Fire-fighting unmanned aerial vehicle speed regulation method based on visual detection

Abstract

The invention relates to the technical field of unmanned aerial vehicle speed regulation, and particularly discloses a fire-fighting unmanned aerial vehicle speed regulation method based on visual detection, which comprises the following steps that S1, an unmanned aerial vehicle shoots a fire scene environment image through an airborne camera, identifies all obstacles and calculates the distance D between the obstacles and the unmanned aerial vehicle; attitude data such as angular velocity and acceleration of the unmanned aerial vehicle, electromagnetic braking current intensity of a slow descent strip and friction plate pressure of mechanical damping are collected; s2, setting a safe distance threshold value Dt, if the obstacle distance D is greater than or equal to Dt, generating a slow descent instruction according to the attitude data, and dynamically adjusting electromagnetic braking current and friction plate pressure, so that the unmanned aerial vehicle slowly descends at a speed of [0.2, 1.0] m / s; if D is smaller than Dt, an emergency stop instruction is triggered, the speed is reduced to 0.3 m / s or below, and the position of the unmanned aerial vehicle is adjusted to avoid the obstacle; intelligent speed regulation and emergency obstacle avoidance are achieved by visually detecting the obstacle distance, and safe slow descent is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle speed regulation, in particular to a fire-fighting unmanned aerial vehicle speed regulation method based on visual detection. BACKGROUND

[0002] Currently, the application of fire-fighting unmanned aerial vehicles in high-rise building rescue faces three core problems: poor slow descent stability, insufficient high-temperature resistance, and single function, which seriously restricts their practicality and reliability in complex fire environment.

[0003] Firstly, in terms of slow descent stability, traditional slow descent ropes are mostly made of steel cables or ordinary high-molecular fiber materials, which have high ductility and are prone to elastic deformation under dynamic load, resulting in speed fluctuations and oscillation during descent. For example, steel cables soften easily at high temperatures, and ordinary nylon ropes have an elongation rate of more than 5% when under stress, making it difficult to accurately control the descent trajectory of the personnel or supplies carried. In addition, traditional braking systems (such as pure mechanical friction braking) have a slow response speed (> 200 ms) and cannot adjust the braking force in real time, resulting in acceleration overshoot or emergency braking failure, increasing the risk of rescue.

[0004] Secondly, in terms of high-temperature resistance, the existing slow descent rope materials cannot adapt to the extreme environment of the fire scene. Conventional polymer fibers (such as polyester and nylon) have a sharp decrease in strength when the temperature exceeds 200°C, while metal cables, although resistant to high temperatures, are heavy and increase the load on the unmanned aerial vehicle. High-temperature radiation, flying sparks and chemical corrosion in the fire scene further accelerate material aging, leading to an increased risk of rope breakage. For example, in a certain fire case, the unmanned aerial vehicle slow descent rope suddenly failed after being locally heated to 300°C, resulting in a failed rescue.

[0005] Finally, in terms of functional expansion, traditional slow descent ropes can only achieve a single vertical descent function and lack integrated design. For example, they cannot transmit power or data simultaneously and need to carry additional equipment, occupying the effective payload of the unmanned aerial vehicle; they also do not integrate rescue modules such as escape ladders and quick-release hooks, resulting in low efficiency in multi-task execution.

[0006] In view of the above problems, there is an urgent need for a high-strength, high-temperature-resistant, multi-functional integrated unmanned aerial vehicle slow descent rope system for unmanned aerial vehicle assisted landing deceleration; through material innovation and mechatronic collaborative design, stable unmanned aerial vehicle slow descent, environmental adaptation and functional expansion are achieved to meet the stringent demands of modern fire rescue. SUMMARY

[0007] The purpose of the present application is to provide a fire-fighting unmanned aerial vehicle speed regulation method based on visual detection, which solves the following technical problems.

