A curtain wall cleaning unmanned aerial vehicle control system and method

Abstract

The application relates to the technical field of intelligent control, and particularly discloses a curtain wall cleaning unmanned aerial vehicle control system and method, which comprises an initialization unmanned aerial vehicle and an integrated remote control terminal, a double-mode low-delay communication is established, and a module state is self-checked; multi-dimensional data of a close state, a distance, wind load and equipment operation are collected through a pressure sensor, an ultrasonic measuring instrument, an air flow rate sensor and the like; based on a PID adaptive control algorithm, a wind load influence coefficient is combined to dynamically adjust the rotating speed of a propeller and negative pressure adsorption force, safe close pressure is maintained, a remote control moving instruction is received, higher-precision three-dimensional movement is realized in cooperation with close data, adsorption force is closed-loop adjusted during movement, adsorption abnormalities, communication interruption and low power states are monitored, and a hierarchical emergency response and safe shutdown are triggered. The application can solve the problems of traditional equipment, such as unfirm adsorption, complicated operation and poor high-altitude adaptability, improve operation stability, convenience and safety, and is suitable for complex high-altitude curtain wall cleaning scenes.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to a control system and method for a curtain wall cleaning drone. Background Technology

[0002] With the increasing number of high-rise buildings in cities, the demand for curtain wall cleaning operations is growing. Small curtain wall cleaning robots are gradually replacing traditional manual cleaning methods due to their flexibility. In existing technologies, the control schemes of small curtain wall cleaning equipment are mainly divided into two categories. One type adopts a fixed negative pressure adsorption mode, which maintains adhesion to the curtain wall by preset adsorption pressure, without considering the impact of high-altitude wind load changes on adsorption stability. The other type adopts a separate control logic, where adsorption adjustment, movement operation, and status monitoring rely on different controllers or operating terminals, requiring frequent manual switching.

[0003] However, the aforementioned existing technologies have some problems in practical applications. The adsorption control lacks adaptive capability, and the fixed adsorption pressure cannot dynamically offset the interference of high-altitude wind loads. This can easily lead to loosening of the adhesion due to poor adsorption, or damage to the curtain wall material due to excessive adsorption pressure. Moreover, the operation process is cumbersome, and switching between multiple controllers causes response delays, making it difficult to meet the real-time control requirements of high-altitude operations. In the existing technologies, the movement and adsorption control are independent of each other. During the movement process, there is a lack of real-time monitoring and closed-loop adjustment of the adhesion status, and the positioning accuracy is insufficient, which can easily lead to cleaning omissions or repeated operations.

[0004] Therefore, there is an urgent need for a control system and method for curtain wall cleaning drones to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a control method for a curtain wall cleaning drone, comprising the following steps: Initialize the curtain wall cleaning drone and integrated remote control terminal, establish a low-latency communication connection, and self-check the status of each module; Real-time collection of multi-dimensional operational data, including data on the fit between the robot and the curtain wall, spacing information, wind load parameters in the high-altitude environment, and equipment operating status data; Based on the multi-dimensional operation data, the propeller speed and negative pressure adsorption force are dynamically calculated and adjusted through an adaptive control algorithm to maintain the robot's stable adhesion to the curtain wall. It receives movement commands from the integrated remote control terminal, combines the adhesion status data with the spacing information, and coordinates the control of the drive wheel to achieve three-dimensional movement, while maintaining closed-loop adjustment of the adsorption force during the movement. The system monitors abnormal conditions and triggers an emergency procedure to shut down the system when it detects that the adsorption force is below the safety threshold, communication is interrupted, or the remaining battery power is too low.

[0006] Furthermore, the present invention also discloses a control system for a curtain wall cleaning drone, comprising: The initialization module is used to initialize the curtain wall cleaning drone and the integrated remote control terminal, establish a low-latency communication connection, and self-check the status of each module. The data acquisition module is used to collect multi-dimensional operational data in real time, including data on the fit between the robot and the curtain wall, spacing information, wind load parameters in the high-altitude environment, and equipment operating status data. The calculation module is used to dynamically calculate and adjust the propeller speed and negative pressure adsorption force based on the multi-dimensional operation data through an adaptive control algorithm to maintain the robot's stable adhesion to the curtain wall. The control module is used to receive movement commands from the integrated remote control terminal, combine the adhesion status data and the spacing information, and coordinate the control of the drive wheel to achieve three-dimensional movement, while maintaining closed-loop adjustment of the adsorption force during the movement. The monitoring module is used to monitor abnormal system conditions. When it detects that the adsorption force is lower than the safety threshold, communication is interrupted, or the remaining battery power is too low, it triggers the emergency handling procedure and performs a safe shutdown.

[0007] Furthermore, the control module includes: The conversion unit is used to parse the movement commands from the integrated remote control terminal, which include directional control signals in three dimensions: up and down, left and right, and forward and backward, and convert them into target speed and steering commands for the drive wheels; The judgment unit is used to determine the bonding state based on the real-time distance. If the real-time distance exceeds the preset maximum safe distance, it is determined that the bonding is loose and the adsorption force enhancement command is triggered first. The calculation unit is used to combine the movement direction command and real-time spacing information to calculate and output the control signal of the drive wheel motor, and at the same time realize closed-loop control of speed and position through motor encoder feedback. The adjustment unit is used to continuously monitor the change of adsorption pressure value during the movement execution phase. If the pressure fluctuation exceeds the set stable threshold, the movement process is paused and the adsorption control algorithm is invoked to readjust the adsorption pressure value to the stable range. The sending unit is used to maintain the current adsorption force output after a single movement command is executed, wait for the next command, and send the robot's real-time position, movement status and adsorption pressure value to the display screen of the integrated remote control terminal for visual display.

[0008] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described curtain wall cleaning drone control method.

[0009] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described curtain wall cleaning drone control method.

[0010] The beneficial effects of this application are as follows: Firstly, this application can improve the adhesion stability of high-altitude operations. By collecting wind load data through the airflow rate sensor and air pressure sensor on the fuselage, and combining the adhesion status data of the pressure sensor and ultrasonic measuring instrument, the propeller speed is dynamically adjusted using a PID adaptive control algorithm to compensate for the fluctuation of adsorption force caused by wind load interference and displacement in real time, so that the adsorption pressure is stabilized within a safe range, effectively solving the problem that the traditional fixed adsorption mode cannot cope with high-altitude wind load and the adhesion is not firm.

