Small self-sensing intelligent special vehicle and control method

Abstract

The invention relates to the technical field of intelligent fire rescue, and discloses a small self-sensing intelligent special vehicle and a control method.The small self-sensing intelligent special vehicle is characterized in that a plurality of supporting foot devices are connected to the outer side face of a lower bottom plate, a gearbox is connected to the terminal of the top of the lower bottom plate, plunger pumps are connected to the output ends of the gearbox, and the number of the plunger pumps corresponds to that of the supporting foot devices; the top of the upper top plate is connected with a recognition device, and a spraying device corresponding to the fire extinguishing agent storage box is arranged at the front end of the top of the upper top plate; one side of the top end of the middle plate is connected with a fire extinguishing agent storage box, and the middle end of the top face of the middle plate is connected with a steering assembly. According to the device, the change of a fire source can be recognized in time in a complex environment, and accurate and effective fire extinguishing measures are taken. According to the device, a brand new thought and method are brought to forest fire fighting, the fire source checking and extinguishing efficiency is improved, secondary reburning of the fire is avoided, and the overall safety and emergency response capacity of fire fighting operation are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent fire and rescue technology, specifically to a small, self-sensing intelligent special vehicle and its control method. Background Technology

[0002] In recent years, global warming and increasingly severe droughts have led to frequent forest fires, posing a serious threat to the ecological environment and human society. Forest fires not only destroy large areas of forest resources but also disrupt the ecological balance, further triggering a series of environmental problems such as soil erosion and climate change. Therefore, preventing and extinguishing forest fires has become a major challenge that urgently needs to be addressed in today's society.

[0003] However, traditional forest firefighting equipment, such as large fire trucks and helicopters, often has limitations in responding to forest fires. While large fire trucks possess powerful firefighting capabilities, they are limited by the complex terrain and inconvenient transportation in forests, making it difficult to reach the fire scene quickly, and their ability to carry firefighting resources is also limited. Helicopters, on the other hand, have strong mobility, but due to their high cost and weather conditions, they are often unable to participate in firefighting operations for extended periods or efficiently. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a small, self-sensing intelligent special vehicle and its control method, which has advantages such as improving the efficiency of fire source detection and extinguishing, preventing secondary reignition of fires, and enhancing the overall safety and emergency response capabilities of firefighting operations, thus solving the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a small self-sensing intelligent special vehicle and a control method, comprising: The lower base plate has multiple support legs connected to its outer side. A gearbox is connected to the top end of the lower base plate. Each output end of the gearbox is connected to a plunger pump, and the number of support legs corresponds to the number of output legs. The top plate is equipped with an identification device connected to its top, and a spraying device is provided at the front end of the top of the top plate corresponding to the fire extinguishing agent storage tank. The middle plate has a fire extinguishing agent storage tank connected to one side of its top, a steering assembly connected to the middle of the top surface of the middle plate, and a control module and a battery module installed on one side of the top of the middle plate. The top plate, middle plate, and bottom plate are fixedly connected from top to bottom by support columns.

[0006] As a preferred embodiment of the present invention, the identification device includes a camera bracket, a camera, a first servo motor and a support rod, and a lidar; The first servo motor consists of two parts. One servo motor is fixedly connected to the support rod and installed on the top of the upper plate. The other servo motor is fixedly installed on one side of the top of the support rod and connected to the camera bracket and the lidar. The camera and lidar are fixed to the side of the camera bracket.

[0007] As a preferred technical solution of the present invention, the spraying device includes a nozzle, a support gimbal, a second servo motor, and a servo motor connector; The support gimbal is fixedly installed on one side of the top of the upper plate, and the support gimbal can extend and retract vertically and rotate 360 ​​degrees. The second servo motor and the servo motor connector are rotatably connected. The servo motor connector is fixedly installed at the corresponding position of the support gimbal. The nozzle is fixedly installed on the top of the second servo motor. The nozzle draws fire extinguishing agent liquid from the fire extinguishing agent storage tank and sprays it through a water pump fixedly installed on the top of the upper plate.

[0008] As a preferred embodiment of the present invention, the gearbox includes a housing, inside which is provided a first bevel gear that can be driven by a steering assembly. The outer end of the first bevel gear is connected to four second bevel gears. The output end of the second bevel gear is provided with two electromagnets, each containing a coil that can generate magnetism when energized. A coupling sleeve is provided between the two electromagnets, through which power is transmitted to a plunger pump. The second bevel gears are rotatably engaged with the housing.