[0008] The purpose of the present application can be achieved by the following technical solutions: A fire-fighting unmanned aerial vehicle speed regulation method based on visual detection, comprising the following steps: Step S1: Obtain environmental images of the fire scene based on the shooting points mounted on the UAV, analyze the environmental images to obtain all obstacles in the environmental images, the shooting points are used to capture images, and monitor the obstacle distance D of the obstacles, the obstacle distance is the distance between the obstacle and the UAV; The descent path of the UAV is determined based on the environmental image, and the attitude data of the UAV, including angular velocity and acceleration, is obtained. The current intensity of the electromagnetic braking of the UAV descent cable and the friction plate pressure of the mechanical damping are also obtained. Step S2: Set a safety threshold Dt. If D≥Dt, generate a start descent command based on the UAV attitude data, transmit the descent command to the UAV descent cable, and dynamically adjust the current intensity and friction plate pressure so that the descent speed of the UAV is in real time at [0.2, 1.0] m / s. If D < Dt, an emergency stop command is generated based on the UAV attitude data. The descent speed is adjusted to below 0.3 m / s based on the emergency stop command, and the UAV's position is adjusted to avoid the obstacle.

[0009] As a further aspect of the present invention: the shooting point is based on an infrared thermal imaging and a visible light dual-spectrum camera, the infrared thermal imaging is used to identify obstacles in a dense smoke environment, and the visible light dual-spectrum camera is used to identify obstacles in a clear environment; the environmental image includes a heat map obtained by infrared thermal imaging and a multispectral image obtained by the visible light dual-spectrum camera.

[0010] As a further aspect of the present invention, the process of analyzing the environmental image includes: The environmental image is divided into several pixels, the pixel value of each pixel is obtained, and a pixel value range threshold is set. If the pixel value belongs to the pixel value range threshold, the pixel is recorded as an obstacle pixel. The total number of obstacle pixels in the environmental image is obtained, and the area ratio of each pixel is obtained. Based on the total number and the area ratio, the area ratio of obstacles in the environmental image is obtained. If the area occupied by the obstacle exceeds 15%, the angular velocity of the drone is obtained, and the relative motion trend between the obstacle and the drone is predicted based on the angular velocity.

[0011] As a further aspect of the present invention: the process of dynamically adjusting the current intensity and friction plate pressure includes: When the shooting point detects that the distance to the obstacle of the drone is greater than or equal to the safety threshold, the electromagnetic braking is based on the permanent magnet synchronous motor bearing 60% to 80% of the braking force, and the descent speed of the drone is controlled by adjusting the excitation current of the permanent magnet synchronous motor. When the shooting point detects that the distance to the obstacle of the drone is less than the safety threshold, the rated value of the friction plate pressure of the mechanical damping is obtained, and the friction plate pressure is increased to 120% of the rated value. The mechanical damping and electromagnetic braking simultaneously adjust the response time so that the response time is less than 100ms. The response time is the time interval between the trigger moment and the deceleration moment. The trigger moment is the moment when the drone generates the descent command, and the deceleration moment is the moment when the drone begins to decelerate.

[0012] As a further aspect of the present invention: the safety threshold is set based on a preset load-threshold mapping table, which is a pre-stored data table used to store the correspondence between different load weights and safety thresholds.

[0013] As a further aspect of the present invention: the process of acquiring environmental images of the fire scene from the shooting point includes acquiring the descent path of the drone, and the shooting point performing a panoramic scan of the fire scene below the drone every 200ms to generate a dynamic heat map containing the distribution of obstacles.

[0014] As a further aspect of the present invention: the process of determining the descent path of the UAV based on the environmental image includes: The dynamic heat map is divided into several regions, the obstacle density in each region is obtained, and the horizontal position of the UAV is adjusted in real time so that the UAV is always directly above the region with the lowest obstacle density, thus obtaining the descent path.