[0011] Secondly, this application simplifies the operation process and achieves integrated and precise control. By constructing a low-latency connection through dual-mode communication, it integrates functions such as adsorption start / stop, movement adjustment, speed fine-tuning, and emergency stop into a single remote control terminal, eliminating the cumbersome operation of switching between multiple controllers and achieving closed-loop interaction between command issuance and status feedback. This solves the pain points of traditional separate control, such as response delay and complex operation.

[0012] Thirdly, this application can improve the accuracy of mobile positioning and work efficiency. By coordinating the movement control and adsorption control, and combining the speed and position feedback of the motor encoder, the positioning accuracy of three-dimensional movement can be achieved. At the same time, the adsorption pressure fluctuation is continuously monitored during the movement. By pausing the movement, adjusting the adsorption, and resuming the operation, the loosening of the adhesion caused by the movement is avoided, thus solving the problems of large positioning deviation and low cleaning efficiency of traditional equipment. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of a method flow proposed in an embodiment of this application.

[0014] Figure 2 This is a schematic diagram of the system structure proposed in an embodiment of the present invention.

[0015] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0017] like Figure 1 As shown, this application provides a control method for a curtain wall cleaning drone, including the following steps: S1, initialize the curtain wall cleaning drone and integrated remote control terminal, establish a low-latency communication connection, and self-check the status of each module; S2 collects multi-dimensional operation data in real time, including data on the fit between the robot and the curtain wall, spacing information, wind load parameters in the high-altitude environment, and equipment operation status data. S3, based on the multi-dimensional operation data, the propeller speed and negative pressure adsorption force are dynamically calculated and adjusted through an adaptive control algorithm to maintain the robot's stable adhesion to the curtain wall; S4, receive the movement command from the integrated remote control terminal, combine the adhesion status data and the spacing information, coordinate the control of the drive wheel to achieve three-dimensional movement, and maintain closed-loop adjustment of the adsorption force during the movement; S5 monitors abnormal system conditions. When it detects that the adsorption force is below the safety threshold, communication is interrupted, or the remaining battery power is too low, it triggers an emergency handling procedure and performs a safe shutdown.

[0018] As described in steps S1-S5 above, this invention achieves stable adhesion, precise movement, and safe operation of the curtain wall cleaning drone in high-altitude curtain wall operation scenarios by initializing the communication connection between the curtain wall cleaning drone and the integrated remote control terminal, collecting multi-dimensional operation data, adjusting the propeller speed and negative pressure adsorption force based on the adaptive control algorithm, coordinating the control of the drive wheel to achieve three-dimensional precise movement and maintaining closed-loop adjustment of adsorption force during movement, monitoring abnormal states and triggering emergency handling. At the same time, it simplifies the operation process and improves the adaptability and reliability of the operation.

[0019] Curtain wall cleaning drones need to operate in high-altitude environments, where there are uncertain wind loads and diverse curtain wall surface materials, posing challenges to the stability of the drone's adhesion to the curtain wall. At the same time, high-altitude operations have extremely high requirements for ease of operation and safety redundancy. Traditional small window cleaning machines generally have problems such as a single adsorption method, inability to dynamically respond to changes in wind load, the need to switch between multiple controllers during operation, and an imperfect emergency handling mechanism. These problems can easily lead to weak adhesion, large positioning deviations, and high operational risks. Therefore, it is necessary to develop a control method that balances stable adhesion, precise control, and safety assurance to solve the core technical pain points in high-altitude operations.

[0020] Traditional solutions often employ fixed adsorption pressure or single-dimensional data acquisition control modes, failing to dynamically adjust adsorption force based on wind load changes and equipment operating status. Furthermore, the independent operation of movement and adsorption control can easily lead to loosening of the seal during movement. Remote control is also fragmented, resulting in high response delays, and the lack of a systematic emergency response design makes it difficult to handle unexpected anomalies in high-altitude operations. This invention addresses the limitations of traditional methods through a comprehensive solution that integrates multi-dimensional data fusion acquisition, adaptive control algorithms linking adsorption and movement control, and unified remote control with all-scenario emergency response. This achieves a synergistic improvement in adsorption stability, operational convenience, and operational safety.

[0021] The core of step S1 is to establish a reliable communication foundation and equipment status assurance for the entire operation process. Specifically, this is achieved by activating the curtain wall cleaning drone and the integrated remote control terminal, using 2.4G and 5G dual-mode communication technology to establish a low-latency connection, ensuring that the transmission delay does not exceed 20ms. At the same time, the drone automatically performs status self-checks on key modules such as the adsorption components, sensors, drive wheels, and power modules. Only when all modules are detected to be normal can the system enter the standby ready state. This step provides a stable hardware and communication prerequisite for subsequent data acquisition, control adjustment, and emergency handling, avoiding operational abnormalities caused by initial equipment failures or unstable communication.

[0022] The technical solution of this application logically follows a complete closed-loop operation encompassing preparation, perception, adjustment, execution, and support. In principle, it is based on multi-dimensional data fusion and adaptive control technology to achieve stable, precise, and safe high-altitude operations. First, an initialization process establishes a 2.4G / 5G dual-mode low-latency communication connection between the curtain wall cleaning drone and the integrated remote control terminal. Simultaneously, self-tests are performed on key modules such as the adsorption components, sensors, and power module, laying a reliable hardware and communication foundation for subsequent operations. Then, multiple types of sensors collect real-time data on adhesion status, spacing information, high-altitude wind load parameters, and equipment operating status, comprehensively capturing the dynamic changes in the operating environment and the equipment itself, providing a comprehensive basis for control decisions. Next, based on this multi-dimensional data, a PID adaptive control algorithm dynamically adjusts the propeller speed (adjustment range 500-3000 rpm) and negative pressure adsorption force. Closed-loop control counteracts high-altitude wind load interference, maintaining a safe adsorption pressure of 1.5-2.5 kPa. The system ensures stable adhesion between the drone and the curtain wall within a defined force range. Based on stable adhesion, it receives three-dimensional movement commands from the integrated remote control terminal. Combining adhesion status data and spacing information, it coordinates the control of the tracked drive wheels to achieve precise movement with an accuracy of ±1 cm. During movement, it continuously adjusts the adhesion force in a closed loop to prevent loosening. Finally, by continuously monitoring the adhesion pressure, communication signal strength, and remaining battery power, it triggers a graded emergency response for key abnormal scenarios such as adhesion anomalies, communication interruptions, and low battery levels, ultimately executing an orderly and safe shutdown. This forms a comprehensive safety assurance mechanism. The overall solution addresses the technical pain points of traditional small window cleaning robots, such as weak adhesion, cumbersome operation, and poor adaptability to high altitudes, through a logical chain of data acquisition, algorithm adjustment, collaborative control, and emergency backup.

