Intelligent fire extinguishing robot for underground garage

Abstract

The invention discloses an intelligent fire extinguishing robot for an underground garage. The intelligent fire extinguishing robot comprises multiple system modules, wherein an intelligent navigation and positioning system constructs a three-dimensional map for positioning and obstacle avoidance; the parking space intelligent identification system scans the parking space to generate a thermodynamic diagram for early warning battery spontaneous combustion; a mechanical arm of the efficient fire extinguishing execution system is integrated with a force control sensor and the like, and multi-fire-extinguishing-agent switching is supported; the multi-robot collaborative operation module realizes data sharing of main and auxiliary robots, and the auxiliary robot moves the vehicle and releases the unmanned aerial vehicle; the multi-mode perception sensor group can identify glass materials and the like; the control system fuses data to adjust mechanical arm parameters; the energy management system adopts a graphene battery and a photovoltaic film; and the remote operation and maintenance module uses a block chain for evidence storage and a universe interface for commanding. All the systems construct a link through a bus, and data collection, instruction generation and linkage of all the systems are achieved. The robot dynamically plans a path based on central axis cruise, fire prevention and control are completed through multi-machine linkage, and full-process intelligent fire prevention and control are achieved through multi-system cooperation.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent fire-fighting robots, and in particular relates to an intelligent fire-fighting robot for underground garages. Background Art

[0002] With the accelerating pace of urbanization, underground garages, as crucial urban infrastructure, are expanding in number and size. These enclosed spaces, densely packed with vehicles, and numerous electrical devices create significant fire hazards. Vehicle battery failure, electrical shorts, and the accumulation of flammable materials can all potentially cause fires. Once a fire occurs, it spreads rapidly, with smoke dispersing quickly. Evacuation and firefighting are extremely difficult, resulting in significant casualties and property damage.

[0003] Currently, fire prevention and control in underground garages primarily relies on traditional firefighting systems, such as automatic sprinkler systems and smoke alarm systems. However, these traditional systems have significant limitations. Traditional smoke alarm systems may experience false alarms or missed alarms in the complex environment of underground garages, failing to promptly and accurately detect early-stage fire hazards. While automatic sprinkler systems can spray water to extinguish fires after a fire occurs, their effectiveness is limited for certain types of fires, such as battery spontaneous combustion, and the water spraying may cause secondary hazards such as electrical short circuits. Furthermore, traditional firefighting systems lack intelligence and collaborative capabilities, making it impossible to flexibly adjust firefighting strategies based on the actual conditions at the fire scene. They also struggle to effectively share data and collaborate with other equipment.

[0004] In terms of fire-fighting equipment, existing underground garage fire-fighting robots have deficiencies in navigation and positioning, fire identification, and fire-fighting execution. Traditional navigation and positioning systems have low accuracy, making it difficult to achieve precise positioning and active obstacle avoidance in complex underground garage environments, which can easily cause the robot to collide or get lost during movement. The fire identification system's ability to detect parking space characteristics and battery spontaneous combustion risks is insufficient, and it is unable to promptly detect early signs of spontaneous combustion, such as abnormal heating of vehicle batteries. The fire-fighting execution system lacks flexibility and accuracy, making it difficult to adapt to different types of fires and complex fire-fighting scenarios. For example, it cannot effectively penetrate bulletproof glass for targeted firefighting, and the selection and switching of fire-extinguishing agents are not intelligent enough. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent fire-fighting robot for an underground garage, so as to solve the problem in the prior art proposed in the background technology that the fire prevention and control in underground garages has insufficient navigation and positioning accuracy, which cannot meet the demand for precise movement of the robot in the underground garage.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] An intelligent fire-fighting robot for an underground garage, comprising:

[0008] Intelligent navigation and positioning system, building three-dimensional high-precision map positioning, integrating multi-sensor data to achieve active obstacle avoidance;

[0009] The intelligent parking space recognition system scans parking space characteristics and generates a heat map to warn of battery spontaneous combustion risks;

[0010] Highly efficient fire extinguishing execution system, the robotic arm integrates force control sensors and multi-degree-of-freedom nozzles, and supports intelligent switching of multiple fire extinguishing agents;

[0011] Multi-robot collaborative operation module: the main and auxiliary robots share data, and the auxiliary robot moves and releases drones to cover blind spots;

[0012] A multi-modal sensor system identifies glass material, tracks fire sources, and detects flying broken glass to trigger safety braking.