[0015] As a further aspect of the present invention, the drone also includes a manual intervention mechanism. When the drone is in the manual intervention mechanism, ground personnel directly set the descent speed of the drone and adjust the current intensity of the electromagnetic brake and the friction plate pressure of the mechanical damping according to the descent speed.

[0016] The beneficial effects of this invention are: The drone of this invention is highly maneuverable and can quickly reach the fire scene. Its descent system enables precise delivery and transport of supplies or personnel, shortening rescue time. Through its payload mounting and release mechanism, power supply and signal transmission scheme, and other design features, the system is compatible with various devices, such as life detectors and communication relay equipment, meeting diverse rescue needs. Furthermore, it can switch between "personnel rescue mode" and "supply delivery mode" within 5 minutes, with a maximum payload extended to 150kg, enhancing rescue flexibility.

[0017] This invention employs a hybrid braking scheme combining mechanical damping and electromagnetic braking, along with dynamic modeling and simulation optimization, to achieve a controllable descent speed of [0.2, 1.0] m / s, with an acceleration not exceeding 0.5g, and a swing amplitude ≤5° under ±15° wind direction deviation, avoiding injury and equipment damage caused by falls or excessive swing. The descent cable uses an ultra-high molecular weight polyethylene-aramid blended composite cable with a breaking strength >2 times the maximum working load, a safety factor ≥3, and can withstand a rated load of 20kg and an impact load of 150kg. Simultaneously, the system features a self-locking structure and a dual-stage unlocking hook. With redundant design, emergency release can be completed via mechanical manual pull ring in the event of communication interruption or power failure, ensuring rescue safety; the drone adopts high-temperature resistant materials (such as lithium batteries with operating temperature of -20℃ to 120℃ and carbon fiber propellers with short-term temperature resistance of 300℃) and protective design, and the surface of the descent cable has a nano-coating, which can work stably in harsh environments such as high temperature, humidity, and smoke; for example, the materials can still maintain stable mechanical properties at a high temperature of 800℃, and the key performance of the descent cable is not less than 90% of the initial value under conditions of -20℃ to 60℃ and humidity of 0% to 95%RH.

[0018] This invention establishes a continuous medium and multi-rigid-body dynamic model through static and dynamic analysis, optimizes the damping system, and maintains the system damping ratio ζ in an underdamped state of 0.7-1.0, with a displacement overshoot ratio of <5%, and the time required for the amplitude to decay to 10% of the initial value of <2 cycles, thus suppressing vibration during the descent process and improving system stability. The system adopts a modular architecture, with each functional component connected through standardized interfaces, facilitating rapid replacement and maintenance. For example, the connection structure design between the ground anchoring device and the UAV platform takes into account the convenience of maintenance, improving the maintainability of the system.

[0019] Furthermore, this invention integrates knowledge from multiple disciplines such as materials mechanics, dynamics modeling, and control theory. It innovatively adopts a hybrid braking scheme combining a permanent magnet synchronous motor and a micro constant current valve, and a verification method combining multi-physics field coupled finite element simulation and field testing. This provides methodological support for the optimized design of fire-fighting drone descent systems and can be extended to fields such as high-altitude rescue and material delivery. By incorporating cables and optical fibers into the tethered cable, it achieves 50W power supply and 1Gbps data transmission, providing stable power and high-definition video links for mounted equipment. Combined with an AI control system and multimodal sensors, it can construct an integrated three-dimensional rescue network, contributing to the modernization and upgrading of smart fire-fighting and emergency rescue capabilities. Attached Figure Description

[0020] The invention will now be further described with reference to the accompanying drawings.

[0021] Figure 1 This is a schematic diagram illustrating the steps of a speed control method for a firefighting drone based on visual detection according to the present invention. Figure 2This is a schematic diagram illustrating the working principle of the descent cable in a speed control method for firefighting drones based on visual detection, as described in this invention. Figure 3 This is an exploded view of the drone-mounted deceleration cable structure in a visual detection-based speed control method for firefighting drones according to the present invention. Figure 4 This refers to the safe working range of the intelligent hook in the speed adjustment method for firefighting drones based on visual detection in this invention. Figure 5 This invention relates to a three-dimensional model of the descent cable structure of a multifunctional tethered firefighting drone, which is part of a speed control method for firefighting drones based on visual detection. Detailed Implementation

[0022] 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.

