CN-120293125-B - Forest fire-extinguishing robot path planning method and system based on ground-air coordination

CN120293125BCN 120293125 BCN120293125 BCN 120293125BCN-120293125-B

Abstract

The invention discloses a forest fire-extinguishing robot path planning method and system based on ground-air coordination, wherein the method utilizes an unmanned aerial vehicle to carry a laser radar and an infrared and visible light camera to scan the disaster area in the air, and acquires the space information and the fire intensity of a fire point in real time; the unmanned aerial vehicle is in communication connection with the wind power fire extinguishing robot, shares a three-dimensional point cloud model, performs environment calibration and positioning, performs path planning by applying an improved A-type algorithm, builds a path planning mathematical model aiming at the minimum disaster degree, dynamically updates and plans the operation path of the wind power fire extinguishing robot through fusion of a global satellite navigation system and an inertial navigation system, identifies a fire source after the target fire point is reached, calculates a safe fire extinguishing distance by using a flame detection model result, and dynamically adjusts the angle of an air duct according to the flame state to perform fire extinguishing operation. According to the invention, through cooperative work between the unmanned aerial vehicle and the fire extinguishing robot, the efficiency and the intelligent degree of fire extinguishing operation are improved.

Inventors

YANG CHUNMEI
ZHAO XIN
LIU LIN
YAN JIE

Assignees

东北林业大学

Dates

Publication Date: 20260512
Application Date: 20250314

Claims (8)

