Cleaning and extinguishing integrated and all-terrain trafficability intelligent inspection vehicle

Abstract

The invention provides a cleaning and extinguishing integrated and all-terrain trafficability intelligent inspection vehicle, which relates to the field of intelligent inspection vehicles, and is characterized by comprising a mechanical arm, a storage box, a PVT thermal photovoltaic panel, an Archimedes wind motor, an aluminum profile vehicle body, a plate spring, wheels, a motor drive axle, a water tank, an emergency fire extinguishing nozzle, a double-fork-arm suspension, a sensing camera and a searchlight. The intelligent inspection vehicle has the advantages that four major breakthroughs can be realized through the intelligent inspection vehicle: 1, wind, light and heat integrated energy supply (Archimedes fan + photovoltaic PVT), the endurance of a forest region without a power grid is 72 hours, and carbon is reduced by 4380kg per year; 2, the passing rate of complex terrains is improved by 55% through front double fork arms, a rear plate spring suspension and direct-drive steering; 3, the RF-YOLO is fused with GSConv and CBAM, the fire point identification accuracy is 98.2%, the recall rate is 96.5%, and infrared verification is assisted to reduce false alarms; and 4, the bionic mechanical arm cooperates with the water mist spray head, precise decontamination and fire extinguishing are achieved, the dosage of chemicals is reduced, and fire spreading is inhibited.

Description

Technical Field

[0001] This invention relates to the field of intelligent inspection vehicles, and in particular to an intelligent inspection vehicle that integrates fire suppression and all-terrain mobility. Background Technology

[0002] Forest fire prevention and control scenarios are characterized by complex terrain (mostly mountains and ravines, with no fixed roads), wide operating range, and variable environment (unstable lighting and wind conditions), which place stringent requirements on the energy supply, terrain adaptability, and functional synergy of prevention and control equipment.

[0003] In existing forest fire fighting technologies, energy supply mostly relies on fuel power or a single new energy source (such as solar energy). The former has safety hazards such as high carbon emissions and the risk of secondary fires, while the latter is limited by environmental conditions, resulting in unstable energy supply. In terms of terrain adaptability, tracked chassis have a certain degree of passability but are cumbersome to turn, consume a lot of energy, and have high maintenance costs, while ordinary wheeled chassis are difficult to cope with complex forest terrain and have poor stability. Fire identification mainly adopts traditional image processing methods based on color space conversion or manual feature extraction, which are easily affected by lighting, obstruction, and fire-like interference, resulting in low recognition rate and high false alarm rate for early small fires. Extinguishing operations generally use fixed spraying devices or manual cleaning methods, which have problems such as waste of agents, insufficient fire extinguishing accuracy, and inability to effectively remove diverse flammable pollutants.

[0004] The aforementioned technical bottlenecks severely restrict the improvement of forest fire fighting equipment in terms of greening, intelligence, autonomy, and all-weather operation capabilities. There is an urgent need for a new type of forest fire fighting robot system that integrates efficient energy management, strong terrain adaptability, high-precision fire perception, and precise extinguishing functions. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent inspection vehicle that integrates fire suppression and all-terrain mobility, solving the problems of unstable energy supply or high carbon emissions, weak terrain adaptability, low fire identification accuracy and high false alarm rate in the prior art, as well as the problems of rough fire suppression operations, waste of agents and inaccurate fire suppression.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0007] An intelligent patrol vehicle integrating fire suppression and all-terrain mobility is characterized by comprising: a robotic arm, a storage box, a PVT thermo-photovoltaic panel, an Archimedes wind turbine, an aluminum profile body, leaf springs, wheels, a motor drive axle, a water tank, an emergency fire extinguishing nozzle, a double wishbone suspension, a sensing camera, and a searchlight.

[0008] The RF-YOLO fire detection system is built on a fusion receptive field module, integrating GSConv convolution, SPD spatial pyramid downsampling module and CBAM attention mechanism, and combined with an infrared temperature sensor for fire source verification.