[0009] As a preferred embodiment of the present invention, the steering assembly includes a key and a steering motor, wherein the output shaft of the steering motor is fixedly mounted with a key, and the steering motor is connected to a first bevel gear via the key.

[0010] As a preferred technical solution of the present invention, the number of the support legs is four, and they are equidistantly connected to the outer side of the bottom plate. The support legs include a connector, a piston cylinder, a robotic arm, a piston rod, a Mecanum wheel, and a motor. The connector is fixedly installed on the outer side of the bottom plate. A rotatable piston cylinder is connected to the side end of the connector. A mechanical arm is rotatably connected to the bottom of the side end of the connector. The mechanical arm is rotatably connected to the piston rod at the output end of the piston cylinder. A motor is provided on one side of the end of the mechanical arm. A Mecanum wheel is connected to the output end of the motor.

[0011] As a preferred embodiment of the present invention, the number of plunger pumps is four. Each plunger pump includes a main shaft, a central lever, plunger assemblies, a housing, and a rear cover. The main shaft is connected to the output end of the second bevel gear through a coupling sleeve. The main shaft is connected to the central lever and the six plunger assemblies, and is rotatably connected to the housing. The rear cover is spherically connected to the central lever, and is fixedly connected to the housing.

[0012] The present invention also provides a small self-sensing intelligent special vehicle and a control method thereof. Based on the above-mentioned small self-sensing intelligent special vehicle, the method includes the following steps: S1: Vehicle initialization, the control module sequentially performs self-tests and communication tests on the lidar, camera, first servo motor, plunger pump, electromagnet and battery module; S2: Receive fire zone information from the task input, generate a global path, and acquire real-time pose. S3: Perform global path planning, generate a global path from the current pose to the fire zone on the map, and achieve local obstacle avoidance and tracking while meeting the safety braking distance constraint; S4: When the vehicle arrives at the fire zone, thermal imaging images are collected by the camera, and fire source identification and center location are performed. Specifically, this includes: initial location and fine location. The initial location specifically involves obtaining the line of sight direction for the same heat source target and using the triangulation method to solve the initial position of the fire source center. The fine location specifically involves: spatially associating the point cloud obtained by the lidar with the candidate heat source area, extracting the corresponding point cloud set and calculating the three-dimensional coordinates of the fire source center, making a second correction to the initial location result, and finally outputting the fire source center position. S5: Drive the spraying device to complete alignment; S6: Perform a review after spraying is completed. That is, when the temperature peak of the fire source area decreases and remains stable below the preset threshold and meets the continuous frame requirement, the fire extinguishing is determined to be completed and the task is ended. Otherwise, S4-S5 are re-executed. Drive the vehicle back to the starting point when the battery is low or the mission is completed.

[0013] Compared with the prior art, the present invention provides a small self-sensing intelligent special vehicle and a control method, which has the following beneficial effects: This invention enables autonomous positioning and map construction by performing multi-sensor self-testing, time synchronization, and origin pose establishment after the vehicle is powered on. Upon receiving navigation information, it employs a synchronous positioning and mapping method tightly coupled with lidar and inertial measurement units. This involves correcting motion distortion in the point cloud through partitioned processing to unify the point cloud to a reference time. At the front-end lidar odometry, principal component analysis and mean clustering are used to extract feature points, complete ground extraction, and constraint construction. In the back-end optimization section, constraint factors from each stage are introduced to optimize the factor graph, resulting in a stable pose output. Based on this pose and map, a navigation obstacle avoidance algorithm is used to complete global path planning and local dynamic obstacle avoidance control, enabling the vehicle to autonomously navigate to a designated fire area in complex forest or obstacle environments. This allows for timely identification of fire source changes and the implementation of precise and effective firefighting measures in complex environments. This device will bring new ideas and methods to forest fire fighting, improve the efficiency of fire source investigation and extinguishing, prevent secondary reignition of fires, and enhance the overall safety and emergency response capabilities of firefighting operations. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the overall invention; Figure 2 This is a schematic diagram of the identification device of the present invention; Figure 3 This is a schematic diagram of the gearbox of the present invention; Figure 4 This is a schematic diagram of the support leg device of the present invention; Figure 5 This is a schematic diagram of the spraying device of the present invention; Figure 6 This is a schematic diagram of the steering component of the present invention; Figure 7 This is a schematic diagram of the plunger pump of the present invention; Figure 8 This is a diagram showing the second bevel gear of the present invention not mating with the housing; Figure 9 This is a diagram showing the fit between the second bevel gear and the housing of the present invention.

