Multi-terrain forest fire prevention troubleshooting system and method for cross-country travel
By designing a off-road marching forest fire prevention inspection system with track structure and multiple sensors, the monitoring blind spot problem caused by the fixed sensor position is solved, high-precision fire detection and timely alarm are achieved in complex terrain, and the scientificity and timeliness of fire monitoring are improved.
Patent Information
- Application Number
- CN202510216194.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-26
- Publication Date
- 2025-05-27
AI Technical Summary
In the prior art, due to the fixed position of the sensor, there are a large number of monitoring blind spots in complex layout places, which makes the fire difficult to be detected as soon as possible and delays the fire extinguishing opportunity.
A multi-terrain forest fire prevention inspection system for off-road marching was designed, using track structure and track material composited with high-strength rubber and metal reinforcement ribs, equipped with high-sensitivity smoke sensors, high-definition cameras, FLIR thermal imager and NVIDIA Jetson Xavier NX high-performance computing module to achieve autonomous movement and high-precision fire detection.
The system can move autonomously in complex terrain, effectively eliminate monitoring blind spots, timely detect fire hazards, and accurately locate the fire source through the fusion of intelligent algorithms and multi-sensor data, improving the accuracy and timeliness of fire monitoring.
Smart Images
Figure CN120039320A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fire prevention inspection, and specifically to a multi-terrain forest fire prevention inspection system and method for cross-country travel. Background Art
[0002] At present, with the continuous rapid development of society, the urbanization process is advancing at an unprecedented speed. All kinds of buildings are springing up, with their structures and functions becoming increasingly complex and diverse, and the degree of concentration of people and property is also rising sharply. However, along with this prosperous scene, the frequency of fires and the potential harm degree are constantly increasing, which undoubtedly brings an unprecedented severe challenge to the safety of life and property; At present, traditional fire detection means mainly rely on sensors at fixed positions, such as common smoke alarms and temperature sensors. These devices have indeed played a certain role in fire monitoring and can capture some key information in the initial stage of a fire. However, when faced with actual complex and changeable fire scenarios, their limitations are fully exposed; Firstly, due to the fixed positions of the sensors, in places with complex layouts such as large warehouses and multi-story buildings, there are a large number of monitoring blind spots. Once a fire breaks out quietly in these blind spots, it is very difficult to be detected in the first time, thus delaying the best fire extinguishing opportunity. Secondly, the fire scene is often accompanied by harsh environments such as thick smoke and high-temperature roasting, which will seriously affect the performance of the fixed sensors, resulting in inaccurate and untimely information obtained by them, and unable to provide a reliable basis for subsequent rescue operations. Thirdly, these fixed sensors can only provide local information of the location where they are located, and it is difficult to conduct all-round and dynamic monitoring of the entire fire scene. Firefighters cannot timely and accurately master key information such as the spread direction of the fire and the combustion intensity, which greatly affects the scientificity and timeliness of fire extinguishing and rescue decisions, is not conducive to the efficient implementation of fire fighting work, and is also difficult to comprehensively protect the safety of life and property; In view of the above-mentioned various difficulties, developing a device that can move autonomously and flexibly shuttle through the fire scene to achieve efficient and accurate fire detection has become an important issue that urgently needs to be solved in the current fire protection field. It is precisely in this context that the tracked fire inspection robot emerges with its unique advantages, bringing a new solution and hope to the fire prevention and control work. Summary of the Invention
[0003] (I) Technical Problems to be Solved In view of the deficiencies of the prior art, the present invention provides a multi-terrain forest fire prevention inspection system and method for off-road travel, which solves the problem that in the prior art, due to the fixed position of sensors, there are a large number of monitoring blind spots in places with complex layouts such as large warehouses and multi-story buildings. Once a fire breaks out quietly in these blind spots, it is very difficult to detect it in the first time, thus delaying the best fire extinguishing opportunity.
