High-rise building fire rescue system and rescue system control method

Abstract

The invention discloses a high-rise building fire rescue system and a rescue system control method, the system comprises an unmanned aerial vehicle, the unmanned aerial vehicle bears a glass breaking bomb, a static rope and a rope climbing robot, and the glass breaking bomb is installed on an ejection device of the unmanned aerial vehicle; during rescue, one end of the static rope is fixed outside a building by the unmanned aerial vehicle through a quick locking device, the other end of the static rope is thrown into a fire room, and the rope body is used for providing a moving track for the rope climbing robot; the rope climbing robot comprises a battery system, a digital motor, a planetary reducer, a safety buckle, an emergency braking device and a control module. According to the method, the flexibility of the unmanned aerial vehicle and the rope climbing robot are utilized, accurate positioning and flexible movement are achieved, the unmanned aerial vehicle and the rope climbing robot system do not need to depend on physical guide rails and flexibly cope with various building structures and complex environments, the static rope system can be rapidly and safely fixed and used for the rope climbing robot, normal operation under various conditions is ensured, and the safety is high. The adaptability and reliability of rescue are improved, and the efficiency and safety of high-rise building fire rescue are improved.

Description

Technical Field

[0001] The invention relates to a high-rise building fire rescue system and a rescue system control method, belonging to the technical field of emergency rescue equipment and fire fighting. Background Art

[0002] In the process of accelerating urbanization and urbanization, the original building form can no longer meet the living and working environment requirements of the urban population, and the building floors are gradually increasing. There are a large number of vertical shafts such as pipeline wells, cable wells, and smoke exhaust ducts in high-rise buildings. Once a fire occurs, these shafts increase the vertical spread of fire in high-rise buildings and accelerate the speed of smoke spread. The difficulty of high-rise building fire rescue is mainly reflected in the following aspects: (1) Limited rescue height: The existing ladder trucks and fire trucks are generally only tens of meters high, which makes it difficult to cover the high-rise fire rescue needs of super-high-rise buildings; (2) Slow rescue speed: Traditional rescue methods require a long time to deploy and operate. Ladder trucks and high-pressure water guns need to be deployed after firefighters arrive at the scene. This process may take several minutes or even longer, which is particularly unfavorable in emergency situations when a fire occurs; (3) Greatly affected by the environment: During the high-rise building fire rescue process, environmental factors such as wind speed and direction have a great impact on traditional rescue equipment, affecting the stability and operation of ladder trucks and high-pressure water guns; (4) Limited rescue coverage: The internal structure of high-rise buildings is complex, and traditional rescue methods are difficult to quickly launch rescue operations inside the building and cannot effectively cover all affected areas. Ladder trucks and high-pressure water guns can only carry out rescue from the outside and it is difficult to enter the building for precise rescue.

[0003] At present, with the continuous application of drone technology in fire rescue, fire rescue in high-rise buildings has made certain progress, but there are still problems, mainly reflected in (1) complex structure and difficult maintenance: the guide rail system and bearing device in the existing technology have complex structures, including multiple telescopic parts and locking components. These complex mechanical structures are prone to failure under high temperature and emergency conditions, which increases the difficulty and cost of maintenance; (2) limited positioning accuracy: due to the limitations of the guide rail system, the movement range and flexibility of the bearing device are affected. In the actual rescue process, the guide rail system may not be able to accurately locate the fire source, resulting in a decrease in rescue efficiency; (3) limited applicable environment: the existing technology relies on the physical connection of the guide rail system. For rooms with closed windows or obstacles, the guide rail is difficult to enter, affecting the rescue effect. In addition, the complex guide rail system is difficult to install and use in certain building structures. Summary of the invention

[0004] The present invention provides a high-rise building fire rescue system and a rescue system control method. The rescue system has a simple structure, is easy to maintain, has a low failure rate, is flexible in positioning, has high precision, and can adapt to a variety of environments. The rescue system control method integrates a variety of data, has high control precision, and can improve rescue efficiency.

[0005] In order to achieve the above-mentioned objectives, the present invention provides a high-rise building fire rescue system, including a drone, the drone carries a glass-breaking bomb, a static rope and a rope-climbing robot, the glass-breaking bomb is installed on the ejection device of the drone, and includes a high-strength spring and an impact head; during rescue, one end of the static rope is fixed to the outside of the building by the drone through a quick locking device, and the other end is dropped into the fire room, and the rope body is used to provide a mobile track for the rope-climbing robot; the rope-climbing robot includes a battery system, a digital motor, a planetary reducer, a safety buckle, an emergency braking device and a control module.

[0006] A high-rise building fire rescue system control method comprises the following steps:

[0007] Step 1: The drone carries the glass-breaking bomb, static rope and rope-climbing robot and flies to the outside of the fire room through the navigation of the positioning module;

[0008] Step 2: The glass-breaking bomb is launched through the ejection device of the drone. The high-strength spring releases energy, pushing the impact head to impact and break the glass on the windows of the fire room;

[0009] Step 3: The drone fixes one end of the static rope outside the fire room and drops the other end into the fire room;

[0010] Step 4: Place the rope-climbing robot on the static rope. Under the control of the control module, the rope-climbing robot moves along the static rope to the fire room through remote control or automatic mode, and then transports personnel or materials to the ground.

[0011] Furthermore, the control process of the rope climbing robot control module in step 4 is:

[0012] (1) Establish a motion model of the rope-climbing robot in the non-powered release state, simulate the descent speed and displacement trajectory of the machine, and realize the estimation and optimization of the motion process of the rope-climbing robot:

[0013] (1-1) Force analysis:

[0014] Gravity analysis: F g =mg; where m is the mass of the rope-climbing robot and g is the acceleration due to gravity;

[0015] Static rope friction analysis: F friction=μmgcos(θ); where μ is the friction coefficient; θ is the inclination angle of the static rope relative to the horizontal plane; cos(θ) is the horizontal component of the static rope, which is used to calculate the friction resistance along the static rope;

[0016] Wind resistance analysis: Where ρ is the air density; C d A is the air resistance coefficient; area is the windward area of ​​the rope-climbing robot; v m The machine descending speed;

[0017] (1-2) Equation of motion:

[0018] The resultant force along the static rope is:

[0019] F T =F g sin(θ)-F friction -F fall cos(θ);

[0020] Substituting into the force analysis formula, we get:

[0021]

[0022] According to Newton's second law, the acceleration a along the static rope is:

[0023]

[0024] Simplifying, we get:

[0025]

[0026] (1-3) Velocity and displacement equations:

[0027] Assuming the initial velocity is v0, the instantaneous velocity v(t) of the rope-climbing robot in the no-power release state is obtained by integrating the acceleration equation:

[0028]

[0029] The displacement x(t) is obtained by further integrating the velocity equation:

[0030]