[0013] The control system integrates multi-sensor data and adaptively adjusts the puncture parameters of the robotic arm;

[0014] Energy management system, graphene battery supports fast charging, photovoltaic film replenishes power to extend battery life;

[0015] Remote operation and maintenance module, blockchain evidence data, and metaverse interface support AR remote command;

[0016] Each system forms a signal link through a bus: the sensor group collects environmental data and transmits it to the control system. The data fusion algorithm generates control instructions, driving the navigation system to plan routes, the execution system to accurately extinguish fires, and the collaborative module to schedule cluster operations. The energy system provides power to the entire system, and the remote module synchronizes data in real time to form a closed-loop management, realizing intelligent fire prevention and control with multi-system linkage.

[0017] The robot implements dynamic path planning based on the central axis cruising mode and completes the entire fire prevention and control process through multi-machine linkage.

[0018] According to the above technical solution, the intelligent navigation and positioning system actively avoids obstacles by:

[0019] The intelligent navigation and positioning system includes a dual-lidar and visual SLAM fusion module and a vehicle-road collaboration module. The dual-lidar and visual SLAM fusion module is used to construct a high-precision 3D map of the garage in real time to improve the robot's positioning accuracy.

[0020] The vehicle-road collaboration module is used for real-time communication between the robot and the garage's intelligent equipment, and combines millimeter-wave radar, ultrasonic waves, and infrared thermal imaging sensors to achieve active obstacle avoidance during the robot's movement.

[0021] According to the above technical solution, the parking space intelligent identification system warns of battery spontaneous combustion specifically as follows:

[0022] The intelligent parking space recognition system includes a super-resolution optical sensor, an infrared thermal imager, a light projection sensor, and a digital twin module. The super-resolution optical sensor and the infrared thermal imager simultaneously scan several parking spaces to realize multi-feature recognition of the license plate, vehicle model, and battery heat source on the parking space.

[0023] The light projection sensor includes a light projection module and a sensing module. The light projection module is used to project structured light covering the rectangular parking space, and the sensing module is used to collect fire data based on the reflected light of the projected structured light.

[0024] The digital twin module generates a three-dimensional heat map of the garage in real time, marking the vehicle location, battery temperature, and risk level to warn of potential battery spontaneous combustion hazards.

[0025] According to the above technical solution, the efficient fire extinguishing execution system is specifically as follows:

[0026] The highly efficient fire extinguishing execution system includes a carbon fiber bionic robotic arm and a fire extinguishing agent storage tank. The carbon fiber bionic robotic arm has height adjustment, multi-angle rotation, and dynamic force-controlled puncture functions. The end of the carbon fiber bionic robotic arm integrates a six-dimensional force control sensor and a multi-degree-of-freedom nozzle, which can penetrate bulletproof glass and spray fire extinguishing agent in a targeted manner.

[0027] The fire extinguishing agent storage tank is installed on the robotic arm, supporting mixed loading and intelligent switching of multiple types of fire extinguishing agents.

[0028] According to the above technical solution, the multi-machine collaborative operation module is specifically as follows:

[0029] The multi-machine collaborative operation module includes a main fire-fighting robot and a secondary car-moving robot cluster; the main fire-fighting robot and the secondary car-moving robot share fire scene data through distributed edge computing; the secondary car-moving robot is equipped with an electromagnetic adsorption chassis to move adjacent vehicles of the burning vehicle, and the secondary car-moving robot can release micro fire-fighting drones to cover high-altitude blind spots.

[0030] According to the above technical solution, the multimodal perception sensor group is specifically:

[0031] The multimodal perception sensor group includes a multispectral glass recognition sensor and a UV-visible fire source detection sensor; the multispectral glass recognition sensor uses a combination of near-infrared light source and CCD array to identify tempered / laminated / ceramic composite glass materials; the UV-visible fire source detection sensor integrates a dual-spectrum module to support dynamic tracking of fire sources and early risk prediction.