[0004] (2) Technical solutions To achieve the above objectives, the present invention is realized through the following technical solutions: A multi-terrain forest fire prevention inspection method for off-road travel includes the following steps: S1. Check various components of the off-road robot to ensure that all indicators of the off-road robot reach the best state, and then start the off-road robot to patrol in the multi-terrain forest; S2. Select a track material made of high-strength rubber and metal reinforcing ribs. On the surface of the track, use a special vulcanization process to make serrated anti-slip patterns. The depth and spacing of the patterns are calculated through mechanical simulation to adapt to different ground conditions such as muddy, slippery, and rugged in the forest; S3. Adopt a dual-motor drive mode, with each motor connected to a driving wheel respectively, and provide strong torque through a speed reducer; S4. Use finite element analysis software to simulate the layout of components such as the battery, sensors, and control module inside the robot to ensure that the center of gravity of the robot always remains within a safe range during complex movements such as climbing slopes and crossing ditches, preventing rollover; S5. During the patrol, the highly sensitive smoke sensor inside the robot monitors smoke to work, and the high-definition camera equipped at the front end has automatic focusing and night vision functions, and is paired with a FLIR thermal imager to accurately capture the thermal signal of the fire source in the dark or smoky environment; S6. The Nvidia Jetson Xavier NX high-performance computing module is simultaneously used inside the robot as the data processing core, equipped with the TensorFlow deep learning framework and a customized fire detection algorithm. The sensor data is transmitted to the processing unit through the CAN bus. The algorithm analyzes the data in real time, uses a convolutional neural network to identify the flame characteristics, and judges the fire development stage through the smoke concentration and temperature change trend; S7. Use 4G / 5G network to upload the fire scene information to the cloud server, and install a high-decibel speaker on the top of the robot. When a fire is detected, the speaker plays a pre-recorded voice warning message, including the fire location, fire intensity, and escape route.
[0005] A multi-terrain forest fire prevention inspection system for off-road travel includes: Robot body: It consists of a crawler unit, a drive and suspension unit, a center of gravity optimization unit, a sensor unit, a data processing unit, and a communication and warning unit; Crawler unit: The crawler material made of high-strength rubber and metal reinforcing ribs is selected and manufactured through an integrated molding process by a mold to ensure the durability and strength of the crawler. On the surface of the crawler, special vulcanization technology is used to produce serrated anti-slip patterns, and the depth and spacing of the patterns are calculated through mechanical simulation to adapt to different ground conditions such as muddy, slippery, and rugged in the forest; Drive and suspension unit: Adopt a dual-motor drive mode, with each motor connected to a driving wheel respectively, and a reducer is used to provide strong torque; Center of gravity optimization unit: Use finite element analysis software to simulate the layout of components such as batteries, sensors, and control modules. Place the heavier battery at the center position at the bottom of the robot, evenly distribute the sensors around the fuselage, and install the control module in a stable area close to the center of gravity. In this way, ensure that the center of gravity of the robot always remains within a safe range during complex movements such as climbing slopes and crossing gullies, preventing rollover; Sensor unit: Select high-sensitivity smoke sensors to monitor smoke, equipped with a high-definition camera at the front end, with automatic focusing and night vision functions, and equipped with a FLIR thermal imager, which can accurately capture the thermal signal of the fire source in the dark or smoky environment; Data processing unit: Use the NVIDIA Jetson Xavier NX high-performance computing module as the data processing core, equipped with the TensorFlow deep learning framework and a customized fire detection algorithm. Transmit the sensor data to the processing unit through the CAN bus. The algorithm analyzes the data in real time, uses a convolutional neural network to identify the flame characteristics, and judges the development stage of the fire through the change trend of smoke concentration and temperature; Communication and warning unit: Adopt a redundant communication method combining 4G / 5G communication modules and Wi-Fi modules, and install a high-decibel speaker on the top of the robot. When a fire is detected, the speaker plays a pre-recorded voice warning message, including the fire location, the size of the fire, and the escape route.