[0023] Please see Figure 1 As shown, this invention is a speed control method for firefighting drones based on visual detection. The designed drone is a multi-functional tethered firefighting drone. Due to its characteristics of fixed position but variable altitude and 24-hour continuous flight, tethered drones are perfectly suited to serve as guardians of command centers and other critical locations. As a firefighting drone, stable operation in fire scenes and rescue operations is a challenge that needs to be addressed. High temperatures, delays, airflow disturbances, and low visibility all pose severe tests for firefighting drones.

[0024] The system is equipped with a brushless motor, high-temperature magnets, and ceramic bearings, along with copper heat sinks. The propellers are made of lightweight and high-temperature resistant carbon fiber, and a short propeller design enhances resistance to currents. High-temperature resistant lithium batteries (operating temperature -20℃ to 120℃) are used, along with phase change material for heat dissipation. Ideally, a tethered power supply system could be used, providing continuous power via a high-temperature resistant cable. However, the limitations of tethered power supply include a limited range of movement, a relatively fixed location, and a radius of operation restricted by the cable length, which is detrimental to rescue operations.

[0025] Based on the above, the basic parameters of a drone under ideal conditions are shown in the table below:

[0026] In the design of the descent cable for this multi-functional tethered firefighting drone, the descent cable structure is crucial for rescue operations. This design incorporates a self-locking structure and a descent escape ladder as the descent cable structure. The self-locking structure consists of a servo motor, servo motor arm, self-locking shell, locking mechanism, locking device, and spring. The descent escape ladder consists of an escape ladder, descent sleeve, and buckle. The primary operating environment is in fire scenes requiring drone-based firefighting rescue. Its working principle is as follows: Figure 2 As shown, in Figure 2 In the diagram, A represents the window where the trapped person was located, B represents the wall containing the window, C represents the entire descent cable structure, and D represents the drone carrying the descent cable to perform the mission. During the rescue operation, the window, wall, descent cable, and drone form a stable triangular structure with stabilizing properties. The descent cable system has the function of securing the window and self-locking.

[0027] In the overall system architecture, the unmanned aerial vehicle platform, the descent cable structure, and the ground control unit constitute the three core modules, such as... Figure 3 As shown, the UAV platform is mainly responsible for flight control and payload docking, while the descent cable structure enables the slow descent of materials or personnel and provides damping and safety protection. The ground control unit is responsible for parameter issuance, real-time monitoring, and emergency response commands.

[0028] In a typical rescue mission, the pilot sends mission instructions via the ground control unit. The UAV platform, equipped with a descent cable structure, takes off and waits for further instructions at the designated location. The system first performs a self-check of cable tension and initial brake reset, then enters a steady-state waiting phase. Upon completion of the rescue and receipt of the "begin descent" command, the electronic control unit outputs drive pulses to the motors and valves according to a preset descent curve, achieving a constant or variable rate descent. During the descent, a camera point monitors the attitude and swing amplitude in real time. This camera point includes an inertial measurement unit and a vision sensor. If the safety threshold is exceeded, the controller immediately adjusts the braking force or sends an "emergency stop" command. After landing is completed and the hook lock disengagement is detected, the system automatically tightens the cable and resets the brakes, awaiting the next mission.

[0029] The descent principle is based on a combination of friction braking and electromagnetic braking: during the initial descent phase, friction limits the maximum acceleration; in the mid-to-late stages, the reverse power generation characteristic of the motor precisely adjusts the descent speed, ensuring that the acceleration does not exceed 0.5g and the speed can dynamically switch between 0.2m / s and 1.0m / s. The requirements analysis indicates that the drone can carry a payload of up to 20kg, with a maximum descent altitude of no more than 50m, and can maintain a swing amplitude of ≤5° even under ±15° wind direction deviation conditions.