[0032] According to the above technical solution, the multimodal perception sensor group also includes an infrared dual-function sensor and an ultrasonic array sensor; the infrared dual-function sensor includes a ToF laser radar and an uncooled thermal imager, which are used to construct a three-dimensional profile and temperature field model of the target vehicle;

[0033] The ultrasonic array sensor has multiple groups of transducers distributed circumferentially, which are used to detect near-field broken glass flying in real time and trigger emergency braking of the robotic arm.

[0034] According to the above technical solution, the control system is specifically:

[0035] The control system includes a multi-sensor data fusion algorithm and an adaptive PID controller; the multi-sensor data fusion algorithm fuses IMU attitude data, force and torque feedback, lidar point cloud and thermal imaging data through Kalman filtering; the adaptive PID controller automatically adjusts the puncture parameters according to the glass material.

[0036] According to the above technical solution, the energy management system is specifically:

[0037] The energy management system includes graphene solid-state batteries and magnetic resonance wireless charging piles; graphene solid-state batteries and magnetic resonance wireless charging piles are used to support fast charging of the robot.

[0038] According to the above technical solution, the remote operation and maintenance module is specifically:

[0039] The remote operation and maintenance module uses blockchain encryption technology to store fire data and operation logs to meet fire audit requirements, and sets up a metaverse digital twin control interface to support remote immersive command through AR glasses and real-time synchronization of the robotic arm's perspective and thermal data.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] This invention achieves intelligent fire prevention and control throughout the entire process through innovative collaboration across multiple systems. The intelligent navigation and positioning system integrates multiple sensors to construct a three-dimensional, high-precision map, enabling precise positioning and active obstacle avoidance, ensuring rapid arrival at the fire scene. The intelligent parking space recognition system, leveraging ultra-resolution optical sensors and other equipment, scans parking spaces in real time to generate heat maps, providing precise warnings of battery spontaneous combustion risks. In the highly efficient fire extinguishing execution system, a carbon fiber bionic robotic arm, combined with force-controlled sensors, multi-degree-of-freedom nozzles, and intelligent switching capabilities for multiple extinguishing agents, enables precise extinguishing based on the fire situation. In the multi-machine collaborative operation module, the primary and secondary robots share data, allowing the secondary robot to move the vehicle and release drones to cover blind spots, optimizing the rescue process. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a schematic diagram of the functional modules of the intelligent fire-fighting robot of the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0044] Example 1

[0045] like Figure 1 As shown, an intelligent fire-fighting robot for an underground garage includes:

[0046] Intelligent navigation and positioning system, building three-dimensional high-precision map positioning, integrating multi-sensor data to achieve active obstacle avoidance;

[0047] The intelligent parking space recognition system scans parking space characteristics and generates a heat map to warn of battery spontaneous combustion risks;

[0048] Highly efficient fire extinguishing execution system, the robotic arm integrates force control sensors and multi-degree-of-freedom nozzles, and supports intelligent switching of multiple fire extinguishing agents;

[0049] Multi-robot collaborative operation module: the main and auxiliary robots share data, and the auxiliary robot moves and releases drones to cover blind spots;

[0050] A multi-modal sensor system identifies glass material, tracks fire sources, and detects flying broken glass to trigger safety braking.

[0051] The control system integrates multi-sensor data and adaptively adjusts the puncture parameters of the robotic arm;

[0052] Energy management system, graphene battery supports fast charging, photovoltaic film replenishes power to extend battery life;

[0053] Remote operation and maintenance module, blockchain evidence data, and metaverse interface support AR remote command;

[0054] Each system forms a signal link through a bus: the sensor group collects environmental data and transmits it to the control system. The data fusion algorithm generates control instructions, driving the navigation system to plan routes, the execution system to accurately extinguish fires, and the collaborative module to schedule cluster operations. The energy system provides power to the entire system, and the remote module synchronizes data in real time to form a closed-loop management, realizing intelligent fire prevention and control with multi-system linkage.

