lightweight track network installed in the ceilings or inside the walls of public areas on each floor; Water supply mains, power supply lines, and data communication bus are integrated into the dedicated fire shaft and horizontal light rail network; The fast-connection base stations, located at key nodes, are configured to have physical locking, automatic water and charging, and data interaction. Key nodes include stairwells on each floor and outside equipment rooms.

3. The intelligent fire protection system for high-rise buildings with built-in autonomous mobile fire-fighting bodies as described in claim 1, characterized in that: The distributed sensing and fire location module is based on the fusion of signals from the building's built-in fire alarm system and sensor group information to construct a distributed sensing network. The sensor group includes smoke and temperature detectors, thermal imaging cameras, multispectral flame detectors, and gas composition sensors.

4. The intelligent fire protection system for high-rise buildings with built-in autonomous mobile fire-fighting units as described in claim 3, characterized in that: The firefighting plan is designed based on fire information and the real-time status of autonomous mobile firefighting units. The plan includes the dispatch of autonomous mobile firefighting units, route selection, task allocation, and dynamic adjustment of building equipment.

5. The intelligent fire protection system for high-rise buildings with a built-in autonomous mobile fire-fighting unit as described in claim 2, characterized in that, The autonomous mobile fire-fighting system structure includes: The mobile chassis is configured to move along three-dimensional fire-fighting tracks and connect to base stations; The task module interface is configured to mount different task modules. The breaching and entry module is configured to enable the removal of obstacles in a fire and rapid entry into building structures. The extinguishing agent spraying module is configured to achieve precise spraying of different types of extinguishing agents; The reconnaissance relay module is configured to achieve real-time image acquisition, environmental parameter monitoring, and communication signal relay of the fire scene environment; The local controller is configured to accept commands from the control center and has edge computing capabilities.

6. The intelligent fire protection system for high-rise buildings with built-in autonomous mobile fire-fighting bodies as described in claim 5, characterized in that: The execution steps of the autonomous mobile fire-fighting unit in carrying out fire-fighting tasks are as follows: Deployment and Autonomous Movement: After receiving instructions from the control center, the autonomous mobile firefighting unit activates its chassis to move along a pre-set three-dimensional firefighting track network; Obstacle removal and entry: In confined space scenarios, the obstacle removal module is used to perform the removal of obstacles. Fire reconnaissance and environmental awareness: Fire information is collected through the reconnaissance relay module, and it interacts with the human-machine interaction and collaborative command module as a communication relay node; Precise fire suppression and sustained fire control: The fire extinguishing agent spraying model selects the appropriate fire extinguishing method based on the fire situation information until the fire is extinguished; Mission Completion and Autonomous Return: After the fire is extinguished, the autonomous mobile firefighting unit returns to the nearest or designated base station to resupply and resume standby status.

7. The intelligent fire protection system for high-rise buildings with built-in autonomous mobile fire-fighting units as described in claim 1, characterized in that, The human-computer interaction and collaborative command module receives manual instructions in the following modes: Supervision mode: Firefighting tasks are deployed based on the fire development simulated by the digital twin model; Remote control mode: In complex situations, it allows direct control of one or more autonomous mobile fire-fighting units for operation; Collaborative mode: The system plans a safe path and adopts a collaborative approach where robots perform tasks first, followed by personnel.

8. A fire-fighting method, said method being applied to a high-rise building intelligent fire-fighting system with a built-in autonomous mobile fire-fighting body as described in any one of claims 1 to 7, characterized in that: The method includes the following steps: S1. After a fire occurs, the distributed sensing and fire location module detects and analyzes the fire situation and collects fire information based on the sensor network of the sensor group. S2, the central intelligent decision-making and cluster dispatch module generates fire-fighting plans based on the digital twin model and multi-agent collaborative path planning according to fire information and dispatches the decision to the autonomous mobile fire-fighting body execution module; S3, the autonomous mobile fire-fighting unit execution module dispatches the fire-fighting unit to move along the track network and perform fire-fighting tasks based on the received fire information; S4. During the firefighting process, the human-machine interaction and collaborative command module conducts human-machine collaborative command based on scheduling decisions, realizing status monitoring and dynamic adjustment until the fire is brought under control.

9. A fire-fighting method as described in claim 8, characterized in that: The multi-agent cooperative path planning includes the following steps: Task allocation: Based on fire information and fire protection system status processing, including task decomposition and classification, as well as matching and allocation; Global path: Based on task allocation results, 3D fire track network topology, and real-time fire dynamic weight processing, including track network modeling and single-machine path generation; Local adjustments: By setting the local controller of the fire protection system to enable autonomous decision-making in multiple scenarios, including when encountering temporary obstacles, when multiple machines converge, and when the fire situation changes suddenly.

10. A fire-fighting method as described in claim 9, characterized in that: The single-machine path generation is based on The algorithm evaluates the priority of each node using a cost function to obtain the optimal path, as shown in the following formula: In the formula: The actual cost represents the distance from the starting point to the current node. The travel time already consumed; As a heuristic function, it estimates the value starting from the current node. The time still needed to reach the destination; This is a dynamic penalty term, reflecting the process from the starting point to the node. The additional risk costs incurred due to passing through areas with high temperatures or dense smoke along the route.

Images