[0031] (2) Introducing intelligent sensors and algorithms, a real-time rope tension feedback control model is established to enable the rope climbing robot to automatically detect the rope tension, ambient wind speed and angle parameters, and automatically adjust the rotor speed to change the speed:

[0032] (2-1) Analysis of static rope tension and force balance of rope climbing robot:

[0033] Assume that the rope-climbing robot moves along the rope, and the tension of the rope is T rope The weight of the rope-climbing robot, the swing angle δ and the wind speed v wind The influence of rope tension is expressed as:

[0034]

[0035] Where, δ is the swing angle; L is the rope length;

[0036] (2-2) Analysis of lateral force under the influence of wind speed:

[0037] The wind speed will generate a lateral force F on the rope climbing robot wind , which affects the balance of the machine. The lateral wind force is approximately expressed as:

[0038]

[0039] (2-3) The control formula for automatically adjusting the wheel speed is:

[0040] In order to realize automatic adjustment control based on sensor feedback, it is assumed that the goal of the rope climbing robot is to maintain a stable moving speed v target , the proportional-integral-differential PID controller is used to adjust the wheel speed according to the deviation. The PID control formula is:

[0041]

[0042] Where u(t) is the control output, i.e. the adjustment amount of the wheel; e(t) = v target -v(t) is the speed deviation, that is, the difference between the target speed and the current speed; K p , K i and K d are proportional, integral and differential gain coefficients respectively;

[0043] (2-4) Combined tension and angle control feedback:

[0044] Feedback control of tension and swing angle is introduced to adjust the rotation speed of the wheel to cope with the swing of the static rope. Assuming that the sensor measures the current tension T rope and angle δ, introduce angle compensation term and tension compensation term to further improve the control formula:

[0045]

[0046] Among them, K δ K is the angle compensation gain, used to cope with angle changes; T is the tension compensation gain; T target It is the preset tension target value, which is used to maintain the balance of the rope climbing robot;

[0047] (2-5) Dynamic adjustment of ambient wind speed:

[0048] Changes in wind speed will affect the stability of the machine. The control system can be further optimized through wind speed feedback. Assuming that the sensor monitors the wind speed v in real time wind , introduce the wind speed compensation term:

[0049]

[0050] Among them, K v is the wind speed compensation gain;

[0051] (2-6) Wheel angular velocity adjustment:

[0052] According to the controller output u(t), the angular velocity ω of the wheel can be calculated:

[0053]

[0054] Where r is the wheel radius;

[0055] (3) In view of the influence of wind speed in the environment, internal shock absorption and windproof stability design are added:

[0056] (3-1) In order to reduce the lateral swing of the machine caused by wind, a damping device is designed inside the rope climbing robot. The damping force F damping Proportional to speed:

[0057] F damping = -b·v;

[0058] Where b is the damping coefficient; v is the instantaneous speed of the machine;

[0059] (3-2) When a rope climbing robot is disturbed by wind, it will generate an elastic restoring force to try to restore the machine to the equilibrium position. The elastic restoring force is expressed by Hooke's law:

[0060] F spring = -k·x;

[0061] Where k is the spring coefficient; x is the displacement, i.e., the distance the rope-climbing robot deviates from the equilibrium position;

[0062] (3-3) Under the combined action of wind force and shock absorption / elastic device, the lateral motion of the rope climbing robot is expressed by the following differential equation:

[0063]

[0064] After expansion:

[0065]

[0066] in, is the lateral acceleration of the rope-climbing robot;

[0067] In summary, by setting reasonable damping coefficients and spring coefficients, the stability of the system in an environment with changing wind speeds can be ensured. Generally, increasing the values ​​of the damping coefficient and spring coefficient can improve the damping effect and restoring force of the system, allowing the machine to quickly return to a balanced position and reduce the vibration amplitude;

[0068] In order to maintain the stability of the machine when the wind speed changes, you can: ① Increase the damping coefficient b to increase the vibration attenuation speed. ② Increase the spring coefficient k to enhance the restoring force. ③ Appropriately adjust the air resistance coefficient Cd and the windward area A. area To reduce wind disturbance to the machine.

[0069] (4) Since the wind speed will affect the stability of the rope climbing robot when it is working at high altitude, a nonlinear controller that is resistant to wind disturbance can be introduced to automatically adjust the grip or release force of the machine based on real-time monitoring of the ambient wind speed to prevent the rope climbing robot from deviating from the control path due to the influence of wind:

[0070] (4-1) To offset the influence of wind on the rope climbing robot, an anti-wind control force F is applied. control , so that the rope climbing robot remains at the target position, using the following control formula:

[0071]

[0072] k p and k d is the gain parameter of position and speed control; x target is the target position, x is the displacement, f nonlinear (x) is a nonlinear compensation function used to cope with disturbances under high wind speeds;

[0073] (4-2) The gripping or releasing force T of the rope climbing robot is related to the wind force, and the following formula is used to calculate the gripping force:

[0074] T=μ·(m·g+F control -F wind );

[0075] (4-3) To achieve more stable wind resistance control, the hyperbolic tangent function tanh is used to deal with rapidly changing wind speeds:

[0076] f nonlinear (x) = β1·tanh(β2·(vv threshold ));

[0077] Among them, β1 and β2 are adjustment parameters; v threshold is the threshold value. When the wind speed exceeds this value, nonlinear compensation begins to become significant.

[0078] (4-4) Update the displacement and velocity of the machine by numerical integration:

[0079] x(t+Δt)=x(t)+v(t·Δt

[0080]

[0081] Through the above formula, the impact of wind on the machine can be dynamically monitored and calculated, so as to adjust the grip or release force of the rope climbing robot in real time to ensure the stability of the machine in an environment with changing wind speed.

[0082] (5) The speed control loop and position control loop in the hierarchical control strategy are designed to achieve smooth and precise movement of the rope-climbing robot on the rope. The position control loop is responsible for ensuring that the machine moves along the predetermined overall motion trajectory, while the speed control loop is used for fine adjustments to enable the rope-climbing robot to land smoothly at the target position. The specific process is as follows:

[0083] (5-1) Position control loop: Assume the target position is x target , the current displacement of the rope climbing robot is x, then the position deviation is e p for:

[0084] e p =x target -x

[0085]

[0086] (5-2) Speed ​​control loop: target speed v target The output from the position control loop, the actual speed v is the current real-time speed of the rope climbing robot, and the speed deviation e v for:

[0087] e v =v target -v;

[0088] The speed is regulated using a PID controller to generate a force F P To overcome resistance and achieve speed adjustment:

[0089]

[0090] k pv , k iv and k dv is the PID gain of the speed control loop;

[0091] (5-3) System dynamics model: Based on Newton's second law, the motion state of the rope-climbing robot is described, and its acceleration a is expressed as:

[0092]

[0093] The velocity v and displacement x can be updated by numerical integration method:

[0094] v(t+Δt)=v(t)+a·Δt

[0095] x(t+Δt)=x(t)+v(t)·Δt;

[0096] Among them, Δt is the time step;

[0097] (5-4) Combination of hierarchical controllers: The core of hierarchical controllers is to achieve an effective combination of position control loops and speed control loops. The specific implementation steps are as follows:

[0098] (5-4-1) The position control loop is based on the position deviation e p Generate target speed v target ;

[0099] (5-4-2) The speed control loop is based on the speed deviation e v Generate control force F P ;

[0100] (5-4-3) Using the kinetic model, according to F P Update the speed and position of the rope climbing robot;

[0101] (5-4-4) Continuously adjust the gains of the speed and position controllers to achieve more precise and smooth control;

[0102] This hierarchical control strategy combines the overall trajectory tracking function of the position control loop and the fine adjustment function of the speed control loop to achieve smooth and precise movement of the rope-climbing robot on the rope.