[0055] The robot implements dynamic path planning based on the central axis cruising mode and completes the entire fire prevention and control process through multi-machine linkage.

[0056] The working process of the intelligent fire-fighting robot in the underground garage includes the following steps:

[0057] Step A1: Fire signal acquisition and ignition location positioning. After receiving the alarm information sent by the fire detection device, the robot uses the central axis cruising mode to realize dynamic path planning and head to the ignition location.

[0058] Step A2: Central axis cruise path planning and movement.

[0059] The intelligent navigation and positioning system activates the central axis cruise mode based on the fire coordinates: dual 16-line laser radars (scanning range 360°×270°) and binocular vision cameras (resolution 1920×1080) use the LOAM fusion algorithm to build a three-dimensional high-precision map of the garage (positioning accuracy at the centimeter level), and combine with the vehicle-road collaboration module (5G-V2X communication) to obtain real-time traffic data from garage smart devices (such as cameras).

[0060] During the movement, millimeter-wave radar (detection distance ≤ 20 meters), ultrasonic sensors (near field ≤ 3 meters) and infrared thermal imaging work together to realize the "pre-perception-obstacle avoidance-resetting the central axis" closed loop: when an obstacle is detected, the robot's maximum lateral displacement is ≤ 30 cm, and the reset time after obstacle avoidance is ≤ 2 seconds, ensuring that it reaches the landing position along the shortest path (error ≤ 0.5 meters).

[0061] Step A3, structured light projection confirmation and coordinated scheduling with the secondary robot. After arriving at the burning parking space, the robot is equipped with multiple light projection sensors, and the multiple light projection sensors project multiple rectangular frames. Each rectangular monitoring frame projected by the light projection sensor corresponds to monitoring one parking space (for example, 3 rectangular monitoring frames correspond to three parking spaces). After the robot arrives at the location of the burning vehicle, it aligns the 3 middle monitoring frames with the starting parking space for secondary identification; if it is detected that the vehicle in the middle parking space is on fire, the rectangular monitoring frames on both sides simultaneously identify the vehicle information of the adjacent parking spaces. If a vehicle is identified, the secondary car moving robot is notified to move the adjacent vehicles of the burning vehicle.

[0062] The light projection module projects structured light (coding pattern frequency 200Hz) covering the entire rectangular area into the parking space. The sensing module analyzes the phase distortion and intensity attenuation of the reflected light based on the principle of triangulation, detects micro-deformations on the parking space surface (accuracy 0.1mm) and smoke particle scattering (particle size detection lower limit 0.1μm), and confirms the specific location of the fire source (such as the battery compartment, engine hood, etc.).

[0063] The control system sends instructions to the auxiliary car moving robot cluster through distributed edge computing nodes (such as AWS Greengrass):

[0064] The auxiliary robot (single unit size 0.8m×0.5m×1.2m) uses an electromagnetic adsorption chassis (suction force ≥5000N) to absorb the chassis of the vehicle adjacent to the burning vehicle and move it to a safe area (≥5 meters from the fire source) within 30 seconds to expand the fire protection distance;

[0065] A micro firefighting drone (with a 20-minute flight time and a 4K camera) is simultaneously released to cover the vehicle roof and high-altitude blind spots, transmitting fire video streams and thermal imaging data in real time (delay ≤ 500ms).

[0066] Step A4: Multi-modal sensing of demolition point positioning and precise operation of the robotic arm.

[0067] A multispectral glass recognition sensor (near-infrared light source + CCD array) analyzes the window glass material (98% accuracy for tempered / laminated / ceramic composite glass) and determines the optimal breaking point (error ≤ 1cm) through a stress distribution algorithm.

[0068] The infrared dual-function sensor (ToF lidar + uncooled thermal imager) constructs the target vehicle's three-dimensional profile (accuracy ±2mm) and temperature field model, marking the battery module position to avoid puncture risks;

[0069] The control system uses a Kalman filter algorithm to fuse IMU attitude data, force and torque feedback (six-dimensional force control sensor measurement range ±500N, torque ±50N·m) and lidar point cloud, driving a carbon fiber bionic robotic arm (weight ≤50kg, extension length 20 meters) to approach the glass at a vertical angle (deviation ≤5°).