[0015] The components include: 1. Identification device; 2. Water pump; 3. Top plate; 4. Control module; 5. Battery module; 6. Gearbox; 7. Support leg device; 8. Spraying device; 9. Steering assembly; 10. Extinguishing agent storage tank; 11. Support column; 12. Plunger pump; 13. Lower base plate; 14. Middle plate; 101. Camera bracket; 102. Camera; 103. First servo motor; 104. Support rod; 601. First bevel gear; 602. Second bevel gear; 603. Electromagnetic... Iron; 604, coupling sleeve; 605, housing; 701, connector; 702, piston cylinder; 703, robotic arm; 704, piston rod; 705, Mecanum wheel; 706, motor; 801, nozzle; 802, support gimbal; 803, second servo motor; 804, servo motor connector; 901, key; 902, steering motor; 1201, spindle; 1202, center bar; 1203, plunger assembly; 1204, housing; 1205, rear cover. Detailed Implementation

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

[0017] Please see Figures 1-7A small self-sensing intelligent special vehicle and its control method are disclosed. The gearbox 6 is composed of a first bevel gear (34x20) 601 and four second bevel gears (17x2) 602. The four second bevel gears (17x2) 602 are rotatably engaged with the internal shaft holes of the housing 605. Each second bevel gear (17x2) 602 has two electromagnets 603, which are fixed on the internal circular holes of the housing 605 and the circular holes of the housing wall, respectively. The two electromagnets 603 are energized to generate magnetism to control the sliding of the coupling sleeve 604. The coupling sleeve 604 is slidably engaged with the main shaft 1201 of the plunger pump 12 structure of the whole vehicle assembly. The main shaft 1201 is engaged with the pin of the second bevel gear (17x2) 602 and rotatably engaged with the hole of the housing 605. The support device 7 is rotatably connected to the piston cylinder 702 and the robotic arm 703 by the connector 701. The piston cylinder 702 is slidably engaged with the piston rod 704, and 704 is rotatably connected to the robotic arm 703 by the slidably engaged piston rod 704. The Mecanum wheel 705 is rotatably connected to the motor 706, and the motor 706 is fixed to the robotic arm 703. Together, they form the support device. The spraying device 8 consists of a nozzle 801 fixed on a second servo motor 803, the second servo motor 803 being rotatably connected to a servo motor connector 804, and then fixedly connected to a support gimbal 802 through the servo motor connector 804. The steering assembly 9 is fixed to the steering motor 902 by a key 901; The plunger pump 12 consists of a main shaft 1201, six plunger assemblies 1203, and a central lever 1202 that spherically engages with the housing 1204. The rear cover 1205 spherically engages with the central lever 1202 and is fixedly connected to the housing 1204.

[0018] The identification device 1, water pump 2, and spraying device 8 are fixedly connected to the top plate 3 to form the upper structure of the vehicle. The control module 4, battery module 5, steering assembly 9, and fire extinguishing agent storage tank 10 are fixedly connected to the middle plate 14 to form the middle structure of the vehicle. The gearbox 6, four support feet 7, and four plunger pumps 12 are fixedly connected to the lower base plate 13. The gearbox 6 and the four plunger pumps 12 are rotatably connected to form the lower structure of the vehicle. The upper structure and the middle structure of the vehicle, and the middle structure and the lower structure of the vehicle are fixedly connected by the support column 11. The steering motor 902 of the steering assembly 9 is connected to the first bevel gear 601 of the gearbox 6 via key 901, forming the lower base plate 13 of the intelligent four-legged terrain vehicle. Multiple support feet 7 are connected to the outer side of the lower base plate 13. The gearbox 6 is connected to the top end of the lower base plate 13. Each output end of the gearbox 6 is connected to a plunger pump 12, and the number of pumps corresponds to the number of support feet 7. The top plate 3 is connected to the top of the top plate 3 and the top front end of the top plate 3 is equipped with a spraying device 8 corresponding to the fire extinguishing agent storage box 10. The middle plate 14 has a fire extinguishing agent storage tank 10 connected to one side of its top, a steering assembly 9 connected to the middle of its top surface, and a battery module 5 installed on the middle plate 14. The top plate 3, the middle plate 14, and the bottom plate 13 are fixedly connected from top to bottom by the support column 11.

[0019] The identification device 1 includes a camera bracket 101, a camera 102, a first servo motor 103, and a support rod 104; There are two first servo motors 103. One first servo motor 103 is fixedly connected to the support rod 104 and installed on the top of the upper plate 3. The other first servo motor 103 is fixedly installed on one side of the top of the support rod 104 and connected to the camera bracket 101. The camera 102 is fixed on the side of the camera bracket 101.