[0103] (6) Design an adaptive control system so that the rope-climbing robot can automatically adjust the clamping force according to the thickness of the rope and the surface friction coefficient. The specific process is as follows:

[0104] (6-1) Establishing friction model: The core of the gripping mechanism is to prevent the rope climbing robot from sliding by friction. The friction force F friction The calculation is as follows:

[0105] F friction =μ·N;

[0106] Where N is the normal force exerted by the gripping mechanism on the rope;

[0107] In this design, the variation of μ and N enables the gripping mechanism to automatically adapt to different friction requirements;

[0108] (6-2) Determination of target friction: Assume that the system requires the minimum friction F required To ensure stability, we have:

[0109] F required ≥ mg;

[0110] To prevent the rope-climbing robot from sliding on different ropes, the friction force F friction Must be greater than or equal to F required ,Right now:

[0111] μ·N≥mg;

[0112] Therefore, the normal force N must satisfy:

[0113]

[0114] (6-3) Adaptive control of clamping force: By measuring the diameter D and friction coefficient μ of the rope, the adaptive system dynamically adjusts the clamping force N to meet

[0115] (6-4) Machine learning-based control strategy: Use one of the machine learning algorithm regression or neural network to predict the required clamping force N. The specific process is as follows:

[0116] First, data collection is performed to collect data sets of different rope diameters D, friction coefficients μ, and required clamping forces N. Second, a model is trained using a regression algorithm to train a model f(D, μ) to predict the required clamping force: N = f(D, μ). Finally, in actual applications, the rope's D and μ are obtained in real time through sensors and input into the model to predict the required N.

[0117] (6-5) Control formula of adaptive materials: When using adaptive materials to adjust the clamping force, it is necessary to use the physical properties of the material to adjust N:

[0118] F m =k m ΔL;

[0119] Among them, k m is the equivalent elastic coefficient of the material; ΔL is the deformation of the material, which can be adjusted by adjusting F m , controls the normal force N applied to the rope;

[0120] (6-6) Mathematical model of control strategy: The adaptive control system is represented by the following control strategy:

[0121] The input layer receives the thickness D and friction coefficient μ of the rope and calculates the clamping force N:

[0122]

[0123] Where f(D,μ) is a machine learning prediction model; real-time adjustment of the clamping force, when the rope diameter or friction coefficient changes, recalculate N, adjust the adaptive material or control the clamping mechanism to achieve the required N;

[0124] (6-7) Feedback loop control: Introduce a feedback loop to monitor the friction force F in real time through sensors actual , and the friction force F required Compare them and if the deviation is large, adjust N:

[0125] ΔN=k f ·(F required -F actual );

[0126] k f is the feedback control gain of the adaptive clamping system;

[0127] The system can automatically adjust the clamping force according to the thickness and friction coefficient of the rope, ensuring sufficient friction in different environments and enhancing the versatility and adaptability of the machine;

[0128] (7) Use laser ranging, visual sensing, gyroscope and other sensors to perform data fusion, plan the best climbing path in real time, avoid obstacles or automatically optimize the climbing speed and direction. The specific process is as follows:

[0129] (7-1) Obstacle detection: Use laser ranging and visual sensors to detect the position of obstacles in the surrounding environment in real time. The distance measured by the laser ranging sensor is expressed as:

[0130]

[0131] Where d is the distance of the obstacle; c is the speed of light; t la is the round trip time of the laser pulse;

[0132] Assume that the current position of the rope climbing robot is (x0, y0, z0), then the three-dimensional coordinates of the obstacle (x obs ,y obs ,z obs ) is expressed as:

[0133] x obs =x0+d·cos(η)·sin(φ)

[0134] y obs =y0+d·sin(η)·sin(φ)

[0135] z obs =z0+d·cos(φ)

[0136] Among them, η and φ are the horizontal and vertical angles, respectively, which are used to determine the relative orientation of obstacles;

[0137] (7-2) Data fusion and state estimation: The position and speed of the rope-climbing robot are estimated through the fusion of visual sensors, laser ranging and gyroscope data to generate a real-time environment map. The specific process is as follows:

[0138] Kalman filtering is a commonly used data fusion method. Assume that the machine state is: The state update formula is:

[0139] X k+1 =A·X k +B·u k +w k

[0140] Among them, A is the state transfer matrix, which describes the dynamics of the system; B is the control matrix; u k is the control input; w k is the process noise;

[0141] The observation equation is:

[0142] Z k =H·X k +v k ;

[0143] Among them, Z k is the observed value, i.e. the position or velocity measured by the sensor; H is the observation matrix; v k is the observation noise;

[0144] The prediction and update steps of the Kalman filter are described as follows:

[0145]

[0146] Update steps:

[0147] K k =P k|k-1 ·H T ·(H·P k|k-1 ·H T +R) -1

[0148]

[0149] P k =(IK k ·H)·P k|k-1

[0150] Among them, K k is the Kalman gain; Q is the process noise covariance matrix; R is the observation noise covariance matrix;

[0151] (7-3) Path planning, the specific process is as follows:

[0152] (7-3-1) The A* algorithm selects a path through the heuristic estimation function f(n), f(n) = g(n) + h(n): g(n) is the known path cost from the starting node to node n; h(n) is the estimated cost from node n to the target node. The specific formula is:

[0153] f(n)=g(n)+h(n);

[0154] At each step, the node with the lowest f(n) value is selected for expansion until the target node is found;

[0155] (7-3-2) Smooth the generated path to avoid large direction changes during the climbing process of the rope climbing robot. The path smoothing formula is:

[0156] path smooth (t) = λ·path raw (t)+(1-λ)·path smooth (t-1);

[0157] Among them, λ is the smoothing coefficient;

[0158] (7-4) Motion control: After the path planning is completed, the control system moves the machine along the path by determining the climbing speed v and the direction angle σ. The speed control formula is:

[0159] Assume the coordinates of the target point are (x goal ,y goal ,z goal ), the current position is (x, y, z), and the unit vector of the target direction is:

[0160]

[0161] The velocity vector v is then:

[0162]

[0163] Among them, v max is the maximum speed of the system;

[0164] (7-5) Direction control: The control formulas for the direction angle σ and the pitch angle ψ are:

[0165]

[0166] The above formulas form the basis for intelligent path planning and real-time obstacle avoidance, and provide computational support for the precise climbing and navigation of the rope-climbing robot.