[0070] Demolition and fire extinguishing execution process:

[0071] Step B1, dynamic force-controlled puncture: When the puncture gun at the end of the robotic arm (travel 3 meters, force control resolution 1N) contacts the glass, the force torque sensor provides real-time feedback of the contact force. When glass cracking is detected (force value sudden change > 200N), the adaptive PID controller automatically reduces the puncture speed by 10% to avoid excessive penetration and damage to the battery;

[0072] Step B2, intelligent switching of multiple fire extinguishing agents: Based on the type of fire source (e.g., battery spontaneous combustion prioritizes perfluorohexanone, oil line fire prioritizes foam), dual 100L fire extinguishing agent storage tanks achieve four-way switching through solenoid valves. The nozzles are switched to "high-pressure penetration mode" (pressure ≥ 10MPa), and the fire extinguishing agent is sprayed directionally through the hollow pipe to the fire source (e.g., the gap in the battery compartment);

[0073] Step B3, closed-loop monitoring of fire extinguishing effectiveness: The UV-visible light sensor continuously tracks the fire source. When the UV light signal disappears and the thermal imager shows a cooling rate of ≥50°C / s and a temperature below 100°C, the fire extinguishing is considered complete and the robotic arm resets to cruise standby mode.

[0074] Full-process data closed-loop and remote control: Energy management system: Graphene solid-state batteries (capacity 20kWh, energy density 500Wh / kg) provide power throughout the entire process, and the perovskite photovoltaic film on the surface of the robotic arm (photoelectric conversion rate 28%) replenishes 1.5kWh of energy per hour when there is sufficient sunlight to ensure continuous operation.

[0075] Remote operation and maintenance module: Fire data and operation logs are encrypted and stored in the IPFS network (retention ≥ 5 years) through blockchain technology (alliance chain architecture). Administrators can view the robotic arm's perspective images and three-dimensional thermal maps in real time through the Metaverse digital twin interface (such as Microsoft HoloLens2). It supports gesture commands to adjust the nozzle angle (accuracy ±1°), with operation delay ≤ 200ms, realizing remote immersive command.

[0076] This invention achieves intelligent fire prevention and control throughout the entire process through innovative collaboration across multiple systems. The intelligent navigation and positioning system integrates multiple sensors to construct a three-dimensional, high-precision map, enabling precise positioning and active obstacle avoidance, ensuring rapid arrival at the fire scene. The intelligent parking space recognition system, leveraging ultra-resolution optical sensors and other equipment, scans parking spaces in real time to generate heat maps, providing precise warnings of battery spontaneous combustion risks. In the highly efficient fire extinguishing execution system, a carbon fiber bionic robotic arm, combined with force-controlled sensors, multi-degree-of-freedom nozzles, and intelligent switching capabilities for multiple extinguishing agents, enables precise extinguishing based on the fire situation. In the multi-machine collaborative operation module, the primary and secondary robots share data, allowing the secondary robot to move the vehicle and release drones to cover blind spots, optimizing the rescue process.

[0077] Example 2

[0078] This embodiment provides a specific implementation of an intelligent navigation and positioning system:

[0079] Hardware configuration: The chassis is equipped with dual 16-line laser radars (scanning range 360°×270°), binocular vision cameras (resolution 1920×1080), and the edge computing unit (ECU) uses NVIDIA Jetson Xavier NX (computing power 32TOPS).

[0080] Software process:

[0081] Dual LiDARs and visual SLAM fusion algorithms (such as LOAM) build a high-precision 3D map of the garage, with centimeter-level positioning accuracy.

[0082] The vehicle-road collaboration module exchanges vehicle location data with the garage smart camera in real time through 5G-V2X, and combines millimeter-wave radar (detection distance ≤ 20 meters), ultrasonic wave (near field ≤ 3 meters), and infrared thermal imaging (temperature anomaly detection) to realize the "pre-perception-obstacle avoidance-resetting the central axis" closed loop. The maximum lateral displacement during obstacle avoidance is ≤ 30 cm, and the reset time is ≤ 2 seconds.

