Control methods, devices, cleaning equipment and storage media for window cleaning robots

Abstract

This specification provides a control method for a window cleaning robot, relating to the field of intelligent cleaning technology. The method uses a vision sensor to collect environmental detection information from the operating surface; then, based on the environmental detection information, it determines the distance parameter between the window cleaning robot and the edge of the operating surface; if the distance parameter meets the distance condition, it controls the window cleaning robot to perform a turning operation and activates the ball-head photoelectric sensor configured on the window cleaning robot; furthermore, by collecting the trigger status signal of the ball-head photoelectric sensor, it judges the risk state and executes corresponding avoidance operations to prevent the window cleaning robot from falling off the operating surface. This solution utilizes a vision sensor for non-contact distance detection, enabling proactive turning and avoidance before reaching the edge, and activates the contact ball-head photoelectric sensor as backup protection in advance, achieving dual protection of proactive prediction and passive contact, reducing the risk of the robot dangling and falling due to delayed protection.

Description

Technical Field

[0001] This specification relates to the field of intelligent cleaning technology, specifically to the control technology of window cleaning robots in the field of intelligent cleaning technology, and more specifically to the control methods, devices, cleaning equipment and storage media of window cleaning robots. Background Technology

[0002] In the technologies related to autonomous mobile cleaning equipment such as automated window cleaning robots, operational safety is always one of the core requirements. These devices typically need to autonomously navigate and perform cleaning tasks on vertical or inclined work surfaces, inevitably encountering edges, steps, or other obstacles during their operation. Therefore, reliably identifying and addressing the edges of work surfaces to prevent the equipment from falling is a significant challenge. Summary of the Invention

[0003] This specification provides a control method, device, cleaning equipment, and storage medium for a window cleaning robot, improving the safety of the window cleaning robot's operation.

[0004] To achieve the above technical objectives, the embodiments of this specification provide the following technical solutions: Firstly, one embodiment of this specification provides a control method for a window cleaning robot, comprising: During the operation of the window cleaning robot, environmental detection information of the surface on which it operates is collected through visual sensors; The distance parameter between the window cleaning robot and the edge of the operating surface is determined based on the environmental detection information; If the distance parameter meets the distance condition, the window cleaning robot is controlled to perform a turning operation, and the ball head photoelectric sensor configured on the window cleaning robot is activated; By collecting the trigger status signal of the ball-head through-beam sensor, the output of the risk status signal is determined. The risk status signal is used to control the window cleaning robot to perform a posture adjustment operation, which is used to prevent the window cleaning robot from falling off the running surface.

[0005] Secondly, one embodiment of this specification provides a control device for a window cleaning robot, comprising: The acquisition unit is used to collect environmental detection information of the operating surface through a vision sensor during the operation of the window cleaning robot; The control unit is used to determine the distance parameter between the window cleaning robot and the edge of the operating surface based on the environmental detection information. The control unit is further configured to control the window cleaning robot to perform a turning operation and activate the ball head through-beam sensor configured on the window cleaning robot if the distance parameter meets the distance condition. The control unit is also used to determine the output of a risk status signal by acquiring the trigger status signal of the ball head through-beam sensor. The risk status signal is used to control the window cleaning robot to perform a posture adjustment operation, which is used to prevent the window cleaning robot from falling off the running surface.

[0006] Thirdly, one embodiment of this specification also provides a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the control method of the window cleaning robot as described above.

[0007] Fourthly, one embodiment of this specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the control method for the window cleaning robot as described above.

[0008] Fifthly, embodiments of this specification provide a computer program product or computer program, the computer program product including a computer program, the computer program being stored in a computer-readable storage medium or in the cloud; the processor of the computer device reads the computer program, and when the processor executes the computer program, it implements the steps of the above-described control method for the window cleaning robot.

[0009] As can be seen from the above technical solution, the control method for the window cleaning robot provided in this specification uses a visual sensor to collect environmental detection information of the operating surface during the operation of the window cleaning robot; then, based on the environmental detection information, it determines the distance parameter between the window cleaning robot and the edge of the operating surface; if the distance parameter meets the distance condition, it controls the window cleaning robot to perform a turning operation and activates the ball-head photoelectric sensor configured on the window cleaning robot; furthermore, by collecting the trigger status signal of the ball-head photoelectric sensor, it judges the output of the risk status signal, which is used to control the window cleaning robot to perform a posture adjustment operation to prevent the window cleaning robot from falling off the operating surface. This solution uses a visual sensor for non-contact distance detection, which can actively turn and avoid the edge before reaching it, and activates the contact ball-head photoelectric sensor as a backup protection in advance, realizing dual protection of active prediction and passive contact, reducing the risk of the robot being suspended in mid-air and falling due to protection lag. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this specification. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0011] Figure 1 This is a schematic diagram illustrating the application environment of the control method for a window cleaning robot provided as one embodiment of this specification.