[0020] The spraying device 8 includes a nozzle 801, a support gimbal 802, a second servo motor 803, and a servo motor connector 804; The support gimbal 802 is fixedly installed on one side of the top of the upper plate 3, and the support gimbal 802 can extend and retract vertically and rotate 360 ​​degrees. The second servo motor 803 and the servo motor connector 804 are rotatably connected. The servo motor connector 804 is fixedly installed at the corresponding position of the support gimbal 802. The nozzle 801 is fixedly installed on the top of the second servo motor 803. The nozzle 801 draws fire extinguishing agent liquid from the fire extinguishing agent storage tank 10 and sprays it through the water pump 2 fixedly installed on the top of the upper plate 3.

[0021] The gearbox 6 includes a housing 605, inside which is a first bevel gear 601 that can be driven by the steering assembly 9. The outer end of the first bevel gear 601 is connected to four second bevel gears 602. The output end of the second bevel gears 602 is provided with two electromagnets 603, and a coupling sleeve 604 is provided between the two electromagnets 603. When the two electromagnets 603 are energized, they generate magnetism to control the sliding of the coupling sleeve 604. Power is transmitted to the plunger pump 12 through the coupling sleeve 604. The second bevel gears 602 and the housing 605 rotate in cooperation.

[0022] The steering assembly 9 includes a key 901 and a steering motor 902. The key 901 is fixedly mounted on the output shaft of the steering motor 902, and the steering motor 902 is connected to the first bevel gear 601 through the key 901.

[0023] The number of support legs 7 is four, and they are equidistantly connected to the outer side of the lower base plate 13. The support legs 7 include a connector 701, a piston cylinder 702, a robotic arm 703, a piston rod 704, a Mecanum wheel 705, and a motor 706. The connector 701 is fixedly installed on the outer side of the bottom plate 13. A rotatable piston cylinder 702 is connected to the side end of the connector 701. A mechanical arm 703 is rotatably connected to the bottom of the side end of the connector 701. The mechanical arm 703 is rotatably connected to the piston rod 704 provided at the output end of the piston cylinder 702. A motor 706 is provided on one side of the end of the mechanical arm 703. A Mecanum wheel 705 is connected to the output end of the motor 706.

[0024] There are four plunger pumps 12. Each plunger pump 12 includes a main shaft 1201, a center rod 1202, plunger assemblies 1203, a housing 1204, and a rear cover 1205. The main shaft 1201 is connected to the output end of the second bevel gear 602 through the engagement sleeve 604. The main shaft 1201 is connected to the center rod 1202 and the six plunger assemblies 1203, and is rotatably connected to the housing 1204. The rear cover 1205 is spherically engaged with the center rod 1202, and is fixedly connected to the housing 1204. The gearbox 6 and plunger pump 12 are crucial for the intelligent vehicle's adaptability to terrain. When the vehicle is driving in mountainous terrain with varying terrain, if it encounters complex terrain, the camera 102 will capture and identify the mountainous terrain and transmit the data to the control module 4. The control module 4 will determine whether the outriggers 7 need to be raised (or lowered) to pass through the terrain, and then control the steering motor 902 to rotate forward (or in the opposite direction). This will drive the first bevel gear 601 to rotate via the key 901, and the second bevel gear 602, which meshes with the first bevel gear 601, will also rotate accordingly.

[0025] If the outrigger 7 needs to be raised (or lowered) to traverse the terrain, the control module 4 controls the electromagnet 603 to be energized, causing the coupling sleeve 604 to move, connecting the second bevel gear 602 and the main shaft 1201, causing the main shaft 1201 of the plunger pump 12 to rotate forward (or backward). The plunger assembly 1203 pumps hydraulic oil into (or out) the piston cylinder 702, changing the hydraulic pressure inside the piston cylinder 702, driving the piston rod 704 to perform piston movement, thereby raising (or lowering) the outrigger 7. Moreover, the Mecanum wheel 705 on the outrigger 7 can turn in place, smoothly and quickly traversing various complex terrains to reach the fire scene, and extinguishing the fire through the nozzle 801 and the water pump 2.