[0083] This embodiment provides a specific implementation of a parking space intelligent recognition system:

[0084] Hardware Configuration: Super-resolution optical sensors (resolution 0.5μm) and uncooled infrared thermal imagers (temperature range -40°C-1500°C) are installed on both sides of the vehicle body, along with a built-in YOLOv7-Bifpn target detection model (inference speed 50FPS). The robot in this invention can proactively proceed to the alarm area based on the alarm information sent by the fire detection device, and then conduct secondary confirmation through the intelligent parking space recognition system. The robot can also proactively detect vehicles in parking spaces and identify fires through intelligent cruising.

[0085] Data processing:

[0086] A super-resolution optical sensor (resolution up to 0.5μm) and an uncooled infrared thermal imager (temperature range -40°C to 1500°C) simultaneously scan parking spaces. Using the built-in YOLOv7-Bifpn object detection model (inference speed 50FPS), they extract and identify multiple features, including license plate color (99.9% accuracy for red plates on new energy vehicles), vehicle model (matching a database of over 200 models), and battery heat sources. The infrared thermal imager uses a temperature gradient analysis algorithm to monitor temperature changes in the battery heat source, triggering an early warning mechanism when it detects an abnormal temperature increase of >5°C / s.

[0087] The digital twin module generates a three-dimensional heat map in real time based on the Unity engine, marks the risk level with different colors (green - safety, yellow - warning, red - fire), and pushes it to the fire control room through the API interface.

[0088] The light projection sensor uses structured light coding technology. Its light projection module projects structured light that can cover the entire rectangular parking space. The sensing module, based on the principle of triangulation, collects and decodes the reflected light of the projected structured light at high speed. By analyzing parameters such as the phase distortion and intensity attenuation of the reflected light, it can accurately capture early fire characteristics such as micro-deformation of the parking space surface and scattering of smoke particles.

[0089] This embodiment provides a specific implementation of an efficient fire extinguishing execution system:

[0090] Robotic arm design: Carbon fiber bionic robotic arm (weight ≤ 50kg, extension length 20 meters) includes 6 rotary joints (accuracy ±0.1°), telescopic puncture gun (stroke 3 meters, dynamic force control resolution 1N), and terminal integration:

[0091] Six-dimensional force control sensor (measuring range ±500N, torque ±50N·m);

[0092] Multi-degree-of-freedom nozzle (360° rotation, spray angle adjustable within ±15°), connected to dual 100L fire extinguishing agent storage tanks (supports four-way switching of dry powder / foam / perfluorohexanone / water-based).

[0093] Demolition logic:

[0094] The multi-spectral glass recognition sensor determines the optimal breaking point (stress concentration area), and the robotic arm approaches the glass at a vertical angle (deviation ≤ 5°);

[0095] During the puncture process, the force-torque sensor provides real-time feedback on the contact force. When glass cracking is detected (force value suddenly changes > 200N), the adaptive PID controller automatically reduces the puncture speed by 10% to avoid excessive penetration and damage to the battery module.

[0096] The nozzle is switched to "high-pressure penetration mode" (pressure ≥ 10 MPa), and the fire extinguishing agent is sprayed directionally to the fire source (such as the gap in the battery compartment) through the hollow pipe. The thermal imager continuously monitors the temperature until the fire is extinguished (the cooling rate is ≥ 50℃ / s and the ultraviolet light signal disappears).

[0097] This embodiment provides a specific implementation method for multi-machine collaborative operation:

[0098] The cluster consists of 1 main fire-fighting robot + 2 auxiliary car-moving robots (single-machine size 0.8m×0.5m×1.2m), and data sharing is achieved through distributed edge computing nodes (such as AWS Greengrass).

[0099] Task Assignment:

[0100] After receiving the fire signal, the main robot starts and drives towards the fire source within 5 seconds. At the same time, it locks the fire point through the ultraviolet-visible light sensor, punctures and demolishes it, and sprays the fire extinguishing agent.