[0167] The present invention uses a drone to carry a glass-breaking bomb, a static rope and a rope-climbing robot to fly to the outside of the fire room through the navigation of a positioning module. The glass-breaking bomb is launched through the ejection device of the drone, and the high-strength spring releases energy to push the impact head to impact, so that the glass on the window of the fire room is broken. The drone fixes one end of the static rope outside the fire room and drops the other end into the fire room. The rope-climbing robot is placed on the static rope. The rope-climbing robot moves to the fire room along the static rope under the control of the control module through remote control or automatic mode, and then transports personnel or materials to the ground. The drone carries a glass-breaking bomb to break the window glass and drops the static rope. The rope-climbing robot enters the target room along the static rope for rescue, which improves the positioning accuracy and rescue efficiency. The flexibility of the drone and the high-performance rope-climbing robot are used to achieve accurate positioning and flexible movement. The drone and rope-climbing robot system do not need to rely on physical guide rails and can flexibly respond to various building structures and complex environments. The static rope system can be quickly and safely fixed and used for the rope-climbing robot to ensure normal operation under various conditions, improve the adaptability and reliability of rescue, and significantly improve the efficiency and safety of high-rise building fire rescue. BRIEF DESCRIPTION OF THE DRAWINGS

[0168] Figure 1 It is a schematic diagram of the structure of the rescue system of the present invention;

[0169] Figure 2 is a schematic diagram of the rescue process of the rescue system of the present invention;

[0170] Figure 3 It is a control process flow chart of the rope climbing robot control module of the present invention.

[0171] In the picture: 1. Drone, 2. Glass-breaking bomb, 3. Static rope, 4. Rope-climbing robot, 5. Building, 6. Fire room. DETAILED DESCRIPTION

[0172] The present invention will be further described below in conjunction with the accompanying drawings.

[0173] like Figure 1 , Figure 2As shown, a high-rise building fire rescue system includes a drone 1, the drone 1 carries a glass-breaking bomb 2, a static rope 3 and a rope-climbing robot 4, the glass-breaking bomb 2 is installed on the ejection device of the drone 1, and includes a high-strength spring and an impact head; during rescue, one end of the static rope 3 is fixed to the outside of the building 5 by the drone 1 through a quick locking device, and the other end is dropped into the fire room 6, and the rope body is used to provide a moving track for the rope-climbing robot 4; the rope-climbing robot 4 includes a battery system, a digital motor, a planetary reducer, a safety buckle, an emergency braking device and a control module.

[0174] A high-rise building fire rescue system control method comprises the following steps:

[0175] Step 1: The drone carries the glass-breaking bomb, static rope and rope-climbing robot and flies to the outside of the fire room through the navigation of the positioning module;

[0176] Step 2: The glass-breaking bomb is launched through the ejection device of the drone. The high-strength spring releases energy, pushing the impact head to impact and break the glass on the windows of the fire room;

[0177] Step 3: The drone fixes one end of the static rope outside the fire room and drops the other end into the fire room;

[0178] Step 4: Place the rope-climbing robot on the static rope. Under the control of the control module, the rope-climbing robot moves along the static rope to the fire room through remote control or automatic mode, and then transports personnel or materials to the ground.

[0179] like Figure 3 As shown, the control process of the rope climbing robot control module in step 4 is:

[0180] (1) Establish a motion model of the rope-climbing robot in the non-powered release state, simulate the descent speed and displacement trajectory of the machine, and realize the estimation and optimization of the motion process of the rope-climbing robot:

[0181] (1-1) Force analysis:

[0182] Gravity analysis: F g =mg; where m is the mass of the rope-climbing robot and g is the acceleration due to gravity;

[0183] Static rope friction analysis: F friction =μmgcos(θ); where μ is the friction coefficient; θ is the inclination angle of the static rope relative to the horizontal plane; cos(θ) is the horizontal component of the static rope, which is used to calculate the friction resistance along the static rope;

[0184] Wind resistance analysis: Where ρ is the air density; C d A is the air resistance coefficient;area is the windward area of ​​the rope-climbing robot; v m The machine descending speed;

[0185] (1-2) Equation of motion:

[0186] The resultant force along the static rope is:

[0187] F T =F g sin(θ)-F friction -F fall cos(θ);

[0188] Substituting into the force analysis formula, we get:

[0189]

[0190] According to Newton's second law, the acceleration a along the static rope is:

[0191]

[0192] Simplifying, we get:

[0193]

[0194] (1-3) Velocity and displacement equations:

[0195] Assuming the initial velocity is v0, the instantaneous velocity v(t) of the rope-climbing robot in the no-power release state is obtained by integrating the acceleration equation:

[0196]

[0197] The displacement x(t) is obtained by further integrating the velocity equation:

[0198]

[0199] (2) Introducing intelligent sensors and algorithms, a real-time rope tension feedback control model is established to enable the rope climbing robot to automatically detect the rope tension, ambient wind speed and angle parameters, and automatically adjust the rotor speed to change the speed:

[0200] (2-1) Analysis of static rope tension and force balance of rope climbing robot:

[0201] Assume that the rope-climbing robot moves along the rope, and the tension of the rope is T rope Affected by the weight of the rope climbing robot, the swing angle δ and the wind speed v wind , the rope tension is expressed as:

[0202]

[0203] Where, δ is the swing angle; L is the rope length;

[0204] (2-2) Analysis of lateral force under the influence of wind speed:

[0205] The wind speed will generate a lateral force F on the rope climbing robot wind , which affects the balance of the machine. The lateral wind force is approximately expressed as:

[0206]

[0207] (2-3) The control formula for automatically adjusting the wheel speed is:

[0208] In order to realize automatic adjustment control based on sensor feedback, it is assumed that the goal of the rope climbing robot is to maintain a stable moving speed v target , the proportional-integral-differential PID controller is used to adjust the wheel speed according to the deviation. The PID control formula is:

[0209]

[0210] Where u(t) is the control output, i.e. the adjustment amount of the wheel; e(t) = v target -v(t) is the speed deviation, that is, the difference between the target speed and the current speed; K p , K i and K d are proportional, integral and differential gain coefficients respectively;

[0211] (2-4) Combined tension and angle control feedback:

[0212] Feedback control of tension and swing angle is introduced to adjust the rotation speed of the wheel to cope with the swing of the static rope. Assuming that the sensor measures the current tension T rope and angle δ, introduce angle compensation term and tension compensation term to further improve the control formula:

[0213]

[0214] Among them, K δ K is the angle compensation gain, used to cope with angle changes; T is the tension compensation gain; T target It is the preset tension target value, which is used to maintain the balance of the rope climbing robot;

[0215] (2-5) Dynamic adjustment of ambient wind speed:

[0216] Changes in wind speed will affect the stability of the machine. The control system can be further optimized through wind speed feedback. Assuming that the sensor monitors the wind speed v in real time wind , introduce the wind speed compensation term:

[0217]

[0218] Among them, K v is the wind speed compensation gain;

[0219] (2-6) Wheel angular velocity adjustment:

[0220] According to the controller output u(t), the angular velocity ω of the wheel can be calculated:

[0221]

[0222] Where r is the wheel radius;

[0223] (3) In view of the influence of wind speed in the environment, internal shock absorption and windproof stability design are added:

[0224] (3-1) In order to reduce the lateral swing of the machine caused by wind, a damping device is designed inside the rope climbing robot. The damping force F damping Proportional to speed:

[0225] F damping = -b·v;

[0226] Where b is the damping coefficient; v is the instantaneous speed of the machine;

[0227] (3-2) When a rope climbing robot is disturbed by wind, it will generate an elastic restoring force to try to restore the machine to the equilibrium position. The elastic restoring force is expressed by Hooke's law:

[0228] F spring = -k·x;

[0229] Where k is the spring coefficient; x is the displacement, i.e., the distance the rope-climbing robot deviates from the equilibrium position;

[0230] (3-3) Under the combined action of wind force and shock absorption / elastic device, the lateral motion of the rope climbing robot is expressed by the following differential equation:

[0231]

[0232] After expansion:

[0233]

[0234] in, is the lateral acceleration of the rope-climbing robot;

[0235] In summary, by setting reasonable damping coefficients and spring coefficients, the stability of the system in an environment with changing wind speeds can be ensured. Generally, increasing the values ​​of the damping coefficient and spring coefficient can improve the damping effect and restoring force of the system, allowing the machine to quickly return to a balanced position and reduce the vibration amplitude;

[0236] In order to maintain the stability of the machine when the wind speed changes, you can: ① Increase the damping coefficient b to increase the vibration attenuation speed. ② Increase the spring coefficient k to enhance the restoring force. ③ Appropriately adjust the air resistance coefficient Cd and the windward area A. area To reduce wind disturbance to the machine.

[0237] (4) Since the wind speed will affect the stability of the rope climbing robot when it is working at high altitude, a nonlinear controller that is resistant to wind disturbance can be introduced to automatically adjust the grip or release force of the machine based on real-time monitoring of the ambient wind speed to prevent the rope climbing robot from deviating from the control path due to the influence of wind:

[0238] (4-1) To offset the influence of wind on the rope climbing robot, an anti-wind control force F is applied. control , so that the rope climbing robot remains at the target position, using the following control formula:

[0239]

[0240] k p and k d is the gain parameter of position and speed control; x target is the target position, x is the displacement; f nonlinear (x) is a nonlinear compensation function used to cope with disturbances under high wind speeds;

[0241] (4-2) The gripping or releasing force T of the rope climbing robot is related to the wind force, and the following formula is used to calculate the gripping force:

[0242] T=μ·(m·g+F control -F wind );

[0243] (4-3) To achieve more stable wind resistance control, the hyperbolic tangent function tanh is used to deal with rapidly changing wind speeds:

[0244] f nonlinear (x) = β1·tanh(β2·(vv threshold ));

[0245] Among them, β1 and β2 are adjustment parameters; v threshold is the threshold value. When the wind speed exceeds this value, nonlinear compensation begins to become significant.

[0246] (4-4) Update the displacement and velocity of the machine by numerical integration:

[0247] x(t+Δt)=x(t)+v(t·Δt

[0248]

[0249] Through the above formula, the impact of wind on the machine can be dynamically monitored and calculated, so as to adjust the grip or release force of the rope climbing robot in real time to ensure the stability of the machine in an environment with changing wind speed.

[0250] (5) The speed control loop and position control loop in the hierarchical control strategy are designed to achieve smooth and precise movement of the rope-climbing robot on the rope. The position control loop is responsible for ensuring that the machine moves along the predetermined overall motion trajectory, while the speed control loop is used for fine adjustments to enable the rope-climbing robot to land smoothly at the target position. The specific process is as follows:

[0251] (5-1) Position control loop: Assume the target position is x target , the current position of the rope climbing robot is x, then the position deviation is e p for:

[0252] e p =x target -x

[0253]

[0254] (5-2) Speed ​​control loop: target speed v target The output from the position control loop, the actual speed v is the current real-time speed of the rope climbing robot, and the speed deviation e v for:

[0255] e v =v target -v;

[0256] The speed is regulated using a PID controller to generate a force F P , used to overcome resistance and achieve speed adjustment:

[0257]

[0258] k pv , k iv and k dv is the PID gain of the speed control loop;

[0259] (5-3) System dynamics model: Based on Newton's second law, the motion state of the rope-climbing robot is described, and its acceleration a is expressed as:

[0260]

[0261] The velocity v and displacement x can be updated by numerical integration method:

[0262] v(t+Δt)=v(t)+a·Δt

[0263] x(t+Δt)=x(t)+v(t)·Δt;

[0264] Among them, Δt is the time step;

[0265] (5-4) Combination of hierarchical controllers: The core of hierarchical controllers is to achieve an effective combination of position control loops and speed control loops. The specific implementation steps are as follows:

[0266] (5-4-1) The position control loop is based on the position deviation e p Generate target speed v target ;

[0267] (5-4-2) The speed control loop is based on the speed deviation e v Generate control force F P ;

[0268] (5-4-3) Using the kinetic model, according to F P Update the speed and position of the rope climbing robot;

[0269] (5-4-4) Continuously adjust the gains of the speed and position controllers to achieve more precise and smooth control;

[0270] This hierarchical control strategy combines the overall trajectory tracking function of the position control loop and the fine adjustment function of the speed control loop to achieve smooth and precise movement of the rope-climbing robot on the rope.