[0012] Figure 2 This is a flowchart illustrating a control method for a window cleaning robot, provided as one embodiment of this specification.

[0013] Figure 3 This is a schematic diagram of the path planning results of a window cleaning robot.

[0014] Figure 4 This is a schematic diagram of the functional modules of the control device for a window cleaning robot provided in one embodiment of this specification.

[0015] Figure 5 This is a schematic diagram of the structure of a computing device provided for one embodiment of this specification. Detailed Implementation

[0016] Unless otherwise defined, the technical or scientific terms used in the embodiments of this specification shall have the ordinary meaning understood by one of ordinary skill in the art to which this specification pertains. The terms "first," "second," and similar terms used in the embodiments of this specification do not indicate any order, quantity, or importance, but are merely used to avoid confusion of constituent elements.

[0017] Unless the context otherwise requires, throughout this specification, "a plurality of" means "at least two," and "including" is interpreted as open-ended or encompassing, that is, "including, but not limited to." In the description of this specification, terms such as "one embodiment," "some embodiments," "exemplary embodiment," "example," "specific example," or "some examples" are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example of this specification. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example.

[0018] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0019] In the technologies related to autonomous mobile cleaning equipment such as automated window cleaning robots, operational safety is always one of the core requirements. These devices typically need to autonomously navigate and perform cleaning tasks on vertical or inclined work surfaces, inevitably encountering edges, steps, or other obstacles during their operation. Therefore, reliably identifying and addressing the edges of work surfaces to prevent the equipment from falling is a significant challenge.

[0020] For edge detection using a contact-type mechanical ball-head probe, a retractable ball-head probe is installed at the bottom of the machine body. When the machine moves linearly to the edge of the surface and the ball head loses its glass support, it pops out under the action of a spring and triggers a microswitch, thereby initiating emergency braking or steering. The detection principle of this method determines that it can only generate a protective action when the ball head has completely passed the edge and part of the machine body is already in a suspended state, resulting in an inherent lag in response. In addition, there are also attempts to use non-contact sensors such as infrared or ultrasonic sensors to detect the edge in advance in order to react early. However, such sensor signals are easily interfered with by operating environment factors such as glass light transmission, specular reflection, water stains, and dust accumulation, making it difficult to meet the reliability and consistency of the detection data for safety protection requirements.

[0021] Therefore, in the anti-fall protection mechanism of window cleaning robots, how to alleviate the conflict between the response lag of contact detection and the unreliability of non-contact detection, and achieve more advanced and stable edge perception and risk avoidance, has become a technical problem that urgently needs to be solved in this field.

[0022] To address the aforementioned problems, this specification provides a control system for a window cleaning robot, and the control method for the window cleaning robot provided in this specification is applied to the control system of the window cleaning robot. This control system, when the visual sensor predicts an impending boundary crossing, not only performs steering but also releases the ball head in advance and simultaneously switches its judgment mode, adjusting the delayed passive protection (triggered only after boundary crossing) to an active collaborative protection mechanism of standby and immediate stop upon contact.

[0023] Specifically, the control system of this window cleaning robot may include a... Figure 1The system comprises a client 110, a server 120, and a robot 130, forming an operating environment. The client 110 communicates with the server 120 via a network. The client 110 is wirelessly connected to the robot 130. The robot 130 communicates with the server 120 via a network connection. The client 110 can be an electronic device with network access capabilities. Specifically, for example, the client 110 can be a desktop computer, tablet, laptop, smartphone, digital assistant, smart wearable device, shopping guide terminal, television, smart speaker, microphone, etc. Smart wearable devices include, but are not limited to, smart bracelets, smartwatches, smart glasses, smart helmets, smart necklaces, etc. Alternatively, the client 110 can also be software that can run on the electronic device. The server 120 can be an electronic device with certain computing power. It can have a network communication module, processor, and memory, etc. Of course, the server 120 can also refer to software running on the electronic device. The server 120 can also be a distributed server, which can be a system with multiple processors, memory, network communication modules, etc., operating collaboratively. Alternatively, the server 120 can also be a cluster formed by several servers 120. Alternatively, with the development of science and technology, server 120 could also be a new technological means capable of realizing the corresponding functions of the implementation method described in the manual. For example, it could be a new form of "server" based on quantum computing.