[0023] In one embodiment, the step of real-time acquisition of multi-dimensional operation data includes: S21, through pressure sensors and ultrasonic meters arranged at the bottom of the robot, the adsorption pressure value and real-time distance between the robot and the curtain wall are obtained in real time, respectively. The adsorption pressure value acquisition accuracy is ±0.1kPa, and the real-time distance acquisition range is 0 to 5 cm. S22 uses airflow rate sensors and air pressure sensors installed on the robot body to collect wind speed and air pressure data in the high-altitude operation area in real time, with a sampling frequency of no less than 10Hz, and identifies strong wind interference scenarios based on wind speed changes. S23, real-time acquisition of propeller speed, drive motor load current, battery pack remaining power and integrated remote control terminal wireless signal strength. S24 integrates the collected adsorption pressure value, real-time spacing, wind speed, air pressure, propeller speed, motor load current, battery remaining power and wireless signal strength data into a unified format multi-dimensional status data packet, and sends it in real time to the integrated remote control terminal and the robot's onboard control unit through a 2.4G and 5G dual-mode communication link.

[0024] As described in steps S21-S24 above, by deploying multiple types of sensors, setting precise acquisition parameters, integrating multi-dimensional data, and adopting dual-mode communication transmission, comprehensive, real-time, and high-precision acquisition of data on adhesion status, high-altitude environment, and equipment operation during the operation of curtain wall cleaning drones is achieved. This provides reliable data support for subsequent adsorption force adjustment, three-dimensional movement control, and emergency handling, ensuring the stability and safety of the operation.

[0025] From the perspective of the operational characteristics of high-altitude curtain wall cleaning, drones need to maintain stable contact with the curtain wall in complex high-altitude environments. This contact is affected by multiple factors, including the operation of their own adsorption system and changes in high-altitude wind load. Simultaneously, the equipment's operational status directly impacts operational safety, and a lack of comprehensive and accurate data acquisition can lead to blind control decisions. Traditional data acquisition methods have significant limitations, typically collecting only single-dimensional data that fails to cover key aspects such as contact, environment, and equipment operation. They either suffer from low acquisition accuracy and insufficient sampling frequency, making it difficult to capture real-time changes, and their data transmission methods are limited, prone to delays or interruptions, resulting in lagging subsequent control adjustments and an inability to respond promptly to complex high-altitude conditions. This step addresses these issues by collecting key data in modules, defining acquisition accuracy and frequency standards, standardizing data formats, and employing dual-mode communication transmission to form a complete data acquisition chain. This specifically solves the problems of incomplete data, insufficient accuracy, and unreliable transmission associated with traditional methods.

[0026] Step S21 focuses on acquiring core data regarding the bonding state. Pressure sensors and ultrasonic transducers deployed on the bottom of the robot acquire the adsorption pressure and real-time distance, respectively. The pressure sensor has an accuracy of ±0.1 kPa, accurately sensing minute changes in adsorption force. The ultrasonic transducer covers a range of 0 to 5 cm, reflecting the real-time distance between the drone and the curtain wall. The combination of these two methods provides a quantitative basis for determining the stability of the bonding, preventing inaccurate adsorption adjustments due to unclear perception of the bonding state.

[0027] Step S22 addresses interference factors in the high-altitude environment by installing airflow rate sensors and air pressure sensors on the robot body to collect wind speed and air pressure data in real time. The sampling frequency is no less than 10Hz to ensure rapid capture of wind load changes. By comparing the magnitude of wind speed changes with preset thresholds, strong wind interference scenarios are accurately identified, providing environmental data support for subsequent adaptive enhancement of adsorption force and solving the problem of unstable adhesion caused by sudden changes in high-altitude wind load.

[0028] Step S23 uses the airborne status monitoring unit to comprehensively collect key operating parameters of the equipment, including real-time propeller speed, drive motor load current, remaining battery power, and wireless signal strength of the integrated remote control terminal. These parameters directly reflect the operating status and energy supply of the core components of the equipment, ensuring that the system can monitor in real time whether there are risks such as overload, low power, or communication abnormalities, providing a data foundation for equipment fault early warning and emergency handling.

[0029] Step S24 integrates all collected multi-dimensional data into a unified format status data packet, which is transmitted using a 2.4G and 5G dual-mode communication link. This ensures communication stability during close-range operations and meets the low-latency requirements for long-range operations. The data is sent to the integrated remote control terminal and the airborne control unit in real time to achieve data sharing. This ensures that both remote control operations and airborne autonomous control can make decisions based on complete data, avoiding control coordination failures caused by untimely data transmission or inconsistent formats.

[0030] In one embodiment, the step of dynamically calculating and adjusting the propeller speed and negative pressure adsorption force through an adaptive control algorithm includes: S31. The real-time collected adsorption pressure value is compared with the preset adsorption pressure safety range. If the adsorption pressure is lower than the lower limit of the safety range, it is determined that the adsorption force is insufficient and the propeller speed needs to be increased. If the adsorption pressure is higher than the upper limit of the safety range, it is determined that the adsorption force is overloaded and the propeller speed needs to be reduced. S32, based on the real-time collected wind speed data, calculate the degree of influence of wind load on the adsorption system. Specifically, based on the ratio of real-time wind speed to preset reference wind speed, and combined with the preset wind load adjustment coefficient, calculate the wind load influence coefficient, which increases with the increase of wind speed. S33, based on the deviation between the measured adsorption pressure and the target value, and the wind load influence coefficient, the propeller speed adjustment is calculated using proportional, integral, and derivative adjustment algorithms. The calculation process integrates the current pressure deviation, the historical cumulative deviation, the deviation change trend, and the compensation amount for the wind load influence. S34, Based on the speed adjustment amount, dynamically adjust the actual speed of the propeller within the range of 500 to 3000 rpm so that the adsorption pressure is stabilized within the preset safety range; S35 predicts in real time the fluctuations in adsorption force that may be caused by displacement before and during the robot's movement, and pre-adjusts the propeller speed in advance using a feedforward control method to counteract the fluctuations.

[0031] As described in steps S31-S35 above, through a series of steps including pressure threshold judgment, wind load influence quantification, proportional-integral-derivative adjustment algorithm to calculate speed adjustment, dynamic adjustment of propeller speed and feedforward control to predict fluctuations, the propeller speed and negative pressure adsorption force are precisely and adaptively adjusted to ensure that the curtain wall cleaning drone maintains stable contact with the curtain wall under complex working conditions such as high-altitude wind load interference and mobile operation.