[0030] Based on the mission scenario and safety specifications, the system must meet the following requirements: breaking strength > 2 × maximum working load, safety factor ≥ 3, brake response time < 100ms, and braking energy recovery efficiency ≥ 20%. Accelerated aging tests are conducted on the cable and brake under temperature (-20°C to 60°C) and humidity (0% to 95%RH) conditions to ensure that the key performance is not lower than 90% of the initial value.

[0031] The intelligent hook employs a dual-stage unlocking design. The main hook seat bears the weight and locks the hook, while the secondary hook seat provides a quick release function. After the release signal is jointly confirmed by the controller and onboard sensors, the secondary hook seat electromagnetically unlocks, while the main hook seat slowly unloads to a zero-tension state and then passively returns to its original position via a spring. Its safe operating range is as follows: Figure 4 As shown, this mechanism ensures that in the event of a communication interruption or power failure, an emergency release can be completed via a mechanical manual pull ring.

[0032] The mooring cable incorporates a four-core high-flexibility cable and optical fiber, capable of supplying 50 W of power and transmitting 1 Gbps of data respectively. This provides up to 30 minutes of stable power and a high-definition video link for mounted equipment such as life detectors or communication repeaters. Simultaneously, the outer layer of the cable employs a tensile- and torsion-resistant braided structure, wound together with the flexible cable to ensure uninterrupted signal and power transmission under ±15 Nm torque conditions.

[0033] Based on the above research and analysis of the working principle of the descent cable, the various parts were disassembled and designed. The structure included a servo motor, servo motor rocker arm, self-locking housing, locking mechanism, locking retainer, and spring self-locking structure. Three-dimensional models of these parts were then created, such as... Figure 5 As shown, the models are the descent cable structure model and the UAV structure model, respectively. The modeling can more intuitively demonstrate the function of the overall system and lay the foundation for further research.

[0034] To ensure that the descent cable has high strength, stability and safety in fire rescue, the material, cross-section design, speed control mechanism and safety redundancy design of the descent cable are selected through mechanical analysis and calculation.

[0035] For the candidate materials, specific strength, elastic modulus, and environmental adaptability are comprehensively considered. The specific strength is calculated as B = σ b / ρ, where σ b ρ represents the tensile strength (also called ultimate tensile strength) of a material, and ρ represents the density of the material; in contrast, ultra-high molecular weight polyethylene (UHMWPE) has a specific strength of 3.09 × 10⁻⁶. 6 Pa·m 3 / kg, low density but low elastic modulus and poor high temperature resistance), aramid (specific strength 2.50×10 6 Pa·m 3 / kg, good fatigue resistance and impact toughness), carbon fiber composite wire harness (specific strength 2.57×10 6 Pa·m 3 / kg, high strength but brittle and costly); ultimately, ultra-high molecular weight polyethylene-aramid blended composite cable was chosen, balancing high specific strength, elastic modulus, and environmental durability; its equivalent ultimate strength σ b =2.8GPa, meeting the requirements for material strength and weather resistance in fire protection scenarios.

[0036] For the cross-sectional dimensions design, based on the maximum load mass of 20kg and the coefficient of inertia of 1.2, the maximum tensile force F is calculated. max =235.4N, combined with the safety factor n s =3, thus obtaining the allowable stress σ allow =9.33×10 8 Pa, minimum cross-sectional area A min =2.52×10 - 7 m 2 The design features a circular cross-section with a diameter d = 0.62 mm, an outer diameter of 1.2 mm, and a mass per unit length of 3.6 × 10⁻⁶ mm. -4 kg / m. Maximum axial stress under rated load σ = 7.85 × 10⁻⁶ kg / m. 8 Pa<σ allow σ under impact load imp =9.81×10 8 Pa < yield strength 1.1 × 10 9 Pa ensures that the cable works in the elastic stage, with only a small amount of plastic deformation under extreme impact.