[0026] The control method of the present invention preferably adopts a closed-loop control system of positioning and mapping, navigation and obstacle avoidance, identification and positioning, fire spraying and extinguishing, and verification and return. Its specific implementation includes the following steps: S1: Control initialization and return-to-origin establishment: After the vehicle is powered on, the controller sequentially performs self-tests and communication tests on the lidar (105), camera (102), first servo motor (103), plunger pump (12), electromagnet (603), and battery module management module; completes signal time synchronization, and records the vehicle address or loaded address at the initialization time as the origin pose for subsequent return positioning and docking after the mission ends. Preferably, when a key sensor is detected to be offline or the battery level is lower than the minimum start-up threshold, the controller enters protection mode, prohibits the execution of tasks, and remains on standby. S1.1 Start controller performs self-test and communication link test on the detection unit, thermal imaging camera, chassis drive, battery module management system and spraying actuator; S1.2 completes multi-sensor time synchronization, loads the extrinsic parameter calibration matrix, and loads the vehicle position or return point at the initialization time. ( ); Indicates the origin point position in the map system. The quaternion representing the origin attitude; S2 (Receive task and point cloud partition motion distortion correction); The system receives fire zone information from a host computer or a preset task. The fire zone can be one of a target coordinate point, a target polygonal region, or a grid region. Based on this, a navigation target and an initial global path are generated. As the vehicle moves toward the target area, it simultaneously collects laser point cloud and inertial data. To improve positioning and mapping accuracy, this embodiment preferably adopts a point cloud motion distortion correction method with partitioned processing: within a single frame point cloud scanning cycle, the point cloud is divided into multiple partitions based on the point cloud timestamp or scanning angle. Combined with the pose sequence calculated by the inertial measurement unit, the point cloud of each partition is interpolated and corrected so that the point clouds at different sampling times are unified to the same reference time coordinate system, thereby reducing the point cloud stretching and distortion caused by the movement of the carrier. S2.1 receives information from the host computer or a preset task and generates a global path; S2.2 Based on laser-acquired point cloud data, the point cloud is divided into sections using the point cloud timestamp or azimuth angle. Each partition is used to interpolate and calibrate the partially lost data to obtain the time and location of each data point. relative to reference time position and sampling points Distortion correction is performed to obtain the reference time and the calibration point. ,satisfy: ; in, Indicates the reference time position The inverse matrix; S3 (Obstacle Avoidance and Path Tracking Control): The controller performs navigation control based on the real-time pose and map output in step S2, preferably including two levels: global path planning and local obstacle avoidance control: (1) Global path planning: Generate a global path from the current pose to the fire zone on the map or grid cost map, and update it dynamically according to the path traversability. (2) Local obstacle avoidance and tracking: Real-time detection of obstacles (tree trunks, shrubs, falling rocks, slopes and impassable areas, etc.), and generation of local control commands based on safe distance and feasible speed range to achieve dynamic detour, deceleration, stopping or replanning. The vehicle uses Mecanum wheels to achieve omnidirectional movement. The controller outputs speed and steering commands to the chassis motor and steering components, so that the vehicle can maintain stable tracking and safely approach the boundary of the target fire area in narrow passages or dense obstacle environments; S3.1: In the global path Under the guidance of local obstacle avoidance generation control quantity ( ), through dynamic window constraints: in, Indicates linear velocity; Indicates angular velocity; This represents the lower limit of linear velocity. This represents the upper limit of linear velocity; Indicates the lower limit of angular velocity. This indicates the upper limit of angular velocity. Indicates the maximum linear velocity; Indicates the maximum angular velocity; Indicates the current linear velocity. Indicates the current angular velocity; Indicates the control period or prediction time step; S3.2: For each group ( Predict short-term trajectories and score them, then select the chassis instruction that minimizes the cost, whose function satisfies: in, The cost of approaching the target / distance to the target; Minimum distance from obstacles or the cost of collision risk; Speed ​​preference factor; Cost of deviating from the global path; This represents the normalized weight coefficients; And meet the safety braking distance constraints: in, To predict the minimum distance from the trajectory to the nearest obstacle; Indicates the linear velocity of the current candidate trajectory; Indicates the safety margin distance; S4 (Fire Source Identification and Center Location): Once the vehicle arrives at the fire zone or a pre-set observation point, the identification device acquires thermal imaging images and performs fire source identification and center location: (1) Fire source identification: The maximum inter-class variance image segmentation algorithm is preferred to perform threshold segmentation on the thermal imaging image to obtain candidate regions of heat source; then, the candidate regions are screened by combining the region area threshold, temperature peak threshold, morphological features and continuous frame consistency rules to eliminate false detection targets such as heat reflection and thermal noise. Acquire thermal imaging images and perform grayscale or temperature mapping; then determine the threshold using the maximum inter-class variance segmentation algorithm. : in; For inter-class variance; The percentage of grayscale pixels at low and high temperatures are respectively represented. These are the average grayscale values ​​for low and high temperatures, respectively. (2) Center positioning: A two-stage positioning method is preferred. The first stage is initial positioning: the vehicle selects at least two different observation poses within a safe distance range, obtains the line of sight direction for the same heat source target, and uses the triangle positioning method to solve for the initial position of the fire source center; Under multiple observation poses that meet safe distance requirements, the vehicle extracts pixel centers from candidate heat source regions and calculates the corresponding line-of-sight directions. The initial position of the heat source center is then determined using a triangulation method. satisfy: in, The vehicle's position on the map system during the two observations; Observe the corresponding fire source direction angle twice; The distance scale along the line of sight between two observations; Initial estimated location of the fire source center; Least squares are used to obtain the results from multiple observations. The second stage is fine positioning: using the fusion algorithm of thermal imaging camera and lidar, the point cloud is spatially correlated with the candidate area of ​​heat source, the corresponding point cloud set is extracted and the three-dimensional coordinates of the fire source center are calculated, the initial positioning result is corrected for a second time, and the final position of the fire source center is output. laser point cloud Transform to camera coordinate system: Select the set of points whose projection falls within the candidate heat source region R as the heat source point set, and calculate its spatial estimation center. ,Will and The final fire source center is obtained by weighted fusion. ,satisfy: , and These are the weight matrix and its corresponding inverse matrix, respectively. Through the above identification and localization process, the accuracy of fire source center location can be improved under conditions of smoke obscuring and complex background interference, providing a reliable target for subsequent spraying. S5 (Spray Direction and Fire Extinguishing Control): The controller calculates the azimuth and pitch of the spraying device 8 based on the fire source center coordinates obtained in step S4, and drives the spraying actuator to complete the alignment; it also simultaneously completes the pressurized water supply and spraying control. Based on the coordinates of the fire source center in the vehicle coordinate system Calculate the azimuth and elevation angles of the spraying device: The three-dimensional coordinates of the fire source center in the vehicle coordinate system; Yaw angle; Pitch angle; The center point of the fire source is transformed sequentially from the world coordinate system to the vehicle coordinate system and the spraying device base coordinate system. Based on the relative coordinates of the fire source in the spraying device base coordinate system, the yaw angle and pitch angle are calculated, thereby converting the geometric pitch relationship between the vehicle body and the fire source into the pitch control quantity of the spraying device relative to the fire source. The fixed external parameters between the spraying device base coordinate system and the vehicle coordinate system are obtained from the assembly dimensions or calibration.