[0009] The multimodal intelligent perception system integrates UWB ultra-wideband positioning modules and inertial sensors to achieve centimeter-level positioning, and fuses LiDAR and binocular vision data to complete terrain scanning, path planning and millisecond-level obstacle avoidance through dynamic window algorithms.

[0010] The Archimedes wind turbine and the PVT thermo-photovoltaic panel form a wind-solar-thermal co-power system.

[0011] The double wishbone suspension, electric drive axle, and aluminum profile body form an all-terrain adaptable chassis structure.

[0012] The robotic arm and emergency fire extinguishing nozzles form an integrated fire suppression and extinguishing device. The robotic arm is used to grab flammable pollutants of different sizes in the forest area, and the emergency fire extinguishing nozzles can target the base of the fire source and spray them, and have the function of pre-wetting the grabbed pollutants.

[0013] The power supply design of the wind-solar-thermal co-generation system has the advantage of highly efficient synergy between the Archimedes wind turbine and the photovoltaic PVT module: the former significantly improves the wind energy utilization coefficient to 40.8% with its omnidirectional air intake design, adapting to complex wind conditions; the latter simultaneously achieves power generation and waste heat recovery, suppressing photovoltaic efficiency degradation while achieving a combined thermal and power efficiency of 64.5%. Together, they support the system's zero-carbon, long-duration power supply capability.

[0014] The multimodal intelligent sensing system, also known as intelligent sensing patrol, achieves integrated collaboration of full-area patrol, centimeter-level positioning, stable communication, and dynamic obstacle avoidance in complex forest areas through multimodal sensor fusion and intelligent algorithms. Firstly, by integrating UWB and RSSI / TDOA hybrid algorithms, centimeter-level accurate positioning is achieved even under dense forest canopy, establishing a spatiotemporal benchmark for global path planning. During patrol, lidar and binocular vision scan the terrain in real time, generating point clouds that are simultaneously used to construct a passable map and perform real-time local path planning based on a dynamic window algorithm, ensuring millisecond-level obstacle avoidance while meeting motion constraints, safe distances, and patrol direction. Another independent dynamic window algorithm synchronously regulates the communication link, adaptively adjusting data transmission based on network latency to ensure stable feedback of status and data in weak signal environments. Finally, the chassis executes control commands, increasing the passability rate in complex terrain by 55%.

[0015] As an improvement, the wind-solar-thermal co-generation energy supply system can support continuous operation for no less than 72 hours without an external power grid, and the overall thermal-electric efficiency of the unit reaches more than 64.5%.

[0016] The Archimedes wind turbine enables omnidirectional air intake and low-wind-speed start-up, while the photovoltaic PVT module integrates a liquid-cooled heat dissipation structure to recover waste heat from the battery cells. The intelligent collaborative control module dynamically schedules the power supply priority of wind and solar energy based on real-time wind speed and energy fluctuations.

[0017] As an improvement, the RF-YOLO fire detection system is deployed on an edge computing platform, achieving an accuracy of no less than 98.2% and a recall rate of no less than 96.5% for detecting early small fires.

[0018] By establishing a fire detection system, this system overcomes the limitations of traditional color space conversion and manual feature extraction, innovatively employing an RF-YOLO detection architecture that integrates a receptive field module. The system automatically learns the deep spatiotemporal features of flames through a convolutional neural network, effectively overcoming false alarms caused by changes in lighting, foliage obstruction, and fire-like interference in complex forest environments. Addressing the deployment requirements of edge computing devices, the model incorporates GSConv convolution and SPD spatial pyramid downsampling modules, significantly reducing the number of model parameters while maintaining high accuracy. Combined with the MobileOne backbone network and CBAM attention mechanism, it enhances feature extraction capabilities while ensuring real-time detection performance. Through an improved feature pyramid network and adaptive training strategy, the system significantly improves the detection recall and localization accuracy of early small fires, achieving a detection accuracy of 98.2% and a recall of 96.5% on the test set, providing a core perceptual foundation for building a fully automated forest fire prevention and control system.