1. A forest fire-extinguishing robot path planning method based on ground-air coordination is characterized by comprising the following steps: Step S1, carrying out aerial range scanning on a forest disaster area by an unmanned aerial vehicle through carrying a laser radar, generating a three-dimensional point cloud model of the unmanned aerial vehicle, and acquiring longitude and latitude of each fire point, the size of the fire point area and the intensity information of the fire point in real time through an infrared camera and a visible light camera; s2, the wind power fire extinguishing robot is in communication connection with the unmanned aerial vehicle, a three-dimensional space model of the surrounding environment is obtained through a laser radar, and the three-dimensional space model is compared with the three-dimensional point cloud model of the unmanned aerial vehicle to position information of the wind power fire extinguishing robot; Step S3, calculating the length and time information of the wind power fire extinguishing robot reaching a fire point, and constructing a path planning mathematical model aiming at the minimum disaster degree according to the information, the fire point area and the fire spreading speed, wherein the method comprises the following specific steps of: step S31 application improvement The path planning algorithm generates an optimal path to each fire, wherein: Improvements in or relating to The cost function of the path planning algorithm is as follows: Wherein, the Representing from a starting node to a current node Is a cost estimate of (2); To be from the initial node to the current node I.e. the distance taken from the starting node to the current node or the actual cost paid; To be from the current node Heuristic estimated cost to the target node, i.e., the predicted distance of the current node to the target node; the weight coefficient of the heuristic function is used for adjusting the influence degree of heuristic estimated cost on the total cost; step S32, constructing a target function of the wind power fire extinguishing robot by taking the minimum disaster degree as a target according to the length and the time of the wind power fire extinguishing robot reaching a fire point route, the fire point area and the fire spreading speed parameters, wherein: The objective function formula is as follows: Wherein, the From the current position for the wind power fire extinguishing robot Reaching the fire point Of (3), wherein , Is a current position set of the wind power fire extinguishing robot, , A target fire point position set; the said The formula of (2) is as follows: Wherein, the For the wind power fire-extinguishing robot slave position To the point of fire Is a path length of (a); for the wind power fire-extinguishing robot slave position Reaching the fire point Time spent; Is the fire point Is the area size of (2); Is the fire point Is a fire spread rate; 、、、 Is that 、、 And Weights of (2); And Punishment term coefficients for path time and fire spreading speed; S4, acquiring speed, position and attitude information of the wind power fire extinguishing robot through fusion of a global satellite navigation system and an inertial navigation system, and dynamically updating and planning a working path of the wind power fire extinguishing robot according to fire point information; and S5, recognizing a fire source through a camera carried by the wind power fire extinguishing robot after reaching a target fire point, calculating a safe fire extinguishing distance by using a flame detection model result, and dynamically adjusting the angle of an air duct according to a flame state to perform fire extinguishing operation, wherein the method comprises the following specific steps of: s51, acquiring a flame area real-time video by the wind power fire extinguishing robot through a carried camera, and extracting flame area and flame space position information by using a YOLO flame detection model; S52, constructing an air duct angle adjustment strategy based on the pixel coordinates and the depth information obtained in the S51, adjusting the angle and the wind power of an extinguishing air duct by a wind power fire extinguishing robot according to different stages of flame, keeping the angle of the air duct close to 0 DEG in parallel with the horizontal direction when the height of the flame is not obviously increased or the flame starts to be reduced, pressing the top of the flame by the air duct, spraying strong wind upwards, swinging the air duct, blowing heat, gradually moving the air duct to the bottom of the flame for extinguishing fire when the height of the flame is continuously reduced or the flame is weakened, gradually increasing the angle of the air duct from 0 DEG, limiting the maximum angle to 45 DEG, finally pointing to the bottom of the flame for cutting, and rapidly extinguishing the fire source; the wind cone angle adjustment strategy is as follows: Wherein, the Air duct for indicating t moment an included angle in the horizontal direction; indicating the change rate of the flame height, i.e., the amount of change in the flame height per unit time; The flame height at time t; is the horizontal distance between the air duct and the flame; And the sensitivity coefficient is used for adjusting the sensitivity of the angle response of the air duct to the flame height change.
2. The forest fire-extinguishing robot path planning method based on the ground-air coordination according to claim 1, wherein the specific steps of the step S1 are as follows: S11, carrying a high-precision laser radar on the unmanned aerial vehicle, and performing aerial scanning on a forest disaster area according to a preset flight path to generate three-dimensional point cloud data of the unmanned aerial vehicle; Step S12, detecting a temperature abnormal region and calibrating fire points by using an infrared camera, and combining a visible light camera to confirm fire point characteristics; and S13, transmitting the fire position, the fire area and the fire intensity information to a control center in real time, and providing input for subsequent fire extinguishing planning.
3. The forest fire-extinguishing robot path planning method based on the ground-air coordination according to claim 1, wherein the specific steps of the step S2 are as follows: s21, starting a laser radar by the wind power fire extinguishing robot, scanning the surrounding environment for multiple times, and obtaining multiple groups of local three-dimensional point cloud data; S22, aligning local three-dimensional point cloud data obtained by different scanning by adopting rigidity transformation to obtain a unified local three-dimensional point cloud set, downsampling the local three-dimensional point cloud set, reducing the density of the point cloud, removing isolated and abnormal noise points, and establishing a coordinate system at the current position of the wind power fire extinguishing robot; Step S23, the wind power fire extinguishing robot pairs each point in the current local three-dimensional point cloud set with the nearest point in the three-dimensional point cloud data of the unmanned aerial vehicle according to a pre-provided unmanned aerial vehicle global point cloud model and by combining with the preliminary positioning of the wind power fire extinguishing robot, calculates rotation and translation transformation according to the matched points, aligns the local point cloud with the global point and positions the wind power fire extinguishing robot according to the characteristics of the point cloud.
4. A forest fire extinguishing robot path planning method based on ground-air coordination as recited in claim 1, characterized in that the heuristic estimated cost The formula of (2) is as follows: Wherein, the method comprises the following steps of , ) For the current node Coordinates of% , ) Is the target node coordinates.
5. The method for planning a forest fire extinguishing robot path based on ground-air coordination according to claim 1, wherein the fire spreading speed is characterized in that The calculation formula of (2) is as follows: Wherein, the Is that Time of day Is defined by the size of the area of (a), Is that Time of day Is a size of the area of (a).
6. The method for planning the forest fire-extinguishing robot path based on the ground-air coordination according to claim 1, wherein the specific steps of the step S4 are as follows: s41, fusing the position, the speed and the gesture by using a global satellite navigation system and an inertial navigation system, and combining the instant positioning and the map construction technology to finish high-precision navigation; and S42, recalculating and adjusting the operation path according to the fire point data updated by the unmanned aerial vehicle in real time and the path deviation state of the wind power fire extinguishing robot.
7. The method for planning the path of the forest fire extinguishing robot based on the ground-air coordination according to claim 1, wherein the flame space position is [ ] , , ) The calculation formula of (2) is as follows: Wherein, the 、 Is the transverse and vertical offset of the flame in the camera coordinate system; the depth distance between the flame and the camera; And The focal length of the camera is set; And Pixel coordinates of flame in the image; And Is the center coordinates of the image; Is the depth value, i.e. the actual distance of the flame.
8. A forest wind fire extinguishing robot path planning system based on ground-air synergy for implementing the method of any one of claims 1-7, characterized in that the system comprises a ground-air communication module, an identification positioning module, a navigation planning module and a detection fire extinguishing module, wherein: The ground-air communication module is used for realizing wireless communication among the ground control station, the unmanned aerial vehicle and the wind power fire extinguishing robot, and guaranteeing real-time transmission of data and receiving of instructions; The identifying and positioning module is used for identifying the surrounding environment in real time through a sensor carried by the wind power fire extinguishing robot and determining the position of the wind power fire extinguishing robot; The navigation planning module is used for combining the position of the target fire point and the fire spreading speed parameters according to the position of the wind power fire extinguishing self and the surrounding environment information provided by the identification positioning module and improving Calculating and optimizing an optimal path of the fire extinguishing robot reaching a target fire point by a path planning algorithm, and adjusting the path in real time according to the dynamic change of the fire point; The detecting and extinguishing module is used for identifying the position, the shape and the flame height state of the flame in real time through a camera and a flame detecting model carried by the wind power fire extinguishing, dynamically adjusting the angle of a wind barrel of the wind power fire extinguishing robot and wind power according to the change of the flame, and executing fire extinguishing operation.