[0037] For the selection of the descent speed control mechanism, a dual-disc friction brake with a friction coefficient μ=0.15 and a normal preload F is chosen. N =500N, average friction radius r m =0.02m, braking torque T=1.5N·m, corresponding braking force F b =T / r m =150N, maximum deceleration 7.5m / s 2 This can limit the peak acceleration in the initial stage by adjusting F. N Achieve continuous adjustment of braking force [50, 200]N. The electromagnetic brake uses a permanent magnet synchronous motor, and the braking force F at a stable speed v = 0.5 m / s... d =196.2N, torque T d =1.962 N·m, feedback power 122.5 W, energy recovery rate 12.5%; the hydraulic solution uses a single-acting miniature constant flow valve, valve port area A v =1.1×10 -7 m 2With an efficiency of 85%, the two options can be flexibly selected according to task requirements.

[0038] For safety redundancy design, the static load safety factor n s ≥3, impact safety factor σ imp ≥1.5 (actually taken as 1.5), fatigue safety factor n f ≥2 ensures the reliability of the descent cable under static load, impact, and fatigue load. Multiple safeguards are provided through material selection, cross-section design, and braking mechanisms, such as hybrid braking schemes and dual-stage unlocking hooks, to enhance the system's safety margin under complex working conditions and ensure stability in repeated use and harsh environments.

[0039] This study conducts multi-dimensional stress analysis and constructs mathematical models to establish multi-scenario stress analysis. Specifically, it clarifies that when the UAV hovers, the descent cable is affected by the weight of the attached load and its own distributed mass, establishes a static tension distribution equation along the cable length, and derives the relationship between the maximum static tension and the attached mass, cable density, and UAV altitude, providing a static basis for cable strength selection. During the descent process, it analyzes the coupling effect of braking force, air resistance (including wind speed interference), and gravity, distinguishes the stress characteristics of uniform deceleration and variable acceleration stages, constructs transient dynamic equations, accurately captures the changes in velocity and acceleration over time, and supports the optimization of braking schemes. For high-temperature thermal radiation in fire scenes (simulating temperature field loading), strong winds (setting wind speed boundary conditions), and impact loads (such as sudden impact from falling attached loads), it quantifies the superposition effect of thermal stress, aerodynamic loads, and impact stress, clarifies dangerous load combinations, and provides critical parameters for safety redundancy design.

[0040] Based on a continuum model, the descent cable is treated as a flexible body. A nonlinear finite element model is established based on the principle of virtual work, considering large deformation and contact friction (with the braking mechanism and hook). The governing equations are derived using the Hamilton variational method to achieve a refined simulation of cable vibration and stress distribution. Based on a multi-rigid-body dynamics model, the UAV-descent cable-load system is simplified into a multi-rigid-body system. Hinge constraints and force elements (such as braking torque and gravity) are defined. The system dynamics model is established using the Lagrange equations to quickly calculate the overall motion response (such as UAV attitude deviation and cable swing angle), adapting to the requirements of preliminary engineering design. Based on a coupled-field model, fluid-structure interaction (aerodynamic interaction between air and cable, UAV) and thermal-structure interaction (the influence of the fire environment on the mechanical properties of the cable) are integrated. Sequential coupling or direct coupling algorithms are used to numerically solve the system response under multi-physics fields, revealing the impact of complex phenomena such as thermal softening and aerodynamic flutter on descent safety.