[0019] As an improvement, the all-terrain adaptable chassis structure has a passability rate of at least 55% higher than that of traditional tracked or ordinary wheeled equipment in complex forest terrain, and a jamming rate of less than 10%.

[0020] As an improvement, the integrated fire extinguishing and suppression device reduces the amount of fire extinguishing agent used by more than 30% through the coordinated control of the robotic arm grasping and the precise spraying of fine water mist from the emergency fire extinguishing nozzle.

[0021] As an improvement, the Archimedes wind turbine can be replaced with a vertical axis wind turbine, which does not require a wind-fighting mechanism.

[0022] As an improvement, the double wishbone suspension is divided into a front double wishbone independent suspension and a rear leaf spring beam non-independent suspension.

[0023] The front double wishbone independent suspension can be replaced with a multi-link independent suspension, and the rear leaf spring beam non-independent suspension can be replaced with an air spring non-independent suspension. The air spring non-independent suspension is equipped with a height sensor and an electronic control unit.

[0024] As an improvement, the obstacle avoidance perception unit composed of the lidar and binocular vision module can be replaced with a combination of millimeter-wave radar and monocular camera, and an obstacle avoidance accuracy of no less than 92% can be achieved by training a forest obstacle sample database.

[0025] As an improvement, the fire extinguishing actuator consisting of the emergency fire extinguishing nozzle and the robotic arm can be replaced with a vehicle-mounted high-pressure water gun and an electric pan-tilt unit.

[0026] The electric pan-tilt unit supports 360° horizontal rotation and ±90° pitch adjustment, and is automatically aligned with the fire source by the fire detection system.

[0027] As an improvement, the PVT thermal photovoltaic panel can be replaced with a flexible thin-film photovoltaic module. The flexible thin-film photovoltaic module is installed to fit the curved surface of the vehicle body, and its weight does not exceed one-third of that of a rigid module of the same power. The loss of photoelectric conversion efficiency is compensated by a multi-panel series layout.

[0028] The all-terrain-adaptive chassis structure, the core mechanical innovation of this project, integrates mature vehicle engineering technologies with lightweight design, effectively improving passability on complex terrains. The front suspension uses a double wishbone independent suspension, ensuring optimal tire contact and handling on rough surfaces; the rear suspension employs a leaf spring beam non-independent structure, balancing high load-bearing capacity and roll suppression. The steering system uses a direct-drive rack and pinion mechanism, offering direct response and adjustable tie rods, balancing steering precision and adaptability. The chassis is equipped with an open differential, adaptively adjusting wheel speed differences to improve grip and reduce tire wear. The body uses lightweight aluminum materials, achieving high rigidity and reducing curb weight. This carefully matched chassis system increases the passability on complex terrain by 55%, providing a stable and efficient mechanical platform for intelligent forest patrol equipment.

[0029] The core advantage of the integrated fire suppression and cleanup system lies in the intelligent collaborative operation of the robotic arm and the fine water mist nozzles. Pollutant removal utilizes a multi-jointed bionic robotic arm, coupled with a vision recognition and gripping module at the front of the vehicle. This allows for precise grabbing of flammable pollutants such as plastic bottles in complex terrain, achieving efficient pickup of waste of different sizes and materials. The success rate of a single grab reaches 92%, far exceeding traditional fixed cleaning devices. Emergency fire suppression is handled by the linked fine water mist nozzles, which are linked in real-time with a fire source sensor at the end of the robotic arm. In the event of a fire, the nozzles automatically switch to fire suppression mode. Through the robotic arm's flexible positioning, the nozzles are precisely aimed at the base of the fire, rapidly suffocating the combustion reaction with micron-level water mist. Simultaneously, a pre-wetting function pre-treats the grabbed flammable materials, reducing the fire spread rate by 60%.