[0006] Preferably: In the communication and warning unit, in the forest, preferably use the 4G / 5G network to upload the fire scene information to the cloud server. When the 4G / 5G signal is poor, automatically switch to the Wi-Fi hotspot mode to communicate with nearby base stations or relay devices. At the same time, equipped with a LoRa long-distance and low-power communication module as a backup communication means to ensure smooth communication in extreme cases.
[0007] Preferably, in the communication warning unit, when a fire occurs, the cloud server pushes the fire information to the monitoring platform of the forest fire prevention command center and the mobile APP of the patrol personnel. The APP interface displays real-time data such as the video, temperature, and smoke concentration at the fire scene, and provides a navigation function to guide the rescue personnel to quickly reach the fire scene.
[0008] (III) Beneficial effects The present invention provides a multi-terrain forest fire prevention inspection system and method for off-road travel, having the following beneficial effects: 1. By setting a crawler structure on the forest fire prevention inspection robot, the crawler can easily roll forward with its large contact area with the ground and strong driving force. Unlike wheeled devices, it is not easy to get stuck in soft soil or blocked by obstacles. When climbing a steep mountain slope, the close fit between the crawler and the ground provides sufficient friction to ensure the robot climbs stably upward, effectively avoiding side slips. At the same time, the adjustable suspension system equipped on the robot can automatically adjust the tension and angle of the crawler according to the terrain undulation, keeping the fuselage always stable, ensuring the normal operation of various sensors and devices, greatly expanding the action range of the robot in the forest, and realizing all-round coverage monitoring of vast forest areas; 2. The robot of the present invention integrates a variety of advanced high-precision sensors, such as a highly sensitive smoke sensor, which can capture extremely subtle smoke signals in the forest, and even the thin smoke generated by an incipient fire in the distance can be detected in time; the infrared temperature sensor is extremely sensitive to the temperature change in the forest environment and can quickly detect local areas with abnormal temperature rise caused by a fire source. With a high-definition camera and a thermal imager, the camera can analyze and judge abnormal situations in the forest through image recognition technology, and the thermal imager can clearly display high-temperature heat sources at night or when the line of sight is blocked by dense vegetation, accurately positioning the fire source location; 3. The present invention has a highly intelligent autonomous decision-making ability. It can automatically plan the optimal travel route according to the real-time acquired terrain information, fire data, and its own position status by using the built-in path planning algorithm. When encountering dangerous situations such as a sudden increase in the fire intensity or a change in the wind direction, the robot can quickly respond, autonomously adjust the travel direction, avoid dangerous areas, and continue to complete the fire inspection task while ensuring its own safety. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 is a framework diagram of the present invention.
[0010] Figure 2 is a schematic structural diagram of the crawler off-road travel forest fire prevention inspection robot of the present invention. DETAILED DESCRIPTION OF THE INVENTION; Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
[0012] Embodiment: As Figure 1-2 shown, the embodiment of the present invention provides a multi-terrain forest fire prevention inspection method for off-road travel, including the following steps: S1. Check various components of the off-road robot to ensure that all indicators of the off-road robot reach the best state, and then start the off-road robot to enter the multi-terrain forest for patrol; S2. Select a track material that combines high-strength rubber and metal reinforcing ribs. On the surface of the track, use a special vulcanization process to make serrated anti-slip patterns. The depth and spacing of the patterns are calculated through mechanical simulation to adapt to different ground conditions such as muddy, slippery, and rugged in the forest; S3. Adopt a dual-motor drive mode, with each motor connected to a driving wheel respectively, and provide strong torque through a speed reducer; S4. Use finite element analysis software to simulate the layout of components such as the battery, sensors, and control modules inside the robot, and ensure that the center of gravity of the robot always remains within a safe range during complex movements such as climbing slopes and crossing gullies to prevent rollover; S5. During the patrol process, the highly sensitive smoke sensor inside the robot monitors the smoke to work, and the high-definition camera equipped at the front end has automatic focusing and night vision functions, and is paired with a FLIR thermal imager, which can accurately capture the thermal signal of the fire source in the dark or smoky environment; S6. The robot internally adopts the NVIDIA Jetson Xavier NX high-performance computing module as the data processing core, is equipped with the TensorFlow deep learning framework and a customized fire detection algorithm, transmits the sensor data to the processing unit through the CAN bus, the algorithm analyzes the data in real time, uses a convolutional neural network to identify the flame characteristics, and judges the fire development stage through the smoke concentration and temperature change trend; S7. Use the 4G / 5G network to upload the fire scene information to the cloud server, and install a high-decibel speaker on the top of the robot. When a fire is detected, the speaker plays a pre-recorded voice warning message, including the fire location, fire size, and escape route.