[0024] Specifically, when controlling the window cleaning robot based on the aforementioned control system, environmental detection information of the operating surface is collected by a vision sensor during the robot's operation. Then, the distance parameter between the window cleaning robot and the edge of the operating surface is determined based on this environmental detection information. If the distance parameter meets the distance condition, the window cleaning robot is controlled to perform a turning operation, and the ball-head photoelectric sensor configured on the window cleaning robot is activated. Furthermore, by collecting the trigger status signal of the ball-head photoelectric sensor, a risk status signal is determined. This risk status signal is used to control the window cleaning robot to perform a posture adjustment operation to prevent the window cleaning robot from falling off the operating surface. This solution utilizes a vision sensor for non-contact distance detection, enabling proactive turning and avoidance before reaching the edge, and activates the contact ball-head photoelectric sensor as backup protection in advance, achieving dual protection of proactive prediction and passive contact, reducing the risk of the robot dangling or falling due to protection lag.

[0025] Based on the above concept, this specification provides a control method for a window cleaning robot. The control method for the window cleaning robot provided in this specification will be described exemplarily below with reference to the accompanying drawings.

[0026] To be applied Figure 1Taking the robot in the example, this specification provides illustrative examples of some implementation methods for controlling the window cleaning robot, such as... Figure 2 As shown, Figure 2 A flowchart illustrating a control method for a window cleaning robot, provided as one embodiment of this specification; the control method for the window cleaning robot includes: 201. During the operation of the window cleaning robot, environmental detection information of the operating surface is collected through a visual sensor.

[0027] In this embodiment, the window cleaning robot includes a control unit, a vision sensor, a ball-head through-beam sensor, an inertial measurement unit (IMU), drive wheels, an adsorption fan, and a current detection module for detecting the current of the drive wheels. The control unit is communicatively connected to the ball-head through-beam sensor, the inertial measurement unit, the drive wheels, the adsorption fan, and the current detection module.

[0028] The term "visual sensor" broadly refers to sensor devices capable of non-contactly sensing environmental features by emitting and receiving electromagnetic or light waves. Specifically, it can refer to any one or a combination of laser rangefinders, microwave radar sensors, infrared rangefinders, structured light sensors, or time-of-flight sensors. In one possible implementation, the visual sensor is a combination of a laser rangefinder and a microwave radar sensor, both mounted on the front bottom of the window cleaning robot, with their detection beams directed towards the robot's direction of travel and positioned close to the surface. This approach allows for the acquisition of edge information before the robot approaches the surface edge, providing a data basis for subsequent predictive avoidance.

[0029] As a specific implementation, a laser rangefinder emits detection pulses at a preset frequency (e.g., 10Hz) towards the operating surface and receives echo signals reflected from the surface and edges. A microwave radar sensor continuously emits microwave signals and receives echoes; the two work together to ensure stable acquisition of environmental detection information even in conditions of reduced visibility, such as dust or water film. This environmental detection information may include, but is not limited to, the distance from the sensor to the surface edge, the intensity distribution of the reflected signal, and the pulse interval characteristics in the echo signal. More generally, environmental detection information can be acquired in various ways. For example, the method may also include: periodically or in real-time reading the raw data stream from a vision sensor during robot operation, and filtering and extracting features from the raw data stream to generate structured data for subsequent distance determination.

[0030] 202. Determine the distance parameters between the window cleaning robot and the edge of the operating surface based on environmental detection information.

[0031] In this embodiment, the specific means to achieve the above-mentioned "determine distance parameters" function may include: at the hardware level, the signal processor in the control unit can receive the analog or digital signals output by the vision sensor; at the software or logic level, it can be achieved by executing edge recognition algorithms and distance calculation algorithms. For example, for a laser rangefinder sensor, the control unit can analyze the time of flight or phase difference in its echo signal to directly calculate the straight-line distance from the robot's front end to the surface edge. For a microwave radar sensor, distance information can be obtained by analyzing the difference frequency signal of Doppler frequency shift or frequency-modulated continuous wave. When multiple sensors are used, the distance values ​​calculated by each sensor can also be fused (e.g., weighted averaging or Kalman filtering) to output a more robust comprehensive distance parameter. This distance parameter characterizes the spatial relationship between the current position of the window cleaning robot and the boundary of the operating surface, and is the core basis for determining whether to trigger an avoidance action.

[0032] In one possible scenario, to further optimize the accuracy and reliability of distance parameter determination in the above embodiments and address the problem of random errors caused by environmental interference such as glass light transmission, specular reflection, or dirt from a single sensor, the process of "determining the distance parameter between the window cleaning robot and the edge of the operating surface based on environmental detection information" can be specifically defined as follows: determining the detection parameter between the window cleaning robot and the edge of the operating surface based on environmental detection information; acquiring the operating map of the window cleaning robot; determining map parameters based on the operating map and the position information of the window cleaning robot; and verifying the detection parameter based on the map parameter to determine the distance parameter. This limitation aims to utilize known environmental map information to cross-validate real-time sensor data to filter out abnormal jumps.