[0271] (6) Design an adaptive control system so that the rope-climbing robot can automatically adjust the clamping force according to the thickness of the rope and the surface friction coefficient. The specific process is as follows:

[0272] (6-1) Establishing friction model: The core of the gripping mechanism is to prevent the rope climbing robot from sliding by friction. The friction force F friction The calculation is as follows:

[0273] F friction =μ·N;

[0274] Where N is the normal force exerted by the gripping mechanism on the rope;

[0275] In this design, the variation of μ and N enables the gripping mechanism to automatically adapt to different friction requirements;

[0276] (6-2) Determination of target friction: Assume that the system requires the minimum friction F required To ensure stability, we have:

[0277] F required ≥ mg;

[0278] To prevent the rope-climbing robot from sliding on different ropes, the friction force F friction Must be greater than or equal to F required ,Right now:

[0279] μ·N≥mg;

[0280] Therefore, the normal force N must satisfy:

[0281]

[0282] (6-3) Adaptive control of clamping force: By measuring the diameter D and friction coefficient μ of the rope, the adaptive system dynamically adjusts the clamping force N to meet

[0283] (6-4) Machine learning-based control strategy: Use one of the machine learning algorithm regression or neural network to predict the required clamping force N. The specific process is as follows:

[0284] First, data collection is performed to collect data sets of different rope diameters D, friction coefficients μ, and required clamping forces N. Second, a model is trained using a regression algorithm to train a model f(D, μ) to predict the required clamping force: N = f(D, μ). Finally, in actual applications, the rope's D and μ are obtained in real time through sensors and input into the model to predict the required N.

[0285] (6-5) Control formula of adaptive materials: When using adaptive materials to adjust the clamping force, it is necessary to use the physical properties of the material to adjust N:

[0286] F m =k m ΔL;

[0287] Among them, k m is the equivalent elastic coefficient of the material; ΔL is the deformation of the material, which can be adjusted by adjusting F m , controls the normal force N applied to the rope;

[0288] (6-6) Mathematical model of control strategy: The adaptive control system is represented by the following control strategy:

[0289] The input layer receives the thickness D and friction coefficient μ of the rope and calculates the clamping force N:

[0290]

[0291] Where f(D,μ) is a machine learning prediction model; real-time adjustment of the clamping force, when the rope diameter or friction coefficient changes, recalculate N, adjust the adaptive material or control the clamping mechanism to achieve the required N;

[0292] (6-7) Feedback loop control: Introduce a feedback loop to monitor the friction force F in real time through sensors actual , and the friction force F required Compare them and if the deviation is large, adjust N:

[0293] ΔN=k f ·(F required -F actual );

[0294] k f is the feedback control gain of the adaptive clamping system;

[0295] The system can automatically adjust the clamping force according to the thickness and friction coefficient of the rope, ensuring sufficient friction in different environments and enhancing the versatility and adaptability of the machine;

[0296] (7) Use laser ranging, visual sensing, gyroscope and other sensors to perform data fusion, plan the best climbing path in real time, avoid obstacles or automatically optimize the climbing speed and direction. The specific process is as follows:

[0297] (7-1) Obstacle detection: Use laser ranging and visual sensors to detect the position of obstacles in the surrounding environment in real time. The distance measured by the laser ranging sensor is expressed as:

[0298]

[0299] Where d is the distance of the obstacle; c is the speed of light; t la is the round trip time of the laser pulse;

[0300] Assume that the current position of the rope climbing robot is (x0, y0, z0), then the three-dimensional coordinates of the obstacle (x obs ,y obs ,z obs ) is expressed as:

[0301] x obs =x0+d·cos(η)·sin(φ)

[0302] y obs =y0+d·sin(η)·sin(φ)

[0303] z obs =z0+d·cos(φ)

[0304] Among them, η and φ are the horizontal and vertical angles, respectively, which are used to determine the relative orientation of obstacles;

[0305] (7-2) Data fusion and state estimation: The position and speed of the rope-climbing robot are estimated through the fusion of visual sensors, laser ranging and gyroscope data to generate a real-time environment map. The specific process is as follows:

[0306] Kalman filtering is a commonly used data fusion method. Assume that the machine state is: The state update formula is:

[0307] X k+1 =A·X k +B·u k +w k

[0308] Among them, A is the state transfer matrix, which describes the dynamics of the system; B is the control matrix; u k is the control input; w k is the process noise;

[0309] The observation equation is:

[0310] Z k =H·X k +v k ;

[0311] Among them, Z k is the observed value, i.e. the position or velocity measured by the sensor; H is the observation matrix; v k is the observation noise;

[0312] The prediction and update steps of the Kalman filter are described as follows:

[0313]

[0314] P k|k-1 =A·P k-1 ·A T +Q

[0315] Update steps:

[0316] K k =P k|k-1 ·H T ·(H·P k|k-1 ·H T +R) -1

[0317]

[0318] P k =(IK k ·H)·P k|k-1

[0319] Among them, K k is the Kalman gain; Q is the process noise covariance matrix; R is the observation noise covariance matrix;

[0320] (7-3) Path planning, the specific process is as follows:

[0321] (7-3-1) The A* algorithm selects a path through the heuristic estimation function f(n), f(n) = g(n) + h(n): g(n) is the known path cost from the starting node to node n; h(n) is the estimated cost from node n to the target node. The specific formula is:

[0322] f(n)=g(n)+h(n);

[0323] At each step, the node with the lowest f(n) value is selected for expansion until the target node is found;

[0324] (7-3-2) Smooth the generated path to avoid large direction changes during the climbing process of the rope climbing robot. The path smoothing formula is:

[0325] path smooth (t) = λ·path raw (t)+(1-λ)·path smooth (t-1);

[0326] Among them, λ is the smoothing coefficient;

[0327] (7-4) Motion control: After the path planning is completed, the control system moves the machine along the path by determining the climbing speed v and the direction angle σ. The speed control formula is:

[0328] Assume the coordinates of the target point are (x goal ,y goal ,z goal ), the current position is (x, y, z), and the unit vector of the target direction is:

[0329]

[0330] The velocity vector v is then:

[0331]

[0332] Among them, v max is the maximum speed of the system;

[0333] (7-5) Direction control: The control formulas for the direction angle σ and the pitch angle ψ are:

[0334]

[0335] The present invention uses a drone carrying a glass-breaking bomb to break the window glass and release a static rope. The rope-climbing robot enters the target room along the static rope for rescue, thereby improving positioning accuracy and rescue efficiency. The flexibility of the drone and the high-performance rope-climbing robot are utilized to achieve precise positioning and flexible movement. The drone and rope-climbing robot system does not need to rely on physical guide rails and can flexibly respond to various building structures and complex environments. The static rope system can be quickly and safely fixed and used for the rope-climbing robot to ensure normal operation under various conditions, thereby improving the adaptability and reliability of rescue, and significantly improving the efficiency and safety of high-rise building fire rescue.

Claims

1. A high-rise building fire rescue system, comprising a drone (1), characterized in that: The drone (1) carries a glass-breaking bomb (2), a static rope (3) and a rope-climbing robot (4); the glass-breaking bomb (2) is mounted on a launching device of the drone (1) and comprises a high-strength spring and an impact head; during rescue, one end of the static rope (3) is fixed to the outside of a building (5) by the drone (1) through a quick locking device, and the other end is dropped into a fire room (6); the rope body is used to provide a moving track for the rope-climbing robot (4); the rope-climbing robot (4) comprises a battery system, a digital motor, a planetary reducer, a safety buckle, an emergency braking device and a control module.