[0032] From the perspective of the physical environment of high-altitude operations, curtain wall cleaning drones need to cope with uncertain wind load interference. Wind load directly weakens the negative pressure adsorption force, leading to loosening of the adhesion. At the same time, changes in the relative position between the drone and the curtain wall during drone movement also cause fluctuations in adsorption force. Different curtain wall materials have different tolerances to adsorption pressure; excessively high pressure may damage the curtain wall, while insufficient pressure cannot guarantee adhesion stability. Traditional adjustment methods often use fixed propeller speeds or simple adjustments based on single pressure data, which cannot dynamically respond to changes in wind load and fluctuations in adsorption force during movement. The adjustment accuracy is low, and the response is lag, making it difficult to meet the stable adhesion requirements of high-altitude operations. This step addresses the problems of adjustment lag and weak anti-interference ability of traditional methods by combining multi-dimensional data fusion, proportional-integral-derivative adjustment algorithms, and feedforward control, achieving dynamic and precise control of adsorption force.

[0033] Step S31 uses adsorption pressure as the core judgment criterion, comparing the real-time adsorption pressure value collected by the pressure sensor with the preset safe range of 1.5 to 2.5 kPa to determine whether the adsorption state is insufficient or overloaded, providing a direct judgment standard for subsequent speed adjustment. When the adsorption pressure is below 1.5 kPa, it indicates that the UAV is at risk of detaching from the curtain wall, and the negative pressure needs to be increased by increasing the propeller speed. When the pressure is above 2.5 kPa, excessive adsorption may damage the curtain wall surface, and the speed needs to be reduced to decrease the negative pressure. This step provides a clear trigger condition for the adjustment action.

[0034] Step S32 addresses the key interference factor of high-altitude wind load by quantifying its impact on the adsorption system. Real-time wind speed data is obtained from information collected by the airflow rate sensor on the fuselage. By calculating the ratio of the real-time wind speed to the preset reference wind speed and then multiplying it by the preset wind load adjustment coefficient, the wind load influence coefficient is obtained. This coefficient increases linearly with increasing wind speed, accurately reflecting the degree to which wind load weakens the adsorption force. This provides a quantitative basis for wind load compensation in subsequent adjustments, avoiding adsorption instability caused by sudden changes in wind load.

[0035] Step S33 uses a proportional-integral-derivative (PID) control algorithm to calculate the propeller speed adjustment. The algorithm inputs include the deviation between the measured adsorption pressure and the target value, the historical cumulative deviation, the deviation trend, and the compensation amount corresponding to the wind load influence coefficient. The pressure deviation is derived from the comparison results in step S31, and the wind load compensation amount is calculated from the wind load influence coefficient in step S32. The algorithm uses a proportional term to quickly respond to the current deviation, an integral term to eliminate accumulated errors, and a derivative term to predict the deviation trend. Simultaneously incorporating the wind load compensation amount ensures the accuracy and timeliness of the adjustment calculation, avoiding the problem of insufficient response to wind load disturbances in simple PID control.

[0036] Step S34 dynamically adjusts the actual propeller speed within a preset range of 500 to 3000 rpm based on the calculated speed adjustment. Changes in propeller speed directly alter the magnitude of the negative pressure adsorption force. When the speed increases, the negative pressure increases, and the adsorption pressure rises; when the speed decreases, the negative pressure decreases, and the adsorption pressure drops. This closed-loop adjustment stabilizes the adsorption pressure within a safe range of 1.5 to 2.5 kPa, balancing adhesion stability with curtain wall protection.

[0037] Step S35 uses feedforward control to proactively address adsorption force fluctuations during movement. Based on the real-time distance change trend collected by the ultrasonic measuring instrument, it predicts the amplitude of adsorption force fluctuations caused by displacement before and during the UAV's movement, and fine-tunes the propeller speed in advance to counteract the fluctuations. The amount of fine-tuning the propeller speed in advance is called the feedforward control speed fine-tuning amount, and the calculation formula for the feedforward control speed fine-tuning amount is: ; Among them, the This indicates the amount of feedforward speed adjustment. This represents the feedforward coefficient (dimensionless, can be 0.5, its value is determined experimentally, and is used to quantify the influence weight of displacement on adsorption force). Indicates the real-time change in spacing. Indicates the change over time. This indicates the current propeller reference speed. This step compensates for the lag in feedback control, ensuring that the adsorption force remains stable during movement and preventing loosening of the bond due to movement.

[0038] In one embodiment, the step of receiving a movement command from an integrated remote control terminal, combining the adhesion state data and the spacing information, and collaboratively controlling the drive wheel to achieve precise three-dimensional movement while maintaining closed-loop adjustment of the adsorption force during movement includes: S41, parse the movement command from the integrated remote control terminal, the command includes directional control signals in three dimensions: up and down, left and right, and forward and backward, and convert them into the target speed and steering command of the drive wheel; S42, based on the real-time distance fed back by the ultrasonic measuring instrument, determine the bonding status. If the real-time distance exceeds the preset maximum safe distance, it is determined that the bonding is loose and the adsorption force enhancement command is triggered first. S43 combines the movement direction command and real-time spacing information to calculate and output the control signal of the drive wheel motor. At the same time, it realizes closed-loop control of speed and position through motor encoder feedback, so that the robot's movement and positioning accuracy on the curtain wall surface reaches ±1 cm. S44, during the moving execution phase, continuously monitor the change in adsorption pressure value. If the pressure fluctuation exceeds the set stable threshold, pause the moving process and call the adsorption control algorithm to readjust the adsorption pressure value to the stable range. S45: After a single movement command is executed, the robot maintains the current adsorption force output, waits for the next command, and sends the robot's real-time position, movement status, and adsorption pressure value to the display screen of the integrated remote control terminal for visual display.

[0039] As described in steps S41-S45 above, by parsing the movement commands of the integrated remote control terminal, combining the adhesion status data and real-time spacing information, controlling the speed and position of the drive wheel in a closed loop, monitoring and adjusting the adsorption force in real time during the movement, and providing feedback on the operation status, the curtain wall cleaning drone can achieve three-dimensional precise movement on the curtain wall, while ensuring stable adhesion during the movement, taking into account both operation efficiency and operational safety.