[0030] The beneficial effects of this invention are as follows: The intelligent patrol vehicle, with its improved mobility, integrates a wind-solar-thermal co-powered system, a multimodal intelligent sensing system, the RF-YOLO high-precision fire detection algorithm, and an integrated fire suppression and extinguishing device, significantly enhancing the greening, intelligence, and autonomous operation capabilities of forest fire fighting equipment. Its beneficial effects include: continuous operation for over 72 hours without an external power grid; a comprehensive thermal-electric efficiency of 64.5%; a 55% increase in the passability rate across complex forest terrain with a jamming rate of less than 10%; an early small fire detection accuracy of no less than 98.2% and a recall rate of no less than 96.5%; a reduction of over 30% in extinguishing agent usage; and efficient removal of diverse flammable pollutants. It also possesses centimeter-level positioning, millisecond-level obstacle avoidance, and stable communication capabilities even with weak signals, comprehensively overcoming the bottlenecks of traditional equipment in energy supply, terrain adaptability, fire detection, and precise fire suppression, providing key technical support for building an all-weather, fully automated, zero-carbon emission forest fire prevention and control system. Attached Figure Description

[0031] Figure 1 This is a framework diagram of a wind-solar-thermal co-powered energy system for an intelligent inspection vehicle that integrates fire suppression and all-terrain mobility, according to the present invention.

[0032] Figure 2 This is a perception and patrol framework diagram of an intelligent patrol vehicle that integrates pest control and all-terrain mobility according to the present invention.

[0033] Figure 3 This is a flowchart of an intelligent patrol vehicle fire detection system that integrates fire suppression and all-terrain mobility, according to the present invention.

[0034] Figure 4 This is a curve showing the suspension parameters of an intelligent inspection vehicle with integrated fire suppression and all-terrain capability as described in this invention, along with its wheel travel.

[0035] Figure 5 This invention presents a diagram showing the various steering relationship curves of an intelligent inspection vehicle with integrated fire suppression and all-terrain capability, featuring an all-terrain adaptable chassis structure.

[0036] Figure 6 This is a perspective view of an intelligent patrol vehicle that integrates fire suppression and all-terrain mobility, according to the present invention.

[0037] In the picture: 1. Robotic arm; 2. Storage box; 3. PVT thermo-photovoltaic panel; 4. Archimedes wind turbine; 5. Aluminum profile body; 6. Leaf spring; 7. Wheel; 8. Motor drive axle; 9. Water tank; 10. Emergency fire extinguishing nozzle; 11. Double wishbone suspension; 12. Sensing camera; 13. Searchlight. Detailed Implementation

[0038] To make the content of this invention easier to understand, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Identical components are represented by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, while the terms "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0039] like Figures 1 to 6 As shown, an intelligent patrol vehicle integrating fire suppression and all-terrain capability is characterized by comprising: a robotic arm, a storage box, a PVT thermo-photovoltaic panel, an Archimedes wind turbine, an aluminum profile body, leaf springs, wheels, a motor drive axle, a water tank, an emergency fire extinguishing nozzle, a double wishbone suspension, a sensing camera, and a searchlight.

[0040] The RF-YOLO fire detection system is built upon a fusion receptive field module, integrating GSConv convolution, SPD spatial pyramid downsampling module, and CBAM attention mechanism, and combined with an infrared temperature sensor for fire source verification. A multimodal intelligent perception system integrates a UWB ultra-wideband positioning module and an inertial sensor to achieve centimeter-level positioning, and fuses LiDAR and binocular vision data, using a dynamic window algorithm to complete terrain scanning, path planning, and millisecond-level obstacle avoidance. The Archimedes wind turbine and PVT thermal photovoltaic panels form a wind-solar-thermal co-generation power supply system. The double wishbone suspension, motor drive axle, and aluminum profile body form an all-terrain adaptable chassis structure. The robotic arm and emergency fire extinguishing nozzles form an integrated fire suppression and extinguishing device; the robotic arm is used to grab flammable pollutants of different sizes from forest areas, and the emergency fire extinguishing nozzles can target the base of the fire source and pre-wet the grabbed pollutants.