2. A high-rise building fire rescue system control method, characterized in that: The steps include: Step 1: The drone (1) carries the glass-breaking bomb (2), the static rope (3) and the rope-climbing robot (4) and flies to the outside of the fire room (6) through the navigation of the positioning module; Step 2: The glass-breaking bullet (2) is launched through the ejection device of the drone (1), and the high-strength spring releases energy to push the impact head to impact, thereby breaking the glass on the window of the fire room (6); Step 3: The drone (1) fixes one end of the static rope (3) outside the fire room (6) and drops the other end into the fire room (6); Step 4: Place the rope-climbing robot (4) on the static rope (3). Under the control of the control module, the rope-climbing robot (4) moves along the static rope (3) to the fire room (6) in remote control or automatic mode, and then transports personnel or materials to the ground.

3. The high-rise building fire rescue system control method according to claim 2, characterized in that: The control process of the control module of the rope-climbing robot (4) in step 4 is as follows: (1) Establish a motion model of the rope-climbing robot (4) in the non-powered release state, simulate the descent speed and displacement trajectory of the machine, and realize the estimation and optimization of the motion process of the rope-climbing robot (4): (1-1) Force analysis: Gravity analysis: F g = mg; where m is the mass of the rope-climbing robot (4) and g is the acceleration due to gravity; Static rope (3) Friction analysis: F friction = μmgcos(θ); where μ is the friction coefficient; θ is the inclination angle of the static rope (3) relative to the horizontal plane; cos(θ) is the component of the static rope (3) in the horizontal direction, which is used to calculate the friction resistance along the static rope (3); Wind resistance analysis: Where ρ is the air density; C d A is the air resistance coefficient; area is the windward area of ​​the rope-climbing robot (4); v m The machine descending speed; (1-2) Equation of motion: The resultant force along the static rope (3) is: F T =F g sin(θ)-F friction -F fall cos(θ); Substituting into the force analysis formula, we get: According to Newton's second law, the acceleration a along the static rope (3) is: Simplifying, we get: (1-3) Velocity and displacement equations: Assuming the initial velocity is v0, the instantaneous velocity v(t) of the rope-climbing robot (4) in the no-power release state is obtained by integrating the acceleration equation: The displacement x(t) is obtained by further integrating the velocity equation: (2) Introducing intelligent sensors and algorithms, a real-time rope tension feedback control model is established, which enables the rope climbing robot (4) to automatically detect the rope tension, ambient wind speed and angle parameters, and automatically adjust the rotor speed to change the speed: (2-1) Static rope (3) Rope tension and rope climbing robot (4) Force balance analysis: Assume that the rope-climbing robot (4) moves along a rope with a tension T rope Affected by the weight, swing angle and wind speed of the rope climbing robot (4), the rope tension is expressed as: Where, δ is the swing angle; L is the rope length; (2-2) Analysis of lateral force under the influence of wind speed: The wind speed generates a lateral force F on the rope-climbing robot (4) wind The lateral wind force that affects the balance of the machine is approximately expressed as: (2-3) The control formula for automatically adjusting the wheel speed is: In order to realize automatic adjustment control based on sensor feedback, it is assumed that the goal of the rope climbing robot (4) is to maintain a stable moving speed v target , the proportional-integral-differential PID controller is used to adjust the wheel speed according to the deviation. The PID control formula is: Where u(t) is the control output, i.e. the adjustment amount of the wheel; e(t) = v target -v(t) is the speed deviation, that is, the difference between the target speed and the current speed; K p , K i and K d are proportional, integral and differential gain coefficients respectively; (2-4) Combined tension and angle control feedback: Feedback control of tension and swing angle is introduced to adjust the rotation speed of the wheel to cope with the swing of the static rope (3). Assume that the sensor measures the current tension T rope and angle δ, introduce angle compensation term and tension compensation term to further improve the control formula: Among them, K δ K is the angle compensation gain, used to cope with angle changes; T is the tension compensation gain; T target is a preset tension target value used to maintain the balance of the rope-climbing robot (4); (2-5) Dynamic adjustment of ambient wind speed: Changes in wind speed will affect the stability of the machine. The control system can be further optimized through wind speed feedback. Assuming that the sensor monitors the wind speed v in real time wind , introduce the wind speed compensation term: Among them, K v is the wind speed compensation gain; (2-6) Wheel angular velocity adjustment: According to the controller output u(t), the angular velocity ω of the wheel can be calculated: Where r is the wheel radius; (3) In view of the influence of wind speed in the environment, internal shock absorption and windproof stability design are added: (3-1) In order to reduce the lateral swing of the machine caused by wind, a damping device is designed inside the rope climbing robot (4). The damping force F damping Proportional to speed: F damping =-b·v; Where b is the damping coefficient; v is the instantaneous speed of the machine; (3-2) Using a gripping pad made of elastic material, when the rope climbing robot (4) is disturbed by wind, an elastic restoring force will be generated to try to restore the machine to the equilibrium position. The elastic restoring force is expressed by Hooke's law: F spring =-k·x; Where k is the spring constant; x is the displacement, i.e., the distance that the rope-climbing robot (4) deviates from the equilibrium position; (3-3) Under the combined action of wind force and shock absorption / elastic device, the lateral motion of the rope climbing robot (4) is expressed by the following differential equation: After expansion: in, is the lateral acceleration of the rope-climbing robot (4); (4) A nonlinear controller that is resistant to wind disturbance is introduced to automatically adjust the grip or release force of the machine based on real-time monitoring of the ambient wind speed to prevent the rope climbing robot (4) from deviating from the control path due to wind influence: (4-1) To offset the influence of wind on the rope climbing robot (4), an anti-wind control force F is applied. control , so that the rope-climbing robot (4) remains at the target position, using the following control formula: k p and k d is the gain parameter of position and speed control; x target is the target position, x is the displacement; f nonlinear (x) is a nonlinear compensation function used to cope with disturbances under high wind speeds; (4-2) The gripping or releasing force T of the rope climbing robot (4) is related to the wind force, and the gripping force is calculated using the following formula: T=μ·(m·g++F control -F wind ); (4-3) To achieve more stable wind resistance control, the hyperbolic tangent function tanh is used to deal with rapidly changing wind speeds: f nonlinear (x)=β1·tanh(β2·(v-v threshold )); Among them, β1 and β2 are adjustment parameters; v threshold is the threshold value. When the wind speed exceeds this value, nonlinear compensation begins to become significant. (4-4) Update the displacement and velocity of the machine by numerical integration: x(t++Δt)=x(t)++v(t)·Δt (5) The speed control loop and position control loop in the hierarchical control strategy are designed to achieve smooth