[0040] From the perspective of high-altitude curtain wall cleaning operations, drones need to maintain stable contact with the curtain wall while performing three-dimensional movement (up / down, left / right, forward / backward) to achieve comprehensive cleaning coverage. During movement, drone displacement may cause changes in the distance between the drone and the curtain wall, leading to fluctuations in adsorption force. If movement control and adsorption control are independent, there is a risk of loosening or even detachment. Furthermore, high-altitude operations require high precision in movement and positioning; otherwise, cleaning omissions or repetitive work may occur. Traditional movement control methods often employ open-loop control or single-dimensional position adjustment, without linkage with the adsorption state, resulting in low positioning accuracy and an inability to respond promptly to adsorption force fluctuations during movement. This cumbersome and risky process necessitates frequent switching of control modes. This step addresses the problems of insufficient coordination between movement and adsorption, low positioning accuracy, and complex operation associated with traditional methods through a comprehensive solution involving instruction parsing and adhesion data fusion, closed-loop control of the drive wheels, real-time linkage adjustment of adsorption force, and visual feedback of status.

[0041] Step S41 forms the basis of the motion control command. The integrated remote control terminal sends motion commands through a 2.4G and 5G dual-mode communication link. The commands include directional control signals in three dimensions: up and down, left and right, and forward and backward. After receiving the commands, the airborne control unit parses them and converts the directional signals into target speed and steering commands for the drive wheels. The drive wheels adopt a DC geared motor and track structure to adapt to the movement requirements of curtain walls made of different materials such as glass and stone. This step provides a clear execution basis for the precise movement of the UAV, ensuring the accuracy and timeliness of command transmission.

[0042] Step S42 focuses on verifying the adhesion status before movement. The real-time distance data comes from an ultrasonic sensor located on the bottom of the robot. This module has a data acquisition range of 0 to 5 centimeters and can accurately reflect the distance relationship between the drone and the curtain wall. The onboard control unit compares the real-time distance collected in real time with the preset maximum safe distance. If the distance exceeds this threshold, it indicates that the adhesion is loose. At this time, the adsorption force enhancement command is triggered first, and the adaptive control algorithm in step S3 is called to increase the propeller speed. The movement command is executed only after the real-time distance returns to the safe range and the adhesion is stable, so as to avoid detachment during movement due to weak adhesion.

[0043] Step S43 achieves precise control of the drive wheels. Combining the analyzed movement direction command with the real-time spacing information fed back by the ultrasonic measuring instrument, the onboard control unit calculates the control signal for the drive wheel motor. Simultaneously, the actual rotational speed and position data of the drive wheels are collected in real time through the motor encoder, forming a closed-loop control. By continuously comparing the actual state with the target state, the control signal is dynamically adjusted to ensure that the UAV's movement and positioning accuracy reaches ±1 cm, meeting the position requirements of fine cleaning operations. At the same time, the tracked drive wheel design improves the UAV's adaptability to movement on different material curtain wall surfaces, enabling it to overcome obstacles up to 15 mm in diameter.

[0044] Step S44 ensures the stability of the adhesion during movement. The adsorption pressure value comes from a pressure sensor located on the bottom of the robot, with an accuracy of ±0.1 kPa, which can capture subtle changes in the adsorption force in real time. During the movement execution phase, the onboard control unit continuously monitors the adsorption pressure value. If the pressure fluctuation exceeds the set stability threshold, it indicates that the movement has caused a change in the adhesion state. At this time, the movement process is immediately paused, and the adsorption control algorithm is invoked to readjust the propeller speed to stabilize the adsorption pressure value within the safe range of 1.5 to 2.5 kPa. Movement continues only after the adhesion state has stabilized, achieving dynamic coordination between movement and adsorption.

[0045] Step S45 completes the feedback loop of the operational status. After a single movement command is executed, the onboard control unit maintains the current adsorption force output to ensure the drone remains in a stable, attached state, awaiting the next command from the integrated remote control terminal. Simultaneously, through a 2.4G and 5G dual-mode communication link, the robot's real-time position, movement status, and adsorption pressure value are sent to the OLED display screen of the integrated remote control terminal, enabling a visual display of the operational status. This allows operators to monitor the equipment's operation in real time, providing a basis for subsequent operational decisions and simplifying the operational process.

[0046] In one embodiment, when the monitoring system detects an abnormal state, such as adsorption force falling below a safety threshold, communication interruption, or low battery power, the steps to trigger an emergency handling procedure and perform a safe shutdown include: S51: Continuously monitor the adsorption pressure value. If the pressure value is lower than the preset safety pressure threshold and the duration exceeds the first set time, it is determined to be an adsorption abnormality. Control the integrated remote control terminal to issue an audible and visual alarm and automatically increase the propeller speed by a fixed ratio. S52 continuously monitors the wireless communication signal strength between the integrated remote control terminal and the robot. If the signal strength is lower than the preset strength threshold and the duration exceeds the second set duration, it is determined that the communication is interrupted, and the drone is controlled to enter the hovering emergency mode to maintain the current adsorption state and position. S53 continuously monitors the remaining battery power. When the remaining power is lower than the preset low power threshold, it triggers a low power alarm and displays a clear low power message on the remote control display. S54, if any of the following conditions—adsorption abnormality, communication interruption, or low battery state—cannot be restored to normal after taking corresponding measures, a safety shutdown procedure is initiated. The procedure includes gradually reducing the propeller speed to a stop, releasing the adsorption pressure, and allowing the robot to smoothly detach from the curtain wall surface. S55: After all emergency events have been handled, the system automatically records the type of the event, the time of occurrence, and the handling measures taken, forming an operation log. This log can be exported through the integrated remote control terminal for analysis.

[0047] As described in steps S51-S55 above, by continuously monitoring three key status parameters—adsorption pressure, communication signal strength, and remaining battery power—a tiered emergency response is implemented for three core abnormal scenarios: adsorption anomaly, communication interruption, and low battery power. Ultimately, through orderly and safe shutdown and fault log recording, the entire process of high-altitude operation of curtain wall cleaning drones is made safe, ensuring equipment operation safety and controllable operational risks.