[0013] Aiming at the problem of poor terrain adaptability of traditional wheeled flammable object detection robots, the present invention adopts a crawler traveling structure. Compared with the wheeled structure, this structure has a larger ground contact area and smaller pressure. In complex terrains such as the ruins after a fire and uneven construction sites, it can move forward stably, is not easy to slip or get stuck in gaps, greatly expands the action range of the robot at the fire scene, enables it to penetrate into various areas, and comprehensively detect flammable objects. At the same time, the equipped robotic arm endows the robot with the ability to handle detected flammable objects such as carrying and isolating them, effectively reducing the risk of fire from the source.
[0014] Facing the dilemma of limited monitoring range and lack of autonomous action ability of the fixed-point visual fire monitoring system, the robot of the present invention can move autonomously. It can shuttle through all corners of the building, effectively eliminate monitoring blind spots, and timely discover hidden flammable object hazards. Moreover, through advanced sensors and intelligent algorithms, the robot can actively conduct investigations and treatments on flammable objects, no longer limited to detection and alarm only, and truly achieve active fire prevention and control.
[0015] Aiming at the disadvantages of insufficient visual recognition ability and poor robotic arm performance of bionic multi-legged flammable object handling robots, the present invention is equipped with a high-precision visual recognition system. This system integrates advanced image recognition algorithms and multi-sensor data fusion technology. In complex environments such as dim light and smoke-filled, it can accurately identify various flammable objects, greatly improving the accuracy of detection. In addition, the carefully designed robotic arm has strong grasping force and extremely high flexibility, can easily handle various types and weights of flammable objects, and efficiently complete the work of carrying and investigation, significantly improving the efficiency of large-area fire hazard investigation.
[0016] Through the all-round optimization of the overall performance of the robot, the present invention is committed to filling the technical gap of existing fire hazard investigation equipment and providing more reliable and efficient technical support for ensuring life and property safety.
[0017] Regarding the multi-terrain forest fire prevention investigation system for cross-country travel, including: Robot body: It consists of a crawler unit, a drive and suspension unit, a center of gravity optimization unit, a sensor unit, a data processing unit, and a communication and warning unit. The entire crawler robot integrates a variety of high-precision sensors, including a highly sensitive smoke sensor that can detect extremely low concentrations of smoke particles; it is equipped with a high-definition camera and a thermal imager, which use image recognition algorithms to identify objects in the scene and determine whether they are flammable, assisting in discovering potential fire sources. Once a fire hazard is detected, the robotic arm carried by the robot starts to work. The robotic arm adopts a multi-joint design, with a flexible range of motion and a powerful grasping force. Through the visual positioning system, the robotic arm can accurately grasp flammable objects and transport them to a safe area for isolation. At the same time, the robot is equipped with an intelligent control system that can automatically plan the best action path according to the detected hazard situation and efficiently complete the inspection and handling work; Crawler unit: The crawler material made of high-strength rubber and metal reinforcing ribs is selected and manufactured through an integrated molding process using a mold to ensure the durability and strength of the crawler. On the surface of the crawler, special vulcanization technology is used to produce serrated anti-slip patterns, and the depth and spacing of