[0033] Specifically, for the data verification process, the control unit first calculates the original detection parameter D_sensor using data acquired by the vision sensor. Simultaneously, the control unit uses positioning sensors such as an inertial measurement unit or wheeled odometers to determine the robot's position information in real time within a pre-generated or instantaneously constructed operational map. This operational map contains the known geometric coordinates of the surface edges. Based on the robot's current pose, the control unit calculates the map parameter D_map from the robot's front end to the known edge from the operational map. Subsequently, the control unit compares and fuses the detection parameter D_sensor with the map parameter D_map. For example, the difference between the two can be calculated; if the difference is less than a preset confidence interval, the weighted average of the two is taken as the final distance parameter; if the difference is too large, it indicates that the vision sensor data may be abnormal, and the map parameter D_map can be prioritized or a degradation processing logic can be triggered.

[0034] As can be seen, this embodiment eliminates the reliance on real-time, easily interfered sensor data for distance parameter determination, instead introducing prior spatial constraints. Its advantage lies in effectively eliminating short-term, large-amplitude ranging errors caused by environmental interference, significantly improving the accuracy of edge recognition and the overall robustness of the system. Those skilled in the art will understand that alternative structures for data verification include: using a Kalman filter to treat the probed parameters as observed values, using predicted values ​​derived from maps and motion models as state estimates, and outputting the optimal distance estimate in real-time through a filtering algorithm; or determining the distance by matching pre-stored offline map data with real-time sensor data using point clouds.

[0035] By adopting the above-mentioned multi-source data verification mechanism, this embodiment can still output accurate and reliable distance parameters under complex lighting or material conditions. This further helps to solve the problem of false triggering or missed protection caused by inaccurate distance judgment, thereby synergistically enhancing the overall technical effect of the present invention in achieving active and precise protection.

[0036] 203. If the distance parameter meets the distance condition, control the window cleaning robot to perform a turning operation and activate the ball head photoelectric sensor configured on the window cleaning robot.

[0037] In this embodiment, the term "distance condition" broadly refers to one or more preset judgment criteria used to determine whether a robot is approaching or has reached a dangerous area. Specifically, it can refer to a single condition that compares a real-time calculated distance parameter with at least one preset distance threshold, or it can encompass composite conditions based on the distance parameter and its changing trend. When the distance parameter indicates that the distance between the window cleaning robot and the edge of the surface enters a preset warning range, the distance condition is determined to be met.

[0038] In one possible scenario, the ball-head through-beam sensor is a contact-type sensing device configured on a window cleaning robot. In this embodiment, the ball-head through-beam sensor includes a retractable ball-head probe, an electromagnetic locking mechanism, and a trigger switch. Under normal operating conditions, the ball-head probe can be locked in the retracted position by the electromagnetic locking mechanism. When the control unit determines that the distance parameter meets the distance condition, it immediately sends an unlocking command to the electromagnetic locking mechanism, causing the ball-head probe to enter a free, ready state under the preload of the elastic element. At this time, the end of the ball head is still in contact with and pressed against the operating surface, in a state awaiting triggering. It should be noted that the aforementioned "activation" action does not necessarily mean that the ball-head through-beam sensor is triggered, but rather that its locking restraint is released, enabling it to immediately respond to boundary crossing events. This action is performed synchronously with the steering operation, ensuring that during steering avoidance, if the vision sensor accidentally misses a detection causing the robot body to partially cross the edge, the ball-head through-beam sensor can be released and triggered within a very short stroke, providing immediate backup protection.

[0039] For controlling the window cleaning robot to perform turning operations, the control unit can specifically plan a turning path in real time based on the edge orientation analyzed by environmental detection information, such as making a 90° turn or a walking path along the edge, and generate corresponding motor drive commands to control the differential operation of the walking motor to change the robot's movement direction so that it moves away from the danger zone before reaching the edge.

[0040] It should be noted that, considering the ball-head beam sensor is a contact-triggered sensor, it has limitations. The robot must first move beyond the edge of the ball head before it can release, at which point part of the robot body is already suspended in the air, posing a high risk. Therefore, this embodiment uses laser / microwave as the primary predictor and the ball-head beam sensor as a backup confirmation. If the laser misses a detection, leading to an out-of-bounds movement, the ball-head beam sensor can still trigger protection, improving operational safety. That is, the application of the vision sensor in steps 101 and 102 provides the necessary space and time for the early execution of the turning operation and activation of the ball-head beam sensor in step 103. The synergy of these two mechanisms ensures that the robot is under dual protection—"active prediction and avoidance" and "passive contact protection"—throughout the entire process from "approaching the edge" to "completing the turn." This jointly solves the problem of delayed protection and partial suspension of the robot body caused by relying solely on contact sensors in related technologies, significantly reducing the risk of fall.