[0048] From the perspective of safety requirements for high-altitude operations, insufficient suction power can cause curtain wall cleaning drones to detach from the curtain wall and fall when operating at high altitudes. Communication interruptions can lead to loss of control, and low battery power can cause mid-operation shutdowns. All three types of anomalies can potentially cause serious safety accidents. Therefore, it is essential to establish a comprehensive emergency response mechanism covering monitoring, response, handling, and traceability. Traditional small window cleaning drones have significant shortcomings in emergency handling, often relying on simple methods such as single alarms or direct shutdowns, lacking targeted compensation measures. For example, direct shutdowns due to insufficient suction power can easily lead to equipment falls; there is no mechanism to maintain suction and hover after communication interruptions; low battery warnings lack clear prompts and buffer time, and there are no complete fault records, making it difficult to trace the root cause of the problem. This step addresses the problems of inadequate traditional emergency mechanisms, insufficient safety redundancy, and poor traceability through a systematic solution of precise monitoring, tiered response, and orderly shutdown log traceability, thereby improving the safety and reliability of high-altitude operations.

[0049] Step S51 is an emergency response to adsorption anomalies. The adsorption pressure value comes from a pressure sensor located at the bottom of the robot. This sensor has an accuracy of ±0.1 kPa and can accurately capture pressure changes. The preset safe pressure threshold is 1.2 kPa, and the first set duration is 2 seconds. When the pressure sensor detects an adsorption pressure value below 1.2 kPa for more than 2 seconds, it is determined to be an adsorption anomaly. At this time, the integrated remote control terminal immediately issues an audible and visual alarm, and the onboard control unit automatically increases the propeller speed by 30% to attempt to restore stable adsorption by increasing the negative pressure. This step performs compensation measures before triggering subsequent shutdown to avoid the risk of equipment falling due to direct shutdown and improve the safety redundancy of emergency response. The formula for calculating the propeller speed after the emergency state increase is as follows: ; Among them, the This indicates the propeller speed after the emergency boost. This indicates the real-time rotational speed of the propeller before the failure. This indicates a fixed enhancement ratio (dimensionless, with a possible value of 0.3; this value can be verified experimentally to ensure rapid enhancement of adsorption force without damaging the curtain wall).

[0050] Step S52 addresses the communication interruption scenario. The wireless communication signal strength data comes from the onboard status monitoring unit, which collects the signal strength between the integrated remote control terminal and the robot in real time. The preset strength threshold is -85dBm, and the second preset duration is 3 seconds. When the signal strength is below -85dBm and the duration exceeds 3 seconds, it is determined to be a communication interruption. At this time, the robot automatically enters the hovering emergency mode, maintaining its current adsorption state and position to prevent the equipment from detaching from the curtain wall due to loss of control. At the same time, it provides the operator with a time window to restore communication, avoiding an overreaction that triggers a shutdown due to a brief communication interruption.

[0051] Step S53 focuses on low battery warning and alerts. The remaining battery power data comes from the onboard status monitoring unit, and the preset low battery threshold is 20%. When the remaining battery power is detected to be below 20%, the system immediately triggers a low battery alarm. The OLED display of the integrated remote control terminal simultaneously displays clear low battery information, intuitively informing the operator of the equipment's energy status and reminding them to end the operation or return to base in time to avoid shutdown or loss of control due to depleted battery power.

[0052] Step S54 enables an orderly and safe shutdown. If, due to an adsorption anomaly, increasing the rotation speed by 30% fails to restore the adsorption pressure to a safe range, communication is interrupted and fails to resume within a preset time, or the operation is not terminated promptly due to low battery, the safety shutdown procedure is initiated. This procedure gradually reduces the propeller speed to a stop via the onboard control unit, avoiding sudden drops in adsorption force caused by a sudden decrease in speed. Simultaneously, it slowly releases the adsorption pressure, allowing the robot to smoothly detach from the curtain wall surface, eliminating the risk of impact or fall caused by a sudden shutdown.

[0053] Step S55 completes fault tracing and data retention. After all emergency events are handled, the system automatically records the event type (adsorption anomaly, communication interruption, or low battery), the time of occurrence (accurate to the second), and the handling measures taken (such as speed increase ratio, hovering duration, and downtime), forming a standardized operation log. This log is stored in the onboard non-volatile memory and can be exported via the integrated remote control terminal, providing data support for subsequent fault investigation and emergency strategy optimization, and improving the system's maintainability and problem tracing capabilities.

[0054] In one embodiment, extended functionality is also included, with specific steps including: S61 acquires images of the curtain wall surface through its onboard vision sensor, identifies the material type of the curtain wall based on image features, and automatically matches and calls the preset adsorption pressure parameters of the material according to the identification results. S62 continuously records the robot's movement trajectory, the range of the cleaned area, and the change of adsorption pressure over time during operation, and stores the data in the onboard non-volatile memory. S63 assesses the system's operating noise level based on the propeller's real-time speed and the motor's load current. If the noise level exceeds the preset comfort threshold, it automatically adjusts the speed control strategy or activates a noise reduction structure to reduce noise. The S64 supports remote updates to the robot's control program and remote adjustments to control parameters via wireless communication networks to adapt to different high-altitude curtain wall cleaning operation scenarios.

[0055] As described in steps S61-S64 above, through a series of extended steps including curtain wall material identification and adaptive matching of adsorption parameters, full recording of operation data, dynamic control of operating noise, and remote updating and debugging of control programs and parameters, the scene adaptability, operation traceability, and maintenance convenience of curtain wall cleaning drones are improved, enabling the system to flexibly cope with different curtain wall materials, complex operating environments, and diverse maintenance needs.

[0056] From a practical application perspective, curtain wall materials vary, such as glass and stone. Different materials have different tolerances to adsorption pressure. Blindly applying uniform adsorption parameters can easily lead to material damage or weak adsorption. After high-altitude operations are completed, it is necessary to trace the operation trajectory and equipment operating status to troubleshoot problems and assess the quality of the work. Excessive system noise may affect the surrounding environment, and different operation scenarios have different requirements for equipment parameters. Traditional equipment lacks a targeted adaptation mechanism, requiring on-site disassembly and adjustment, which is cumbersome and inefficient. Traditional solutions do not consider the impact of material differences on adsorption parameters, lack operation data recording functions, lack dynamic adjustment methods for noise control, and require on-site operation for parameter updates, making it difficult to meet diverse and refined operation and maintenance needs. This step addresses the problems of traditional methods, such as limited scenario adaptation, poor traceability, noise pollution, and cumbersome maintenance, by adding four extended functions: material identification and adaptation, data recording, noise control, and remote updates. This expands the application scope and practical value of the system.