[0041] Design scaled-down tests (such as building a test bench for the descent cable to simulate different loads and wind speeds), collect data on cable strain, descent speed, and UAV attitude, compare these data with model calculations, and correct material constitutive models (such as strength decay models at high temperatures) and boundary conditions (such as aerodynamic drag coefficients) to ensure model accuracy. Based on the validated model, optimize the descent cable cross-sectional parameters (such as blending ratio and diameter) and braking system parameters (such as braking torque adjustment range), predict system failure risks under extreme conditions, guide safety redundancy design (such as dual-stage braking and self-locking mechanism configuration), and promote the transformation of theoretical models into actual rescue equipment.

[0042] A simulation experiment was established, and an integrated test bench was built, which included a UAV, braking device, sensor acquisition, and environmental simulation. A development experiment management platform was also built, which included collaborative control, data fusion, and model verification. Based on the integrated test bench and the development experiment management platform, an experimental platform was obtained. Through the experimental platform, simulation experiments were conducted on the above-mentioned UAV descent cable system architecture to carry out basic performance (static load, dynamic descent) and extreme working condition (thermal environment, strong wind) tests. The theoretical model was calibrated, the design was optimized (material ratio, control algorithm), and the indicators were extracted to form test specifications to facilitate the implementation of the technology.

[0043] This study focuses on the design and development of a multifunctional tethered fire-fighting drone descent system, conducting a systematic investigation from fundamental theoretical analysis to engineering practice verification. First, based on the specific needs of fire emergency rescue, the working principle of the descent system was analyzed in depth, with a focus on the drone-descent cable collaborative working mechanism, the kinematic characteristics of the self-locking mechanism, and the energy dissipation principle. Building upon this, 3D modeling software such as CATIA and SolidWorks were used to create detailed models of the drone platform and the descent cable structure. The descent cable structure includes core components such as a servo drive unit, a locking mechanism, a damping braking system, and a folding escape ladder. Through parametric modeling and virtual assembly technology, the collaborative working process of each functional module was visually demonstrated.

[0044] Regarding material selection, a rigorous material selection process was conducted for key components of the drone and the descent cable structure, taking into account the harsh environments of high temperatures and corrosion at fire scenes. Through comparative analysis of the mechanical properties of ultra-high molecular weight polyethylene (UHMWPE), aramid, and carbon fiber composites, UHMWPE-aramid blended composite cable was ultimately chosen as the main material for the descent cable, combined with nickel-based high-temperature alloys to manufacture the core transmission components. Theoretical calculations verified that this material combination maintains stable mechanical properties even at 800℃, fully meeting the stringent requirements of fire rescue operations.

[0045] To further verify the feasibility of the design scheme, the research team conducted systematic calculation and analysis: First, based on the theory of mechanics of materials, the cross-sectional dimensions and strength of the descent cable were designed and checked, and the safety factor under a rated load of 20kg was calculated to be 3.2; second, by establishing a multi-rigid-body dynamic model, the dynamic characteristics during the descent process were simulated and analyzed, and the electromagnetic-friction hybrid braking system was optimized, achieving a controllable descent speed range of [0.2, 1.0] m / s; finally, using the finite element analysis method, static and dynamic simulations of the overall structure were conducted through simulation experiments on the above-mentioned UAV descent cable system architecture, focusing on the stress distribution and deformation characteristics under a 150kg impact load. The analysis results show that the maximum equivalent stress of the system is 168.53MPa, the maximum deformation is only 0.47mm, and the equivalent elastic strain of key parts is within the allowable range, fully demonstrating the feasibility of the design scheme.

[0046] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A speed control method for firefighting drones based on vision detection, characterized in that, Includes the following steps: Step S1: Obtain environmental images of the fire scene based on the shooting points mounted on the UAV, analyze the environmental images to obtain all obstacles in the environmental images, the shooting points are used to capture images, and monitor the obstacle distance D of the obstacles, the obstacle distance is the distance between the obstacle and the UAV; The descent path of the UAV is determined based on the environmental image, and the attitude data of the UAV, including angular velocity and acceleration, is obtained. The current intensity of the electromagnetic braking of the UAV descent cable and the friction plate pressure of the mechanical damping are also obtained. Step S2: Set a safety threshold Dt. If D≥Dt, generate a start descent command based on the UAV attitude data, transmit the descent command to the UAV descent cable, and dynamically adjust the current intensity and friction plate pressure so that the descent speed of the UAV is in real time at [0.2, 1.0] m / s. If D < Dt, an emergency stop command is generated based on the UAV attitude data. The descent speed is adjusted to below 0.3 m / s based on the emergency stop command, and the UAV's position is adjusted to avoid the obstacle.