[0027] (1) Pressure-boosting water supply control: The steering motor 902 drives the first bevel gear 601 through the key 901, and the first bevel gear 601 drives the second bevel gear 602 for transmission; the electromagnet controls the coupling sleeve to realize power coupling and distribution, so that the output end of the second bevel gear 602 drives the main shaft of the plunger pump 12 to rotate, thereby driving the center bar 1202 and the plunger assembly to work, boosting the water supply to the fire extinguishing agent storage tank to form a stable jet flow; (2) Spraying closed-loop control: During the fire extinguishing process, the controller continuously reads the thermal imaging temperature feedback and performs closed-loop adjustment, preferably including spray intensity adjustment, fine-tuning of direction, and switching of sweeping strategy. When the drift of the fire source center or insufficient temperature drop is detected, the controller fine-tunes the direction of the spraying device or adopts sweeping spray within a certain angle range to improve the fire extinguishing coverage and hit efficiency; when an obstacle is detected to block the spraying path, the controller drives the vehicle to adjust its position before executing spraying. S6 (Firefighting Ends and Return to Base): During and after firefighting, the controller continuously collects thermal imaging data to verify the firefighting effect: when the peak temperature of the fire source area decreases and remains stable below the preset threshold and meets the continuous frame rate requirement, the firefighting is deemed complete and the task ends; if a temperature rise or suspected reignition is detected, the controller returns to steps S4-S5 to perform re-identification, repositioning, and re-spraying to prevent secondary reignition. When the task ends or the battery is low, the controller triggers the return-to-home process, using the origin pose established in step S1 as the return-to-home target, performing path planning based on the constructed map, and executing obstacle avoidance and path tracking control to drive the vehicle safely back to the origin or resupply point and stop; preferably, the speed is reduced and attitude is corrected when approaching the origin to achieve safe stopping or entering the charging / resupply state; S6.1 Fire Extinguishing Effect Verification and Mission Completion Judgment: Set fire extinguishing threshold When the thermal imaging measures the maximum temperature ,satisfy: And continue If the frame is cleared, the fire is considered extinguished; if the above conditions are not met, the fire will be located and controlled again. S6.2 Return to Home and Return to Dock Point: When firefighting is completed, power is insufficient, or a recall command is received, a return-to-home process is triggered. Using the point recorded in step S1.2 as the return-to-home target, the route is replanned, and obstacle avoidance and path tracking are performed. Upon approaching the docking point, position error control is used for precise regression. The corresponding precise positioning regression algorithm is as follows: and For error control rate; wrap means angle normalization: mapping any angle error to a continuous, unique principal value interval, here it is... in Current location of the vehicle; Return docking point; Distance error to the target; The difference between the angle pointing to the target point and the current heading; Ensure alignment with the target heading upon arrival; Error control rate; Through the above steps, the present invention can automatically identify, accurately locate and effectively extinguish fire sources in complex environments, and autonomously return to the origin when the fire is extinguished or the power is insufficient, thereby improving the efficiency of fire source investigation and the safety of fire extinguishing operations, avoiding secondary reignition, and enhancing emergency response capabilities. After the vehicle is powered on, it performs multi-sensor self-testing, time synchronization, and origin pose establishment. Upon receiving navigation information, it employs a synchronous positioning and mapping method tightly coupled with a lidar and inertial measurement unit to achieve autonomous positioning and map construction. This involves point cloud motion distortion correction through partitioning to unify the point cloud to a reference time. In the front-end laser odometry, principal component analysis and mean clustering are used to extract feature points, complete ground extraction, and constraint construction. In the back-end optimization section, constraint factors from each stage are introduced for factor graph optimization, resulting in a stable pose output. Based on this pose and map, a navigation obstacle avoidance algorithm is used to complete global path planning and local dynamic obstacle avoidance control, enabling the vehicle to autonomously drive to a designated fire area in complex forest or obstacle environments. Upon reaching the target area, thermal imaging is used... A maximum inter-class variance (MOL) image segmentation algorithm for visual images is used to automatically identify fire sources and extract candidate regions. A control flow for a fire sprinkler system is designed to achieve closed-loop pointing and spray control of the sprinkler's azimuth and elevation. Furthermore, a two-stage fire source center localization method is designed. First, the fire source center is initially located using a triangulation method. Then, a fusion algorithm combining thermal imaging and LiDAR is used for secondary localization, fusing the coordinates of both to obtain the final fire source center location, thus achieving precise targeting and effective extinguishing of the fire. During the firefighting process, continuous verification and dynamic correction are performed based on thermal imaging temperature feedback to prevent reignition. When the fire is extinguished or the battery is low, the vehicle autonomously returns to its initial position or resupply point based on its original pose and the constructed map. This device enables timely identification of fire source changes and the implementation of precise and effective firefighting measures even in complex environments. This equipment will bring new ideas and methods to forest fire fighting, improve the efficiency of fire source investigation and extinguishing, prevent secondary reignition of fires, and enhance the overall safety and emergency response capabilities of firefighting operations.