[0041] The power supply design of the wind-solar-thermal co-generation system has the advantage of highly efficient synergy between the Archimedes wind turbine and the photovoltaic PVT module: the former significantly improves the wind energy utilization coefficient to 40.8% with its omnidirectional air intake design, adapting to complex wind conditions; the latter simultaneously achieves power generation and waste heat recovery, suppressing photovoltaic efficiency degradation while achieving a combined thermal and power efficiency of 64.5%. Together, they support the system's zero-carbon, long-duration power supply capability.

[0042] like Figures 2 to 6As shown, the multimodal intelligent sensing system, also known as intelligent sensing patrol, achieves integrated collaboration of full-area patrol, centimeter-level positioning, stable communication, and dynamic obstacle avoidance in complex forest areas through multimodal sensor fusion and intelligent algorithms. Firstly, by integrating UWB and RSSI / TDOA hybrid algorithms, centimeter-level accurate positioning is achieved even under dense forest canopy, laying a spatiotemporal benchmark for global path planning. During patrol, lidar and binocular vision scan the terrain in real time, and the generated point cloud is used simultaneously to construct a passable map and real-time local path planning based on a dynamic window algorithm, ensuring millisecond-level obstacle avoidance while meeting motion constraints, safe distances, and patrol direction. Another independent dynamic window algorithm synchronously regulates the communication link, adaptively adjusting data transmission based on network latency to ensure stable feedback of status and data in weak signal environments. Finally, the chassis executes control commands, increasing the passability rate in complex terrain by 55%.

[0043] like Figures 3 to 6 As shown, the wind-solar-thermal co-generation energy supply system supports continuous operation for no less than 72 hours without an external power grid, with a comprehensive thermal efficiency of over 64.5%. The Archimedes wind turbine enables omnidirectional air intake and low-wind-speed start-up, while the photovoltaic PVT module integrates a liquid-cooled heat dissipation structure to recover waste heat from the solar cells. An intelligent collaborative control module dynamically prioritizes wind and solar power supply based on real-time wind speed and energy fluctuations. The RF-YOLO fire detection system, deployed on an edge computing platform, achieves an accuracy rate of no less than 98.2% and a recall rate of no less than 96.5% for detecting early-stage small fires.

[0044] By establishing a fire detection system, this system overcomes the limitations of traditional color space conversion and manual feature extraction, innovatively employing an RF-YOLO detection architecture that integrates a receptive field module. The system automatically learns the deep spatiotemporal features of flames through a convolutional neural network, effectively overcoming false alarms caused by changes in lighting, foliage obstruction, and fire-like interference in complex forest environments. Addressing the deployment requirements of edge computing devices, the model incorporates GSConv convolution and SPD spatial pyramid downsampling modules, significantly reducing the number of model parameters while maintaining high accuracy. Combined with the MobileOne backbone network and CBAM attention mechanism, it enhances feature extraction capabilities while ensuring real-time detection performance. Through an improved feature pyramid network and adaptive training strategy, the system significantly improves the detection recall and localization accuracy of early small fires, achieving a detection accuracy of 98.2% and a recall of 96.5% on the test set, providing a core perceptual foundation for building a fully automated forest fire prevention and control system.

[0045] like Figures 1 to 6As shown, the all-terrain adaptable chassis structure improves the passability of complex forest terrain by at least 55% compared to traditional tracked or ordinary wheeled equipment, with a jamming rate of less than 10%. The integrated fire extinguishing and clearing device reduces the amount of fire extinguishing agent used by more than 30% through the coordinated control of the robotic arm's grasping and the precise spraying of fine water mist from the emergency fire extinguishing nozzles. The Archimedes wind turbine can be replaced with a vertical axis wind turbine, which does not require a wind-fighting mechanism.