and precise movement of the rope-climbing robot (4) on the rope. The position control loop is responsible for ensuring that the machine moves along the predetermined overall motion trajectory, and the speed control loop is used for fine adjustment to enable the rope-climbing robot (4) to stop smoothly at the target position. The specific process is as follows: (5-1) Position control loop: Assume the target position is x target , the current displacement of the rope-climbing robot (4) is x, then the position deviation e p for: e p =x target -x (5-2) Speed ​​control loop: target speed v target The output from the position control loop, the actual speed v is the current real-time speed of the rope-climbing robot (4), and the speed deviation e v for: yes v =v target -v; The speed is regulated using a PID controller to generate a force F P , used to overcome resistance and achieve speed adjustment: k pv , k iv and k dv is the PID gain of the speed control loop; (5-3) System dynamics model: Based on Newton's second law, the motion state of the rope-climbing robot (4) is described, and its acceleration a is expressed as: The velocity v and displacement x are updated by numerical integration method: v(t+Δt)=v(t)+a·Δt x(t++Δt)=x(t)++v(t)·Δt; Among them, Δt is the time step; (5-4) Combination of hierarchical controllers: The core of hierarchical controllers is to achieve an effective combination of position control loops and speed control loops. The specific implementation steps are as follows: (5-4-1) The position control loop is based on the position deviation e p Generate target speed v target ; (5-4-2) The speed control loop is based on the speed deviation e v Generate control force F P ; (5-4-3) Using the kinetic model, according to F P Update the speed and position of the rope-climbing robot (4); (5-4-4) Continuously adjust the gains of the speed and position controllers to achieve more precise and smooth control; (6) Design an adaptive control system so that the rope climbing robot (4) automatically adjusts the clamping force according to the thickness of the rope and the surface friction coefficient. The specific process is as follows: (6-1) Establishing friction model: The core of the gripping mechanism is to prevent the rope-climbing robot (4) from sliding by friction. The friction force F friction The calculation is as follows: F friction =μ·N; Where N is the normal force exerted by the gripping mechanism on the rope; (6-2) Determination of target friction: Assume that the system requires the minimum friction F required To ensure stability, we have: F required ≥mg; To prevent the rope-climbing robot (4) from sliding on different ropes, the friction force F friction Must be greater than or equal to F required ,Right now: μ·N≥mg; Therefore, the normal force N must satisfy: (6-3) Adaptive control of clamping force: By measuring the diameter D and friction coefficient μ of the rope, the adaptive system dynamically adjusts the clamping force N to meet (6-4) Machine learning-based control strategy: Use one of the machine learning algorithm regression or neural network to predict the required clamping force N. The specific process is as follows: First, data collection is performed to collect data sets of different rope diameters D, friction coefficients μ, and required clamping forces N. Second, a model is trained using a regression algorithm to train a model f(D, μ) to predict the required clamping force: N = f(D, μ). Finally, in actual applications, the rope's D and μ are obtained in real time through sensors and input into the model to predict the required N. (6-5) Control formula of adaptive materials: When using adaptive materials to adjust the clamping force, it is necessary to use the physical properties of the material to adjust N: F m =k m ·ΔL; Among them, k m is the equivalent elastic coefficient of the material; ΔL is the deformation of the material, which can be adjusted by adjusting F m , controls the normal force N applied to the rope; (6-6) Mathematical model of control strategy: The adaptive control system is represented by the following control strategy: The input layer receives the thickness D and friction coefficient μ of the rope and calculates the clamping force N: Where f(D,μ) is a machine learning prediction model; real-time adjustment of the clamping force, when the rope diameter or friction coefficient changes, recalculate N, adjust the adaptive material or control the clamping mechanism to achieve the required N; (6-7) Feedback loop control: Introduce a feedback loop to monitor the friction force F in real time through sensors actual , and the friction force F required Compare them and if the deviation is large, adjust N: ΔN=k f ·(F required -F actual ); k f is the feedback control gain of the adaptive clamping system; (7) Use laser ranging, visual sensing, gyroscope and other sensors to perform data fusion, plan the best climbing path in real time, avoid obstacles or automatically optimize the climbing speed and direction. The specific process is as follows: (7-1) Obstacle detection: Use laser ranging and visual sensors to detect the position of obstacles in the surrounding environment in real time. The distance measured by the laser ranging sensor is expressed as: Where d is the distance of the obstacle; c is the speed of light; t la is the round trip time of the laser pulse; Assume that the current position of the rope-climbing robot (4) is (x0, y0, z0), then the three-dimensional coordinates of the obstacle (x obs ,y obs ,z obs ) is expressed as: x obs =x0++d·cos(η)·sin(φ) y obs =y0++d·sin(η)·sin(φ) With obs =z0+d cos(φ) Among them, η and φ are the horizontal and vertical angles, respectively, which are used to determine the relative orientation of obstacles; (7-2) Data fusion and state estimation: The position and speed of the rope-climbing robot (4) are estimated by fusing the data of the visual sensor, laser ranging and gyroscope to generate a real-time environment map. The specific process is as follows: Kalman filtering is a commonly used data fusion method. Assume that the machine state is: The state update formula is: X k+1 =A·X k +B·u k +w k Among them, A is the state transfer matrix, which describes the dynamics of the system; B is the control matrix; u k is the control input; w k is the process noise; The observation equation is: Z k =H·X k ++v k ; Among them, Z k is the observed value, i.e. the position or velocity measured by the sensor; H is the observation matrix; v k is the observation noise; The prediction and update steps of the Kalman filter are described as follows: P k|k-1 =A·P k-1 ·A T ++Q Update steps: K k =P k|k-1 ·H T ·(H·P k|k-1 ·H T ++R) -1 P k =(I-K k ·H)·P k|k-1 Among them, K k is the Kalman gain; Q is the process noise covariance matrix; R is the observation noise covariance matrix; (7-3) Path planning, the specific process is as follows: (7-3-1) The A* algorithm selects a path through the heuristic estimation function f(n), f(n) = g(n) + h(n): g(n) is the known path cost from the starting node to node n; h(n) is the estimated cost from node n to the target node. The specific formula is: f(n)=g(n)++h(n); At each step, the node with the lowest f(n) value is selected for expansion until the target node is found; (7-3-2) Smooth the generated path. The path smoothing formula is: path smooth (t)=λ·path raw (t)++(1-λ)·path smooth (t-1); Among them, λ is the smoothing coefficient; (7-4) Motion control: After the path planning is completed, the control system makes the machine move along the path by determining the climbing speed v and the direction angle σ. The speed control formula is: Assume the coordinates of the target point are (x goal ,y goal ,z goal ), the current position is (x, y, z), and the unit vector of the target direction is: The velocity vector v is then: Among them, v max is the maximum speed of the system; (7-5) Direction control: The control formulas for the direction angle σ and the pitch angle ψ are:

Images