[0041] As mentioned above, the control strategy when the distance parameter meets the distance condition can be implemented in various ways. In one possible implementation, in order to further solve the problem of avoidance failure caused by the robot's excessive speed or late turning start, step 103 is specifically defined as a phased execution process.

[0042] In this embodiment, the feature "if the distance parameter meets the distance condition, control the window cleaning robot to perform a turning operation and activate the ball-head photoelectric sensor configured on the window cleaning robot" is refined as follows: determine the distance threshold indicated by the distance condition; when the distance parameter is close to the distance threshold, control the window cleaning robot to perform a deceleration operation; when the distance parameter is less than the distance threshold, indicate that the distance condition is met, control the window cleaning robot to perform a turning operation and activate the ball-head photoelectric sensor configured on the window cleaning robot.

[0043] Specifically, the control unit pre-determines a distance threshold, such as 30mm. Simultaneously, a "proximity detection zone" slightly larger than this threshold can be set. When the real-time calculated distance parameter falls within this proximity detection zone (e.g., between 30mm and 45mm), the control unit determines that the robot is approaching a danger zone and immediately generates a deceleration command, controlling the walking motor to reduce the travel speed from the normal cleaning speed to a preset safe low speed, such as a 50% reduction. During this stage, the robot's direction of movement has not yet changed. Only when the robot continues to move, causing the distance parameter to eventually fall below the 30mm distance threshold, does the control unit determine that the distance condition has been formally met. At this point, the robot's speed has been significantly reduced, and the control unit then issues a steering command and a ball-end activation command to drive the robot to complete the turn.

[0044] As can be seen, this embodiment allows the robot to have lower momentum and a longer response time before performing a turning maneuver. Its advantage lies in greatly improving the success rate of turning operations and effectively avoiding turning failures caused by high-speed collisions with edges. Those skilled in the art will understand that alternative solutions to achieve the same "segmented control" objective include: dynamically adjusting the distance threshold based on the robot's real-time speed, with the threshold and proximity determination zone set larger as the speed increases; or, issuing only visual or audible warnings during the approach phase without immediate deceleration, and only executing forced deceleration and turning when the threshold is exceeded.

[0045] Furthermore, the aforementioned distance threshold is not limited to a fixed value and can be adjusted according to the actual application scenario. To further address the problem of false deceleration or untimely protection caused by the inability of a fixed threshold to adapt to glass of different thicknesses, this application also provides another possible threshold determination scheme. Specifically, the process of determining the distance threshold includes: determining a preset threshold for distance condition indication; acquiring the thickness information corresponding to the operating surface based on a vision sensor; configuring an adjustment factor based on the thickness information, wherein the thickness information is negatively correlated with the adjustment factor; and adjusting the preset threshold using the adjustment factor to obtain the distance threshold.

[0046] This embodiment utilizes the echo characteristics of a visual sensor to invert the physical properties of the operating surface. As a specific implementation, after a laser rangefinder emits a probe pulse towards the operating surface, the received echo signal will contain double pulses or obvious interference characteristics due to reflections from both the upper and lower surfaces of the glass. The control unit can calculate the thickness of the current operating surface by analyzing the time intervals of these pulses or the distribution of the interference pattern, and thus identify its thickness type, such as 4mm single-pane glass, 25mm insulated glass, or laminated glass.

[0047] The control unit internally stores a lookup table or function that reflects the mapping relationship between glass thickness type and warning distance threshold. Following the physical principle that "the thicker the surface, the smoother the change in edge reflection characteristics, and the greater the required warning distance," this mapping relationship is reflected as a negative correlation between thickness information and the adjustment factor. For example, if a single-layer glass is used as a benchmark, its adjustment factor is 1.0, and the preset threshold is 30mm; for thicker insulated glass, the calculated thickness value is greater, so an adjustment factor greater than 1.0 is configured, such as 1.6, thereby adjusting the final distance threshold to approximately 50mm. In this way, the system can adjust the warning distance online and adaptively, providing a more sufficient protection distance on thick glass and avoiding premature and ineffective deceleration due to an excessively large threshold on thin glass, thus balancing safety coverage and cleaning efficiency. Those skilled in the art can easily combine the adaptive threshold scheme in this embodiment with the aforementioned staged control scheme. This combination can simultaneously achieve accurate warnings adapted to different glass types and high-success-rate step-by-step avoidance operations.