[0046] like Figures 2 to 6 As shown, the double wishbone suspension is divided into a front double wishbone independent suspension and a rear leaf spring beam non-independent suspension. The front double wishbone independent suspension can be replaced with a multi-link independent suspension, and the rear leaf spring beam non-independent suspension can be replaced with an air spring non-independent suspension. The air spring non-independent suspension is equipped with a height sensor and an electronic control unit. The obstacle avoidance perception unit composed of a lidar and a binocular vision module can be replaced with a combination of millimeter-wave radar and a monocular camera. By training with a forest obstacle sample database, an obstacle avoidance accuracy of no less than 92% can be achieved.

[0047] like Figures 1 to 6 As shown, the fire extinguishing actuator consisting of the emergency fire extinguishing nozzle and the robotic arm can be replaced with a vehicle-mounted high-pressure water gun and an electric pan-tilt unit. The electric pan-tilt unit supports 360° horizontal rotation and ±90° pitch adjustment, and is automatically guided by a fire identification system to align with the fire source. The PVT thermophotovoltaic panel can be replaced with a flexible thin-film photovoltaic module. The flexible thin-film photovoltaic module is installed against the curved surface of the vehicle body, and its weight does not exceed one-third of that of a rigid module of the same power. It also compensates for the photoelectric conversion efficiency loss through a multi-panel series layout.

[0048] The all-terrain-adaptive chassis structure, the core mechanical innovation of this project, integrates mature vehicle engineering technologies with lightweight design, effectively improving passability on complex terrains. The front suspension uses a double wishbone independent suspension, ensuring optimal tire contact and handling on rough surfaces; the rear suspension employs a leaf spring beam non-independent structure, balancing high load-bearing capacity and roll suppression. The steering system uses a direct-drive rack and pinion mechanism, offering direct response and adjustable tie rods, balancing steering precision and adaptability. The chassis is equipped with an open differential, adaptively adjusting wheel speed differences to improve grip and reduce tire wear. The body uses lightweight aluminum materials, achieving high rigidity and reducing curb weight. This carefully matched chassis system increases the passability on complex terrain by 55%, providing a stable and efficient mechanical platform for intelligent forest patrol equipment.

[0049] The core advantage of the integrated fire suppression and cleanup system lies in the intelligent collaborative operation of the robotic arm and the fine water mist nozzles. Pollutant removal utilizes a multi-jointed bionic robotic arm, coupled with a vision recognition and gripping module at the front of the vehicle. This allows for precise grabbing of flammable pollutants such as plastic bottles in complex terrain, achieving efficient pickup of waste of different sizes and materials. The success rate of a single grab reaches 92%, far exceeding traditional fixed cleaning devices. Emergency fire suppression is handled by the linked fine water mist nozzles, which are linked in real-time with a fire source sensor at the end of the robotic arm. In the event of a fire, the nozzles automatically switch to fire suppression mode. Through the robotic arm's flexible positioning, the nozzles are precisely aimed at the base of the fire, rapidly suffocating the combustion reaction with micron-level water mist. Simultaneously, a pre-wetting function pre-treats the grabbed flammable materials, reducing the fire spread rate by 60%.