the patterns are calculated through mechanical simulation to adapt to different ground conditions such as muddy, slippery, and rugged terrains in the forest; Drive and suspension unit: It adopts a dual-motor drive mode, with each motor connected to a driving wheel respectively, and a reducer provides powerful torque; Center of gravity optimization unit: Use finite element analysis software to simulate the layout of components such as batteries, sensors, and control modules. Place the heavier battery at the center position at the bottom of the robot, evenly distribute the sensors around the fuselage, and install the control module in a stable area close to the center of gravity. In this way, ensure that the center of gravity of the robot always remains within a safe range during complex movements such as climbing slopes and crossing ditches, preventing rollovers; Sensor unit: Select a highly sensitive smoke sensor to monitor smoke, equipped with a high-definition camera at the front end with automatic focus and night vision functions, and paired with a FLIR thermal imager, which can accurately capture the thermal signal of the fire source in the dark or smoky environment; Data processing unit: Use the NVIDIA Jetson Xavier NX high-performance computing module as the data processing core, equipped with the TensorFlow deep learning framework and a customized fire detection algorithm. Transmit sensor data to the processing unit through the CAN bus. The algorithm analyzes the data in real time, uses a convolutional neural network to identify flame characteristics, and judges the development stage of the fire through the change trend of smoke concentration and temperature; Communication and warning unit: Adopt a redundant communication method combining 4G / 5G communication modules and Wi-Fi modules, and install a high-decibel speaker on the top of the robot. When a fire is detected, the speaker plays a pre-recorded voice warning message, including the fire location, fire size, and escape route.
[0018] In the communication warning unit, in the forest, the 4G / 5G network is preferentially used to upload the fire scene information to the cloud server. When the 4G / 5G signal is poor, it automatically switches to the Wi-Fi hotspot mode to communicate with nearby base stations or relay devices. At the same time, a LoRa long-distance low-power communication module is equipped as a backup communication means to ensure smooth communication even in extreme cases.
[0019] In the communication warning unit, when a fire occurs, the cloud server pushes the fire information to the monitoring platform of the forest fire prevention command center and the mobile APP of the patrol personnel. The APP interface displays real-time data such as the video, temperature, and smoke concentration of the fire scene, and provides a navigation function to guide the rescue personnel to quickly reach the fire scene. In the complex and changeable forest environment, the tracked cross-country forest fire prevention inspection robot demonstrates a highly intelligent autonomous decision-making ability. It can automatically plan the optimal travel route based on the terrain information, fire data, and its own position status obtained in real time using the built-in path planning algorithm. When encountering dangerous situations such as a sudden increase in the fire intensity or a change in the wind direction, the robot can quickly respond, autonomously adjust the travel direction, avoid dangerous areas, and continue to complete the fire inspection task while ensuring its own safety. This autonomous decision-making ability enables the robot to efficiently and quickly execute tasks in the forest without excessive manual intervention, greatly improving the response speed. Compared with the traditional manual patrol and inspection method, the robot can cover a larger forest area in a shorter time, detect and report fire hazards in a timely manner, gain valuable time for forest fire prevention work, effectively reduce the losses caused by forest fires, and become a reliable guardian of forest resources and ecological security.