[0048] 204. By collecting the trigger status signal of the ball head through-beam sensor, the output of the risk status signal is determined. The risk status signal is used to control the window cleaning robot to perform posture adjustment operation. The posture adjustment operation is used to prevent the window cleaning robot from falling off the running surface.

[0049] In this embodiment, the risk status signal is used to control the window cleaning robot to perform a posture adjustment operation, which is used to prevent the window cleaning robot from falling off the operating surface. During the posture adjustment process, the system continuously monitors the current of the drive wheel motor: if the current does not exceed the current threshold (e.g., 700mA), the posture adjustment or obstacle-crossing action continues, and then the normal cleaning process is resumed; if the current exceeds the threshold, it is determined that there is an extreme risk of the frame or suspension, triggering a system alarm. At the same time, the fan adsorption power is increased to ensure safety, and finally the robot stops, locks its posture, and waits for recovery to prevent it from falling and being pulled down.

[0050] It is understandable that the ball-head through-beam sensor contains an infrared through-beam. During the ejection of the ball-head probe 131 due to loss of support after crossing the surface edge, its connecting rod immediately blocks or opens the optical path between the infrared through-beams, thus generating a state switching event, which is manifested as a trigger state signal. The control unit collects this trigger state signal in real time. Once a valid state switch is detected, a risk state signal is generated based on preset logic (e.g., simple trigger logic, or composite judgment logic combining other sensor signals).

[0051] In response to this risk status signal, the control unit instructs the window cleaning robot to perform a posture adjustment operation. The term "posture adjustment operation" broadly refers to any emergency action taken to prevent the robot from falling off the operating surface. Examples include: immediately braking all walking motors, performing a reverse movement to return to a safe area, or activating an adsorption enhancement mechanism to increase negative pressure. For instance, the control unit can urgently cut off the drive current to the forward motor while simultaneously applying a short pulse to the reverse motor, causing the robot to generate a momentary displacement away from the edge, thereby preventing a fall.

[0052] To further address the issue of potential false triggering of protection during turning or normal operation due to reliance on a single signal from the ball-head photoelectric sensor, step 104, "determining the output of a risk status signal by acquiring the trigger status signal of the ball-head photoelectric sensor," can be further elaborated as follows: acquire the trigger status signal of the ball-head photoelectric sensor; in response to the switching process of the trigger status signal, acquire the attitude angle signal through the inertial measurement unit; determine the tilt angle of the window cleaning robot indicated by the attitude angle signal; if the tilt angle exceeds a preset angle threshold, output a risk status signal.

[0053] For the component coordination process involving complex decision-making logic, the ball-head photoelectric sensor includes a flexible contact rod and an infrared photoelectric tube. Its state switching process is triggered when the flexible contact rod is pushed by an obstacle or loses support: during normal operation, the flexible contact rod is pressed against the glass surface, blocking the light path between the infrared photoelectric tubes; when the robot moves beyond the edge, the flexible contact rod pops out, its connecting rod moves away, and the light path between the infrared photoelectric tubes becomes open. This change in connection or disconnection generates a trigger state signal.

[0054] Upon detecting this state transition, the control unit does not simply output a risk status signal directly; instead, it simultaneously wakes up or reads data from the inertial measurement unit (IMU). The IMU typically includes a gyroscope and an accelerometer, capable of sensing the aircraft's angular velocity and linear acceleration in real time, thereby calculating the aircraft's tilt angle. The control unit then compares the currently calculated tilt angle with a preset angle threshold (e.g., 15 degrees).

[0055] As can be seen, the principle of this embodiment lies in distinguishing between "tilts caused by normal operation" and "tilts caused by falls." When the robot makes sharp turns, crosses small protrusions, or moves across uneven surfaces, the robot body may tilt briefly and to a limited extent. At this time, the ball joint may also be triggered momentarily due to deformation. By introducing an AND logic judgment of the attitude angle, only when the robot body tilts beyond the range that can be generated by normal operation (i.e., exceeding the angle threshold) is it finally determined to be a real risk of falling beyond the limit, and then a risk status signal is output to perform attitude adjustment operations such as emergency braking. Its advantage is that it can accurately distinguish between normal operating disturbances and real falls, significantly reducing the false alarm rate and ensuring the accuracy of protective actions. Those skilled in the art will understand that alternative solutions to achieve this composite judgment purpose include: judging the duration of the trigger status signal, ignoring it if it is only a very short pulse; or, combining wheel speed sensors to judge whether there is a sudden change in rotation speed caused by unilateral suspension, etc.

[0056] By employing the judgment logic that combines mechanical triggering and inertial sensing, this embodiment can reliably eliminate false triggering under non-hazardous working conditions. This further helps to improve the availability of the system and the user's trust, thereby synergistically strengthening the integrity and accuracy of the safety protection of this invention.