[0057] Step S61 achieves precise matching between the curtain wall material and the adsorption parameters. An onboard vision sensor acquires images of the curtain wall surface, extracts texture and hardness features, and compares these features against a pre-set material feature database to identify the curtain wall material type. The onboard control unit pre-stores adsorption pressure parameter ranges for different materials: 1.5 to 2.0 kPa for glass and 2.0 to 2.5 kPa for stone. After identification, the matching adsorption pressure parameters are automatically retrieved, replacing manual adjustment. This avoids damage to the curtain wall or loosening of the adsorption due to improper adsorption pressure and improves operational efficiency.

[0058] Step S62 completes the recording and storage of all work data. During the operation, the onboard positioning module collects the robot's position information in real time, forming movement trajectory data. Combined with the running status of the drive wheels, it records the range of the cleaned area. Simultaneously, it records the change in adsorption pressure value collected by the pressure sensor over time. All this data is stored in the onboard non-volatile memory, ensuring that the data will not be lost due to power failure. This provides complete data support for subsequent work quality verification and equipment fault diagnosis, improving the traceability of the operation.

[0059] Step S63 achieves dynamic control of operating noise. The real-time rotational speed of the propeller is obtained from the data collected by the airborne condition monitoring unit, and the load current of the drive motor is obtained through the motor controller. Based on the correspondence between rotational speed and load current, and combined with the preset noise level evaluation model, the current system operating noise level is calculated. The preset comfort threshold is 60 decibels. When the evaluated noise level exceeds this threshold, the airborne control unit automatically adjusts the propeller speed control strategy, reducing the speed while ensuring adsorption stability, or activating the propeller duct noise reduction structure to reduce noise by optimizing the airflow path and reducing the impact on the surrounding environment.

[0060] Step S64 improves the ease of operation and maintenance of the system. Through the 2.4G and 5G dual-mode communication links, operators can remotely access the robot's onboard control unit via the wireless communication network to update the control program online without on-site disassembly. At the same time, core control parameters such as the adsorption pressure safety range and speed adjustment range can be remotely debugged, quickly adapting to high-altitude curtain wall cleaning operation scenarios of different heights and environments, reducing on-site operation and maintenance costs, and improving the system's flexibility and adaptability.

[0061] like Figure 2 As shown, the present invention also discloses a control system for a curtain wall cleaning drone, comprising: Initialization module 1 is used to initialize the curtain wall cleaning drone and the integrated remote control terminal, establish a low-latency communication connection, and self-check the status of each module; The data acquisition module 2 is used to collect multi-dimensional operation data in real time, including the bonding status data between the robot and the curtain wall, spacing information, high-altitude environmental wind load parameters, and equipment operation status data. The calculation module 3 is used to dynamically calculate and adjust the propeller speed and negative pressure adsorption force based on the multi-dimensional operation data through an adaptive control algorithm to maintain the robot's stable adhesion to the curtain wall. The control module 4 is used to receive movement commands from the integrated remote control terminal, combine the adhesion status data and the spacing information, and coordinate the control of the drive wheel to achieve three-dimensional movement, while maintaining closed-loop adjustment of the adsorption force during the movement. Monitoring module 5 is used to monitor abnormal system conditions. When it detects that the adsorption force is lower than the safety threshold, communication is interrupted, or the remaining battery power is too low, it triggers the emergency handling procedure and performs a safe shutdown.

[0062] In one embodiment, the control module includes: The conversion unit is used to parse the movement commands from the integrated remote control terminal, which include directional control signals in three dimensions: up and down, left and right, and forward and backward, and convert them into target speed and steering commands for the drive wheels; The judgment unit is used to determine the bonding state based on the real-time distance. If the real-time distance exceeds the preset maximum safe distance, it is determined that the bonding is loose and the adsorption force enhancement command is triggered first. The calculation unit is used to combine the movement direction command and real-time spacing information to calculate and output the control signal of the drive wheel motor, and at the same time realize closed-loop control of speed and position through motor encoder feedback. The adjustment unit is used to continuously monitor the change of adsorption pressure value during the movement execution phase. If the pressure fluctuation exceeds the set stable threshold, the movement process is paused and the adsorption control algorithm is invoked to readjust the adsorption pressure value to the stable range. The sending unit is used to maintain the current adsorption force output after a single movement command is executed, wait for the next command, and send the robot's real-time position, movement status and adsorption pressure value to the display screen of the integrated remote control terminal for visual display.

[0063] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described curtain wall cleaning drone control method.

[0064] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described curtain wall cleaning drone control method.

[0065] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0066] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0067] The above description is merely a preferred embodiment of the present invention and does not limit the scope of this application. Any equivalent results or equivalent process transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.

Claims

1. A control method for a curtain wall cleaning drone, characterized in that, Includes the following steps: Initialize the curtain wall cleaning drone and integrated remote control terminal, establish a low-latency communication connection, and self-check the status of each module; Real-time collection of multi-dimensional operational data, including data on the fit between the robot and the curtain wall, spacing information, wind load parameters in the high-altitude environment, and equipment operating status data; Based on the multi-dimensional operation data, the propeller speed and negative pressure adsorption force are dynamically calculated and adjusted through an adaptive control algorithm to maintain the robot's stable adhesion to the curtain wall. It receives movement commands from the integrated remote control terminal, combines the adhesion status data with the spacing information, and coordinates the control of the drive wheel to achieve three-dimensional movement, while maintaining closed-loop adjustment of the adsorption force during the movement. The system monitors abnormal conditions and triggers an emergency procedure to shut down the system when it detects that the adsorption force is below the safety threshold, communication is interrupted, or the remaining battery power is too low.

2. The control method for curtain wall cleaning drones according to claim 1, characterized in that, The steps for real-time acquisition of multi-dimensional operation data include: Real-time acquisition of the adsorption pressure value and real-time distance between the robot and the curtain wall; Real-time collection of wind speed and air pressure data in high-altitude work areas, and identification of strong wind interference scenarios based on wind speed changes; Real-time acquisition of propeller speed, drive motor load current, battery pack remaining power, and wireless signal strength of integrated remote control terminal; The collected adsorption pressure value, real-time spacing, wind speed, air pressure, propeller speed, motor load current, battery remaining power and wireless signal strength data are integrated into a unified format multi-dimensional status data packet and sent in real time to the integrated remote control terminal and the robot's onboard control unit.