[0050] During the implementation process,

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent patrol vehicle integrating fire suppression and all-terrain capability, characterized in that, include: Robotic arm (1), storage box (2), PVT thermo-photovoltaic panel (3), Archimedes wind turbine (4), aluminum profile body (5), leaf spring (6), wheel (7), motor drive axle (8), water tank (9), emergency fire extinguishing nozzle (10), double wishbone suspension (11), sensing camera (12) and searchlight (13); The RF-YOLO fire detection system is built on a fusion receptive field module, integrating GSConv convolution, SPD spatial pyramid downsampling module and CBAM attention mechanism, and combined with an infrared temperature sensor for fire source verification. The multimodal intelligent perception system integrates UWB ultra-wideband positioning modules and inertial sensors to achieve centimeter-level positioning, and fuses LiDAR and binocular vision data to complete terrain scanning, path planning and millisecond-level obstacle avoidance through dynamic window algorithms. The Archimedes wind turbine (4) and the PVT thermophotovoltaic panel (3) form a wind-solar-thermal co-generation energy supply system; The double wishbone suspension (11), the electric motor drive axle (8), and the aluminum profile body (5) form an all-terrain adaptable chassis structure; The robotic arm (1) and the emergency fire extinguishing nozzle (10) form an integrated fire extinguishing device. The robotic arm (1) is used to grab flammable pollutants of different sizes in the forest area. The emergency fire extinguishing nozzle (10) can be aimed at the root of the fire source and has the function of pre-wetting the grabbed pollutants.

2. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability as described in claim 1, characterized in that, The wind-solar-thermal co-generation system can support continuous operation for no less than 72 hours without an external power grid, and the overall thermal-electric efficiency of the unit reaches more than 64.5%. The Archimedes wind turbine enables omnidirectional air intake and low-wind-speed start-up, while the photovoltaic PVT module integrates a liquid-cooled heat dissipation structure to recover waste heat from the battery cells. The intelligent collaborative control module dynamically schedules the power supply priority of wind and solar energy based on real-time wind speed and energy fluctuations.

3. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability as described in claim 1, characterized in that, The RF-YOLO fire detection system is deployed on an edge computing platform and has an accuracy rate of no less than 98.2% and a recall rate of no less than 96.5% for detecting early small fires.

4. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability as described in claim 1, characterized in that, The all-terrain adaptable chassis structure improves the passability of complex forest terrain by at least 55% compared with traditional tracked or ordinary wheeled equipment, and the jamming rate is less than 10%.

5. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability according to claim 2, characterized in that, The integrated fire extinguishing and extinguishing device reduces the amount of fire extinguishing agent used by more than 30% through the coordinated control of the mechanical arm (1) grabbing and the emergency fire extinguishing nozzle (10) precisely spraying fine water mist.

6. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability according to claim 2, characterized in that, The Archimedes wind turbine can be replaced by a vertical axis wind turbine, which does not require a wind-fighting mechanism.

7. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability as described in claim 6, characterized in that, The double wishbone suspension (11) is divided into a front double wishbone independent suspension and a rear leaf spring beam non-independent suspension; The front double wishbone independent suspension can be replaced with a multi-link independent suspension, and the rear leaf spring beam non-independent suspension can be replaced with an air spring non-independent suspension. The air spring non-independent suspension is equipped with a height sensor and an electronic control unit.

8. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability according to claim 7, characterized in that, The obstacle avoidance perception unit, composed of lidar and binocular vision module, can be replaced by a combination of millimeter-wave radar and monocular camera. By training a forest obstacle sample database, an obstacle avoidance accuracy of no less than 92% can be achieved.

9. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability according to claim 8, characterized in that, The fire extinguishing mechanism consisting of the emergency fire extinguishing nozzle (10) and the robotic arm (1) can be replaced by a vehicle-mounted high-pressure water gun and an electric gimbal. The electric pan-tilt unit supports 360° horizontal rotation and ±90° pitch adjustment, and is automatically aligned with the fire source by the fire detection system.

10. The intelligent patrol vehicle with integrated fire suppression and all-terrain capability according to claim 9, characterized in that, The PVT thermal photovoltaic panel (3) can be replaced by a flexible thin-film photovoltaic module. The flexible thin-film photovoltaic module is installed in line with the curved surface of the vehicle body, and its weight does not exceed one-third of that of a rigid module with the same power. The loss of photoelectric conversion efficiency is compensated by a multi-panel series layout.

Images