[0028] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A small, self-sensing intelligent special vehicle, characterized in that: include: The lower base plate (13) has multiple support feet (7) connected to its outer side. The top end of the lower base plate (13) is connected to a gearbox (6). Each output end of the gearbox (6) is connected to a plunger pump (12), and the number of the support feet (7) corresponds to the number of the gearboxes (7). The top plate (3) is connected to an identification device (1) and a spraying device (8) is provided at the front end of the top of the top plate (3) corresponding to the fire extinguishing agent storage box (10). The middle plate (14) is connected to a fire extinguishing agent storage tank (10) on one side of the top of the middle plate (14), and a steering assembly (9) is connected to the middle of the top surface of the middle plate (14). A control module (4) and a battery module (5) are provided on one side of the top of the middle plate (14). The top plate (3), middle plate (14) and bottom plate (13) are fixedly connected from top to bottom by the support column (11).

2. The small self-sensing intelligent special vehicle according to claim 1, characterized in that: The identification device (1) includes a camera bracket (101), a camera (102), a first servo motor (103), a support rod (104), and a lidar (105). There are two first servo motors (103). One first servo motor (103) is fixedly connected to the support rod (104) and installed on the top of the upper plate (3). The other first servo motor (103) is fixedly installed on one side of the top of the support rod (104) and connected to the camera bracket (101) and the laser radar (105). The camera (102) and the laser radar (105) are fixed on the side of the camera bracket (101).