3. The control method for curtain wall cleaning drones according to claim 1, characterized in that, The step of dynamically calculating and adjusting the propeller speed and negative pressure adsorption force through an adaptive control algorithm includes: The real-time collected adsorption pressure value is compared with the preset adsorption pressure safety range. If the adsorption pressure is lower than the lower limit of the safety range, it is determined that the adsorption force is insufficient and the propeller speed needs to be increased. If the adsorption pressure is higher than the upper limit of the safety range, it is determined that the adsorption force is overloaded and the propeller speed needs to be reduced. Based on the real-time collected wind speed data, the degree of wind load on the adsorption system is calculated, and the wind load influence coefficient is calculated by combining the preset wind load adjustment coefficient. Based on the deviation between the measured adsorption pressure and the target value, and the wind load influence coefficient, the propeller speed adjustment is calculated using proportional, integral, and derivative adjustment algorithms. Based on the speed adjustment amount, the actual speed of the propeller is dynamically adjusted so that the adsorption pressure is stabilized within the preset safety range; Before and during the robot's movement, the fluctuations in adsorption force that may be caused by displacement are predicted in real time, and the propeller speed is finely adjusted in advance by feedforward control to counteract these fluctuations.

4. The control method for curtain wall cleaning drones according to claim 1, characterized in that, The steps of receiving movement commands from the integrated remote control terminal, combining the adhesion status data and the spacing information, and coordinating the control of the drive wheels to achieve three-dimensional movement, while maintaining closed-loop adjustment of the adsorption force during movement, include: The movement commands from the integrated remote control terminal are parsed. The commands include directional control signals in three dimensions: up and down, left and right, and forward and backward. These commands are then converted into target speed and steering commands for the drive wheels. Based on the real-time spacing, the bonding status is determined. If the real-time spacing exceeds the preset maximum safe distance, the bonding is determined to be loose, and the adsorption force enhancement command is triggered first. By combining the movement direction command and real-time spacing information, the control signal of the drive wheel motor is calculated and output, and closed-loop control of speed and position is achieved through feedback from the motor encoder. During the migration execution phase, the changes in adsorption pressure value are continuously monitored. If the pressure fluctuation exceeds the set stable threshold, the migration process is paused and the adsorption control algorithm is invoked to readjust the adsorption pressure value to a stable range. After a single movement command is executed, the robot maintains its current adsorption force output, waits for the next command, and sends its real-time position, movement status, and adsorption pressure value to the display screen of the integrated remote control terminal for visual display.

5. The control method for curtain wall cleaning drones according to claim 1, characterized in that, When the monitoring system detects an abnormal state, such as adsorption force falling below a safety threshold, communication interruption, or low battery power, the steps to trigger the emergency handling procedure and execute a safe shutdown include: The adsorption pressure value is continuously monitored. If the pressure value is lower than the preset safety pressure threshold and the duration exceeds the first set time, it is determined to be an adsorption abnormality. The integrated remote control terminal is then controlled to issue an audible and visual alarm and automatically increase the propeller speed by a fixed ratio. The wireless communication signal strength between the integrated remote control terminal and the robot is continuously monitored. If the signal strength is lower than the preset strength threshold and the duration exceeds the second set duration, it is determined that the communication is interrupted, and the drone is controlled to enter the hovering emergency mode to maintain the current adsorption state and position. It continuously monitors the remaining battery power. When the remaining power is lower than the preset low power threshold, it triggers a low power alarm and displays a clear low power message on the remote control screen. If any of the following conditions—adsorption abnormality, communication interruption, or low battery state—cannot be resolved after taking corresponding measures, a safety shutdown procedure will be initiated. This procedure includes gradually reducing the propeller speed to a stop, releasing the adsorption pressure, and allowing the robot to smoothly detach from the curtain wall surface. After all emergency events have been handled, the system automatically records the type of the event, the time of occurrence, and the handling measures taken, forming an operation log. This log can be exported through the integrated remote control terminal for analysis.

6. The control method for curtain wall cleaning drones according to claim 1, characterized in that, It also includes extended functionality, with specific steps including: The system acquires images of the curtain wall surface, identifies the material type of the curtain wall based on image features, and automatically matches and calls the preset adsorption pressure parameters of the material according to the identification results. During the operation, the robot's movement trajectory, the range of the cleaned area, and the change of adsorption pressure over time are continuously recorded and stored in the onboard non-volatile memory. Based on the real-time speed of the propeller and the load current of the motor, the noise level of the system is evaluated. If the noise level exceeds the preset comfort threshold, the speed control strategy is automatically adjusted or the noise reduction structure is activated to reduce the noise. It supports remote updates to the robot's control program and remote debugging of control parameters via wireless communication networks to adapt to different high-altitude curtain wall cleaning operation scenarios.

7. A control system for a curtain wall cleaning drone, characterized in that, include: The initialization module is used to initialize the curtain wall cleaning drone and the integrated remote control terminal, establish a low-latency communication connection, and self-check the status of each module. The data acquisition module is used to collect multi-dimensional operational data in real time, including data on the fit between the robot and the curtain wall, spacing information, wind load parameters in the high-altitude environment, and equipment operating status data. The calculation module is used to dynamically calculate and adjust the propeller speed and negative pressure adsorption force based on the multi-dimensional operation data through an adaptive control algorithm to maintain the robot's stable adhesion to the curtain wall. The control module is used to receive movement commands from the integrated remote control terminal, combine the adhesion status data and the spacing information, and coordinate the control of the drive wheel to achieve three-dimensional movement, while maintaining closed-loop adjustment of the adsorption force during the movement. The monitoring module is used to monitor abnormal system conditions. When it detects that the adsorption force is lower than the safety threshold, communication is interrupted, or the remaining battery power is too low, it triggers the emergency handling procedure and performs a safe shutdown.

8. The curtain wall cleaning drone control system according to claim 7, characterized in that, The control module includes: The conversion unit is used to parse the movement commands from the integrated remote control terminal, which include directional control signals in three dimensions: up and down, left and right, and forward and backward, and convert them into target speed and steering commands for the drive wheels; The judgment unit is used to determine the bonding state based on the real-time distance. If the real-time distance exceeds the preset maximum safe distance, it is determined that the bonding is loose and the adsorption force enhancement command is triggered first. The calculation unit is used to combine the movement direction command and real-time spacing information to calculate and output the control signal of the drive wheel motor, and at the same time realize closed-loop control of speed and position through motor encoder feedback. The adjustment unit is used to continuously monitor the change of adsorption pressure value during the movement execution phase. If the pressure fluctuation exceeds the set stable threshold, the movement process is paused and the adsorption control algorithm is invoked to readjust the adsorption pressure value to the stable range. The sending unit is used to maintain the current adsorption force output after a single movement command is executed, wait for the next command, and send the robot's real-time position, movement status and adsorption pressure value to the display screen of the integrated remote control terminal for visual display.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images