2. The speed control method for a firefighting drone based on vision detection according to claim 1, characterized in that, In step S1, the shooting point is based on an infrared thermal imaging and visible light dual-spectrum camera. The infrared thermal imaging is used to identify obstacles in a dense smoke environment, and the visible light dual-spectrum camera is used to identify obstacles in a clear environment. The environmental image includes a heat map obtained by infrared thermal imaging and a multispectral image obtained by visible light dual-spectrum camera.

3. The speed control method for a firefighting drone based on vision detection according to claim 1, characterized in that, In step S1, the process of analyzing the environmental image includes: The environmental image is divided into several pixels, the pixel value of each pixel is obtained, and a pixel value range threshold is set. If the pixel value belongs to the pixel value range threshold, the pixel is recorded as an obstacle pixel. The total number of obstacle pixels in the environmental image is obtained, and the area ratio of each pixel is obtained. Based on the total number and the area ratio, the area ratio of obstacles in the environmental image is obtained. If the area occupied by the obstacle exceeds 15%, the angular velocity of the drone is obtained, and the relative motion trend between the obstacle and the drone is predicted based on the angular velocity.

4. The speed control method for a firefighting drone based on vision detection according to claim 1, characterized in that, In step S2, the process of dynamically adjusting the current intensity and friction plate pressure includes: When the shooting point detects that the distance to the obstacle of the drone is greater than or equal to the safety threshold, the electromagnetic braking is based on the permanent magnet synchronous motor bearing 60% to 80% of the braking force, and the descent speed of the drone is controlled by adjusting the excitation current of the permanent magnet synchronous motor. When the shooting point detects that the distance to the obstacle of the drone is less than the safety threshold, the rated value of the friction plate pressure of the mechanical damping is obtained, and the friction plate pressure is increased to 120% of the rated value. The mechanical damping and electromagnetic braking simultaneously adjust the response time so that the response time is less than 100ms. The response time is the time interval between the trigger moment and the deceleration moment. The trigger moment is the moment when the drone generates the descent command, and the deceleration moment is the moment when the drone begins to decelerate.

5. The speed control method for a firefighting drone based on vision detection according to claim 1, characterized in that, In step S2, the safety threshold is set based on a preset load-threshold mapping table, which is a pre-stored data table used to store the correspondence between different load weights and safety thresholds.

6. The speed control method for a firefighting drone based on vision detection according to claim 1, characterized in that, In step S1, the process of acquiring environmental images of the fire scene by the shooting point includes acquiring the descent path of the drone, and the shooting point performing a panoramic scan of the fire scene below the drone every 200ms to generate a dynamic heat map containing the distribution of obstacles.

7. The speed control method for a firefighting drone based on vision detection according to claim 6, characterized in that, In step S1, the process of determining the descent path of the UAV based on the environmental image includes: The dynamic heat map is divided into several regions, the obstacle density in each region is obtained, and the horizontal position of the UAV is adjusted in real time so that the UAV is always directly above the region with the lowest obstacle density, thus obtaining the descent path.

8. The speed control method for a firefighting drone based on vision detection according to claim 1, characterized in that, In step S2, the drone also includes a manual intervention mechanism. When the drone is in the manual intervention mechanism, ground personnel directly set the descent speed of the drone and adjust the current intensity of the electromagnetic brake and the friction plate pressure of the mechanical damping according to the descent speed.

Images