Description

Forest fire-extinguishing robot path planning method and system based on ground-air coordination Technical Field The invention belongs to the technical field of forest fire-fighting robots, relates to a forest fire-fighting robot path planning method, and in particular relates to a forest fire-fighting robot path planning method and system based on ground-air coordination. Background Forest fires are one of the major disasters that seriously threaten the ecological environment and the safety of human lives and properties. Traditional fire extinguishing methods mainly rely on manual extinguishing or ground mechanical equipment, but in the fire environment of large-scale and complex terrains, the methods have the defects of low response speed, low fire extinguishing efficiency, limited coverage range and the like. With the rapid development of unmanned aerial vehicle technology and robot technology, an intelligent fire extinguishing system with ground and air cooperation gradually becomes a research hot spot. In the prior art, many studies and technical solutions have attempted to use unmanned aerial vehicles in cooperation with robots for fire suppression. For example, some unmanned aerial vehicles are equipped with sensors such as lidar, infrared cameras and the like for detecting information such as fire position, fire spread, forest topography and the like, and some fire extinguishing robots acquire environmental data and plan paths using the sensors to perform fire extinguishing tasks. However, these techniques still have some limitations, especially in terms of path planning, fire point identification, prediction of fire spread, and dynamic adjustment of fire extinguishing strategies, and no truly intelligent synergy has been achieved. Although the prior art provides a new idea for fire suppression, a plurality of technical bottlenecks exist, for example, the existing path planning method often ignores the dynamic change of fire points and the fire spreading speed, so that the planned path is not optimal, and the change caused by fire spreading cannot be dealt with in real time. In the prior art, various sensors and algorithms are adopted for fire point identification and fire extinguishing operation, but static judgment is mostly relied on, dynamic change of flame and flexibility of fire extinguishing strategies are not fully considered, and improvement of fire extinguishing efficiency is limited. Most robots and unmanned aerial vehicles in the prior art have imperfect cooperative work, and lack of effective information sharing and cooperative mechanisms, so that fire extinguishing tasks cannot be completed efficiently in a complex environment. Therefore, a new technical scheme is urgently needed, an optimal path can be planned in real time in a dynamically-changed fire environment, a fire source is accurately positioned, a fire extinguishing strategy is intelligently adjusted, and the fire extinguishing efficiency and the success rate are improved. Disclosure of Invention The invention aims to provide a forest fire-extinguishing robot path planning method and system based on ground-air coordination, which improve the efficiency and the intelligent degree of fire-extinguishing operation through the cooperative work between an unmanned aerial vehicle and a fire-extinguishing robot. The invention aims at realizing the following technical scheme: A forest fire-extinguishing robot path planning method based on ground-air coordination comprises the following steps: Step S1, carrying out aerial range scanning on a forest disaster area by an unmanned aerial vehicle through carrying a laser radar, generating a three-dimensional point cloud model of the unmanned aerial vehicle, and acquiring longitude and latitude of each fire point, the size of the fire point area and the intensity information of the fire point in real time through an infrared camera and a visible light camera; s2, the wind power fire extinguishing robot is in communication connection with the unmanned aerial vehicle, a three-dimensional space model of the surrounding environment is obtained through a laser radar, and the three-dimensional space model is compared with the three-dimensional point cloud model of the unmanned aerial vehicle to position information of the wind power fire extinguishing robot; s3, calculating the length and time information of a wind power fire extinguishing robot reaching a fire point, and constructing a path planning mathematical model aiming at the minimum disaster degree according to the information, the fire point area and the fire spreading speed; S4, acquiring speed, position and attitude information of the wind power fire extinguishing robot through fusion of a global satellite navigation system and an inertial navigation system, and dynamically updating and planning a working path of the wind power fire extinguishing robot according to fire point information; And S5, recognizing a fire source through a came