[0057] As mentioned earlier, the ball-head through-beam sensor is activated and enters a rapid response mode during the turning process. To further optimize system operation and prevent the sensor from remaining active after the robot has moved away from the edge risk area from interfering with subsequent normal operation, for scenarios without edge risks, time-series statistics can be performed on the distance parameters to obtain the corresponding change information. If the change information indicates that the window cleaning robot is moving away from the surface edge in the operating surface, the ball-head through-beam sensor is turned off, and the drive parameters of the window cleaning robot are adjusted. This limitation aims to achieve timely exit of the backup protection mechanism and seamless transition to the normal operating state.

[0058] Specifically, after successfully executing a turning maneuver, the control unit continuously monitors the distance parameters detected by the vision sensors. By calculating the rate of change of the distance parameters over time or the trend of changes in values ​​at multiple consecutive sampling points, the current direction of robot movement can be determined. When the change information indicates that the distance parameters are continuously increasing, meaning the robot is moving away from the surface edge, the control unit determines that the robot has returned to an absolutely safe area.

[0059] At this point, the control unit immediately executes a recovery operation. On one hand, it sends a locking command to the electromagnetic locking mechanism to retract and lock the ball-head probe, thus removing it from the free-ready state and preventing accidental triggering by minor undulations during subsequent routine cleaning. On the other hand, it adjusts the drive parameters of the window cleaning robot to restore them to normal, for example, restoring the trigger filtering time of the ball-head sensor from 10 milliseconds in fast response mode to the normal 200 millisecond anti-shake filtering time, while simultaneously restoring the travel speed to the set value for maximum cleaning efficiency.

[0060] As can be seen, this embodiment enables a smooth transition between safety protection and routine cleaning, ensuring rapid response capabilities in hazardous areas while avoiding frequent false alarms due to excessive sensitivity in safe areas, thereby significantly improving the robot's autonomous continuous operation capability. Furthermore, the triggering condition for the aforementioned recovery operation is not limited to a distance increasing trend; it can be adjusted according to the actual application scenario: when the robot travels a straight distance exceeding a preset safety threshold (e.g., 10 cm) after turning, the recovery process is also triggered. Those skilled in the art can easily combine the safety recovery mechanism in this embodiment with the aforementioned adaptive threshold or segmented control embodiments to form a closed-loop protection scheme covering the entire process from "hazard prediction - graded response - backup protection - safety recovery".

[0061] In another possible implementation, in order to make full use of the successfully identified edge information to optimize the global cleaning path and reduce the risk of triggering edge protection a second time when covering the remaining area, this application also provides a path planning method, which includes: in response to the triggering of a distance condition, performing path planning according to the edge distribution through the running map configured by the window cleaning robot to obtain edge paths and non-edge paths; controlling the window cleaning robot to clean according to the edge path; and executing the cleaning task corresponding to the non-edge path after the edge path is completed.

[0062] In this solution, for path planning scenarios such as Figure 3 As shown, Figure 3 This is a schematic diagram illustrating the path planning results of a window cleaning robot. When the robot first detects a surface edge and triggers a distance condition, its control unit does not limit itself to performing a simple turning avoidance maneuver. Instead, it uses the edge coordinate information acquired to plan an edge path on the operating map that is adjacent to, surrounds, or covers the edge area. This path is designated as the area requiring priority processing. After planning is complete, the control unit directly guides the robot to begin cleaning along this edge path.

[0063] The driving parameters for edge paths differ from those for non-edge paths (also known as internal paths). For example, considering the higher protection priority of edge areas, the robot can employ a lower constant speed and more sensitive collision avoidance logic when executing edge paths, while non-edge paths revert to the normal, efficient cleaning speed. After all edge paths have been traversed and cleaned, the robot's subsequent activity area will be confined to the internal safety zone surrounded by edge paths. At this point, when the robot is scheduled to perform non-edge path cleaning tasks in the internal area, it will almost never have the opportunity to touch the edges during movement. This significantly reduces the risk of repeatedly triggering edge protection or even accidents from a macroscopic perspective of path planning, while ensuring complete coverage of the cleaning area. Those skilled in the art can combine the aforementioned embodiments, for example, by synchronously feeding back the planned edge path information to the map parameter verification module, to further improve the intelligence level and operational efficiency of the entire system.