[0101] The auxiliary robot simultaneously releases a micro firefighting drone (with a flight time of 20 minutes and a camera resolution of 4K) to cover the roof and high-altitude blind spots, and transmit fire data in real time;

[0102] The auxiliary robot uses an electromagnetic adsorption chassis (suction force ≥ 5000N) to absorb the chassis of adjacent vehicles and move to a safe area (≥ 5 meters from the fire source) within 30 seconds to expand the fire protection distance.

[0103] This embodiment provides a specific implementation of an energy management system:

[0104] Battery configuration: Graphene solid-state battery (capacity 20kWh, energy density 500Wh / kg), paired with a garage magnetic resonance wireless charging station (power 120kW, charging efficiency 92%), supporting "senseless docking" (automatic parking error ≤ 10cm);

[0105] Photovoltaic energy replenishment: The surface of the robotic arm is covered with a perovskite photovoltaic film (area 2㎡, photoelectric conversion rate 28%). When the light intensity is ≥1000Lux, 1.5kWh of electricity is replenished per hour, and the battery life in emergency mode is extended from 8 hours to 12 hours.

[0106] This embodiment provides a specific implementation of a remote operation and maintenance module:

[0107] Data storage: Fire data (temperature, video, operation log) is encrypted and stored in the IPFS distributed network through blockchain nodes (alliance chain architecture) to ensure that it cannot be tampered with and meet the traceability requirements of fire audits (data retention ≥5 years);

[0108] Remote control: Administrators access the Metaverse digital twin interface through AR glasses (such as Microsoft HoloLens 2) to view the robotic arm's first-person perspective image, 3D thermal map, and sensor data in real time. Gesture commands are supported to adjust the robotic arm's angle (accuracy ±1°) or switch the type of fire extinguishing agent, with an operation delay of ≤200ms.

[0109] The multimodal perception sensor group integrates multiple sensors to identify glass material, track fire sources, and detect flying glass fragments, triggering safety brakes to ensure operational safety. The control system integrates multi-source data and uses adaptive algorithms to automatically adjust the robotic arm's penetration parameters. The energy management system utilizes graphene battery fast charging and photovoltaic thin-film recharge to extend battery life. The remote operation and maintenance module uses blockchain evidence storage and Metaverse AR command to achieve remote and precise control. Each system establishes a signal link through a bus, forming a closed-loop management system from data collection, command generation, task execution, and feedback optimization, comprehensively improving the efficiency and reliability of underground garage fire prevention and control.

[0110] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0111] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An intelligent fire-fighting robot for underground garages, characterized by: include: Intelligent navigation and positioning system, building three-dimensional high-precision map positioning, integrating multi-sensor data to achieve active obstacle avoidance; The intelligent parking space recognition system scans parking space characteristics and generates a heat map to warn of battery spontaneous combustion risks; Highly efficient fire extinguishing execution system, the robotic arm integrates force control sensors and multi-degree-of-freedom nozzles, and supports intelligent switching of multiple fire extinguishing agents; Multi-robot collaborative operation module: the main and auxiliary robots share data, and the auxiliary robot moves and releases drones to cover blind spots; A multi-modal sensor system identifies glass material, tracks fire sources, and detects flying broken glass to trigger safety braking. The control system integrates multi-sensor data and adaptively adjusts the puncture parameters of the robotic arm; Energy management system, graphene battery supports fast charging, photovoltaic film replenishes power to extend battery life; Remote operation and maintenance module, blockchain evidence data, and metaverse interface support AR remote command; Each system builds a signal link through the bus: the sensor group collects environmental data and transmits it to the control system. The data fusion algorithm generates control instructions to drive the navigation system to plan the path, the execution system to accurately extinguish the fire, and the collaborative module to schedule cluster operations. The energy system provides power to the entire system, and the remote module synchronizes data in real time to form a closed-loop management system, enabling intelligent fire prevention and control with multi-system linkage. The robot implements dynamic path planning based on the central axis cruising mode and completes the entire fire prevention and control process through multi-machine linkage.

2. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The intelligent navigation and positioning system's active obstacle avoidance is specifically: The intelligent navigation and positioning system includes a dual-lidar and visual SLAM fusion module and a vehicle-road collaboration module. The dual-lidar and visual SLAM fusion module is used to construct a high-precision 3D map of the garage in real time to improve the robot's positioning accuracy. The vehicle-road collaboration module is used for real-time communication between the robot and the garage's intelligent equipment, and combines millimeter-wave radar, ultrasonic waves, and infrared thermal imaging sensors to achieve active obstacle avoidance during the robot's movement.

3. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The specific warning of battery spontaneous combustion by the intelligent parking space identification system is: The intelligent parking space recognition system includes a super-resolution optical sensor, an infrared thermal imager, a light projection sensor, and a digital twin module. The super-resolution optical sensor and the infrared thermal imager simultaneously scan several parking spaces to realize multi-feature recognition of the license plate, vehicle model, and battery heat source on the parking space. The light projection sensor includes a light projection module and a sensing module. The light projection module is used to project structured light covering the rectangular parking space, and the sensing module is used to collect fire data based on the reflected light of the projected structured light. The digital twin module generates a three-dimensional heat map of the garage in real time, marking the vehicle location, battery temperature, and risk level to warn of potential battery spontaneous combustion hazards.

4. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The efficient fire extinguishing execution system is specifically: The highly efficient fire extinguishing execution system includes a carbon fiber bionic robotic arm and a fire extinguishing agent storage tank. The carbon fiber bionic robotic arm has height adjustment, multi-angle rotation, and dynamic force-controlled puncture functions. The end of the carbon fiber bionic robotic arm integrates a six-dimensional force control sensor and a multi-degree-of-freedom nozzle, which can penetrate bulletproof glass and spray fire extinguishing agent in a targeted manner. The fire extinguishing agent storage tank is installed on the robotic arm, supporting mixed loading and intelligent switching of multiple types of fire extinguishing agents.

5. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The multi-machine collaborative operation module is specifically as follows: The multi-machine collaborative operation module includes a main fire-fighting robot and a secondary car-moving robot cluster; the main fire-fighting robot and the secondary car-moving robot share fire scene data through distributed edge computing; the secondary car-moving robot is equipped with an electromagnetic adsorption chassis to move adjacent vehicles of the burning vehicle, and the secondary car-moving robot can release micro fire-fighting drones to cover high-altitude blind spots.

6. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The multimodal perception sensor group is specifically: The multimodal perception sensor group includes a multispectral glass recognition sensor and a UV-visible fire source detection sensor; the multispectral glass recognition sensor uses a combination of near-infrared light source and CCD array to identify tempered / laminated / ceramic composite glass materials; the UV-visible fire source detection sensor integrates a dual-spectrum module to support dynamic tracking of fire sources and early risk prediction.

7. The intelligent fire-fighting robot for underground garages according to claim 6, characterized in that: The multimodal perception sensor group also includes an infrared dual-function sensor and an ultrasonic array sensor; the infrared dual-function sensor includes a ToF lidar and an uncooled thermal imager, which are used to construct a three-dimensional profile and temperature field model of the target vehicle; The ultrasonic array sensor has multiple groups of transducers distributed circumferentially, which are used to detect near-field broken glass flying in real time and trigger emergency braking of the robotic arm.

8. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The control system is specifically: The control system includes a multi-sensor data fusion algorithm and an adaptive PID controller; the multi-sensor data fusion algorithm fuses IMU attitude data, force and torque feedback, lidar point cloud and thermal imaging data through Kalman filtering; the adaptive PID controller automatically adjusts the puncture parameters according to the glass material.

9. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The energy management system is specifically: The energy management system includes graphene solid-state batteries and magnetic resonance wireless charging piles; graphene solid-state batteries and magnetic resonance wireless charging piles are used to support fast charging of the robot.

10. The intelligent fire-fighting robot for underground garages according to claim 1, characterized in that: The remote operation and maintenance modules are as follows: The remote operation and maintenance module uses blockchain encryption technology to store fire data and operation logs to meet fire audit requirements, and sets up a metaverse digital twin control interface to support remote immersive command through AR glasses and real-time synchronization of the robotic arm's perspective and thermal data.

Images