3. A small, self-sensing intelligent special vehicle according to claim 2, characterized in that: The spraying device (8) includes a nozzle (801), a support gimbal (802), a second servo motor (803), and a servo motor connector (804). The support gimbal (802) is fixedly installed on one side of the top of the upper plate (3), and the support gimbal (802) can extend and retract vertically and rotate 360 ​​degrees. The second servo motor (803) and the servo motor connector (804) are rotatably connected. The servo motor connector (804) is fixedly installed at the corresponding position of the support gimbal (802). The nozzle (801) is fixedly installed on the top of the second servo motor (803). The nozzle (801) draws fire extinguishing agent liquid from the fire extinguishing agent storage tank (10) and sprays it through the water pump (2) fixedly installed on the top of the upper plate (3).

4. A small, self-sensing intelligent special vehicle according to claim 1, characterized in that: The gearbox (6) includes a housing (605), inside which is provided a first bevel gear (601) that can be driven by the steering assembly (9). The outer end of the first bevel gear (601) is connected to four second bevel gears (602). The output end of the second bevel gear (602) is provided with two electromagnets (603). The electromagnets (603) contain coils that can generate magnetism when energized. A coupling sleeve (604) is provided between the two electromagnets (603). Power is transmitted to the plunger pump (12) through the coupling sleeve (604). The second bevel gear (602) and the housing (605) are rotated together.

5. A small, self-sensing intelligent special vehicle according to claim 4, characterized in that: The steering assembly (9) includes a key (901) and a steering motor (902). The key (901) is fixedly mounted on the output shaft of the steering motor (902). The steering motor (902) is connected to the first bevel gear (601) through the key (901).

6. A small, self-sensing intelligent special vehicle according to claim 4, characterized in that: The number of the support legs (7) is four, and they are equidistantly connected to the outer side of the bottom plate (13). The support legs (7) include a connector (701), a piston cylinder (702), a robotic arm (703), a piston rod (704), a Mecanum wheel (705), and a motor (706). The connector (701) is fixedly installed on the outer side of the bottom plate (13). A rotatable piston cylinder (702) is connected to the side end of the connector (701). A mechanical arm (703) is rotatably connected to the bottom of the side end of the connector (701). The mechanical arm (703) is rotatably connected to the piston rod (704) provided at the output end of the piston cylinder (702). A motor (706) is provided on one side of the end of the mechanical arm (703). A Mecanum wheel (705) is connected to the output end of the motor (706).

7. A small, self-sensing intelligent special vehicle according to claim 6, characterized in that: The number of plunger pumps (12) is four. Each plunger pump (12) includes a main shaft (1201), a center rod (1202), a plunger assembly (1203), a housing (1204), and a rear cover (1205). The main shaft (1201) is connected to the output end of the second bevel gear (602) in cooperation with the coupling sleeve (604). The main shaft (1201) is connected to the center rod (1202) and the six plunger assemblies (1203) in cooperation, and is rotatably connected to the housing (1204). The rear cover (1205) is spherically connected to the center rod (1202), and the rear cover (1205) is fixedly connected to the housing (1204).

8. A small self-sensing intelligent special vehicle and its control method, based on the small self-sensing intelligent special vehicle according to any one of claims 1-7, characterized in that: Includes the following steps: S1: Vehicle initialization, the control module (4) sequentially performs self-test and communication tests on the lidar (105), camera (102), first servo motor (103), plunger pump (12), electromagnet (603) and battery module (5); S2: Receive fire zone information from the task input, generate a global path, and acquire real-time pose. S3: Perform global path planning, generate a global path from the current pose to the fire zone on the map, and achieve local obstacle avoidance and tracking while meeting the safety braking distance constraint; S4: When the vehicle arrives at the fire zone, thermal imaging images are collected by the camera (102), and fire source identification and center positioning are performed. Specifically, this includes: initial positioning and fine positioning. The initial positioning specifically involves obtaining the line of sight direction for the same heat source target and using the triangle positioning method to solve the initial position of the fire source center. The fine positioning specifically involves: spatially associating the point cloud obtained by the lidar (105) with the heat source candidate area, extracting the corresponding point cloud set and calculating the three-dimensional coordinates of the fire source center, making a second correction to the initial positioning result, and finally outputting the fire source center position. S5: Drive the spraying device (8) to complete the alignment; S6: Perform a review after spraying is completed. That is, when the temperature peak of the fire source area decreases and remains stable below the preset threshold and meets the continuous frame requirement, the fire extinguishing is determined to be completed and the task is ended. Otherwise, S4-S5 are re-executed. Drive the vehicle back to the starting point when the battery is low or the mission is completed.

Images