[0020] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A multi-terrain forest fire prevention and investigation method for off-road travel, characterized in that: The following steps are involved: S1. Check various components of the off-road robot to ensure that all indicators of the off-road robot are in the best condition, and then start the off-road robot to patrol in the multi-terrain forest; S2. The track material is a composite of high-strength rubber and metal reinforcement ribs. A special vulcanization process is used to produce serrated anti-skid patterns on the track surface. The depth and spacing of the patterns are calculated through mechanical simulation to adapt to different ground conditions such as muddy, slippery, and rugged in the forest. S3, adopts dual-motor drive mode, each motor is connected to a driving wheel, and provides strong torque through the reducer; S4. Use finite element analysis software to simulate the layout of the robot's internal components such as batteries, sensors, and control modules to ensure that the robot's center of gravity always remains within a safe range to prevent rollover when performing complex actions such as climbing slopes and crossing gullies; S5. During patrol, the highly sensitive smoke sensor inside the robot monitors the smoke for work, while the high-definition camera at the front end has autofocus and night vision functions, and is equipped with a FLIR thermal imager, which can accurately capture the heat signal of the fire source in a dark or smoky environment; The S6 robot uses NVIDIA's Jetson Xavier NX high-performance computing module as the data processing core, equipped with the TensorFlow deep learning framework and a customized fire detection algorithm. The sensor data is transmitted to the processing unit via the CAN bus. The algorithm analyzes the data in real time, uses a convolutional neural network to identify flame characteristics, and determines the fire development stage based on the smoke concentration and temperature change trend. S7. Use 4G / 5G network to upload fire scene information to the cloud server, and install a high-decibel speaker on the top of the robot. When a fire is detected, the speaker plays a pre-recorded voice warning information, including the fire location, fire size and escape route.
2. A multi-terrain forest fire prevention and inspection system for off-road travel, characterized by: include: Robot body: It consists of a track unit, a drive and suspension unit, a center of gravity optimization unit, a sensor unit, a data processing unit, and a communication and early warning unit; Track unit: The track material is a composite of high-strength rubber and metal reinforcement ribs, and is manufactured through a mold-integrated molding process to ensure the durability and strength of the track. A special vulcanization process is used to produce a serrated anti-skid pattern on the track surface. The depth and spacing of the pattern are calculated through mechanical simulation to adapt to different ground conditions in the forest, such as muddy, slippery, and rugged. Drive and suspension unit: adopts dual-motor drive mode, each motor is connected to a driving wheel, and provides strong torque through the reducer; Center of gravity optimization unit: Finite element analysis software is used to simulate the layout of components such as batteries, sensors, and control modules. The heavier battery is placed at the center of the bottom of the robot, the sensors are evenly distributed around the body, and the control module is installed in a stable area close to the center of gravity. In this way, the center of gravity of the robot is always kept within a safe range to prevent rollover when performing complex actions such as climbing slopes and crossing gullies. Sensor unit: A highly sensitive smoke sensor is used to monitor smoke. The front end is equipped with a high-definition camera with autofocus and night vision functions. It is equipped with a FLIR thermal imager to accurately capture the heat signal of the fire source in a dark or smoky environment. Data processing unit: NVIDIA Jetson Xavier NX high-performance computing module is used as the data processing core, equipped with TensorFlow deep learning framework and customized fire detection algorithm, sensor data is transmitted to the processing unit through CAN bus, the algorithm analyzes the data in real time, uses convolutional neural network to identify flame characteristics, and determines the fire development stage through smoke concentration and temperature change trend; Communication warning unit: It adopts a redundant communication method that combines 4G / 5G communication module with Wi-Fi module, and installs a high-decibel speaker on the top of the robot. When a fire is detected, the speaker plays a pre-recorded voice warning information, including the location of the fire, the size of the fire and the escape route.
3. The multi-terrain forest fire prevention and inspection system for off-road travel according to claim 2 is characterized in that: In the communication warning unit, in the forest, 4G / 5G network is used preferentially to upload fire scene information to the cloud server. When the 4G / 5G signal is poor, it automatically switches to Wi-Fi hotspot mode to communicate with nearby base stations or relay devices. At the same time, it is equipped with a LoRa long-distance and low-power communication module as a backup communication means to ensure smooth communication even in extreme situations.
4. The multi-terrain forest fire prevention and inspection system for off-road travel according to claim 2 is characterized in that: In the communication warning unit, when a fire occurs, the cloud server pushes the fire information to the monitoring platform of the forest fire prevention command center and the patrol personnel's mobile phone APP. The APP interface displays the video, temperature, smoke concentration and other data of the fire scene in real time, and provides navigation function to guide rescue personnel to quickly reach the fire scene.