[0064] In summary, this embodiment utilizes a visual sensor to collect environmental detection information of the operating surface during the operation of the window cleaning robot. Based on this information, it determines the distance parameter between the robot and the edge of the operating surface. If the distance parameter meets the requirements, the robot is controlled to perform a turning operation, and its ball-head photoelectric sensor is activated. Furthermore, by collecting the trigger status signal of the ball-head photoelectric sensor, a risk status signal is generated. This risk status signal controls the robot to perform a posture adjustment operation to prevent it from falling off the operating surface. This solution utilizes a visual sensor for non-contact distance detection, enabling proactive turning and avoidance before reaching the edge. The contact ball-head photoelectric sensor is activated in advance as backup protection, achieving dual protection of proactive prediction and passive contact, reducing the risk of the robot dangling or falling due to delayed protection.

[0065] It should be noted that the various embodiments described in this specification emphasize the parts that differ from other embodiments, and the embodiments can be explained by comparison with each other. Any combination of the various embodiments described in this specification based on general technical knowledge is covered within the scope of this specification.

[0066] In one exemplary embodiment of this specification, a control device 400 for a window cleaning robot is also provided, such as... Figure 4 As shown, Figure 4 This specification provides a functional module diagram of a control device for a window cleaning robot according to one embodiment. The control device 400 includes: The acquisition unit 401 is used to collect environmental detection information of the operating surface through a vision sensor during the operation of the window cleaning robot; Control unit 402 is used to determine the distance parameter between the window cleaning robot and the edge of the operating surface based on the environmental detection information; The control unit 402 is also used to control the window cleaning robot to perform a turning operation and activate the ball head photoelectric sensor configured on the window cleaning robot if the distance parameter meets the distance condition. The control unit 402 is also used to determine the output of a risk status signal by acquiring the trigger status signal of the ball head through-beam sensor. The risk status signal is used to control the window cleaning robot to perform a posture adjustment operation, which is used to prevent the window cleaning robot from falling off the running surface.

[0067] Optionally, in one possible embodiment, the control unit 402 is specifically used to determine the detection parameters of the distance between the window cleaning robot and the edge of the operating surface based on the environmental detection information; The control unit 402 is specifically used to acquire the operating map of the window cleaning robot; The control unit 402 is specifically used to determine map parameters based on the operating map and the location information corresponding to the window cleaning robot; The control unit 402 is specifically used to verify the detection parameter based on the map parameters in order to determine the distance parameter.

[0068] Optionally, in one possible embodiment, the control unit 402 is specifically configured to determine the distance threshold indicated by the distance condition; The control unit 402 is specifically used to control the window cleaning robot to perform a deceleration operation when the distance parameter is close to the distance threshold. The control unit 402 is specifically used to indicate that the distance condition is met when the distance parameter is less than the distance threshold, control the window cleaning robot to perform a turning operation, and activate the ball head photoelectric sensor configured on the window cleaning robot.

[0069] Optionally, in one possible embodiment, the control unit 402 is specifically configured to determine a preset threshold for the distance condition indication; The control unit 402 is specifically used to acquire thickness information corresponding to the operating surface based on a vision sensor. The control unit 402 is specifically used to configure an adjustment factor based on the thickness information, wherein the thickness information is negatively correlated with the adjustment factor; The control unit 402 is specifically used to adjust the preset threshold through the adjustment factor to obtain the distance threshold.

[0070] Optionally, in one possible embodiment, the control unit 402 is specifically used to acquire the trigger status signal of the ball head through-beam sensor; The control unit 402 is specifically used to respond to the switching process of the trigger state signal by acquiring attitude angle signals through an inertial measurement unit. The ball head through-beam sensor includes an elastic contact rod and an infrared through-beam tube. The switching process is triggered when the elastic contact rod is pushed by an obstacle. After the elastic contact rod is pushed, it blocks the optical path between the infrared through-beam tubes. The optical path between the infrared through-beam tubes is opened after the elastic contact rod passes the obstacle. The control unit 402 is specifically used to determine the tilt angle of the window cleaning robot body indicated by the posture angle signal; The control unit 402 is specifically used to output a risk status signal if the tilt angle of the machine exceeds a preset angle threshold.

[0071] Optionally, in one possible embodiment, the control unit 402 is specifically used to perform time-series statistics on the distance parameters to obtain the change information corresponding to the distance parameters; The control unit 402 is specifically configured to shut down the ball head photoelectric sensor and adjust the drive parameters of the window cleaning robot if the change information indicates that the window cleaning robot is away from the surface edge in the operating surface.

[0072] Optionally, in one possible embodiment, the control unit 402 is specifically configured to, in response to the triggering of a distance condition, perform path planning according to the edge distribution using the operating map configured by the window cleaning robot, to obtain edge paths and non-edge paths; The control unit 402 is specifically used to control the window cleaning robot to clean along the edge path; The control unit 402 is specifically used to perform the cleaning task corresponding to the non-edge path after the edge path is completed. The driving parameters of the edge path are different from those of the non-edge path.

[0073] Specifically, the acquisition unit and control unit in this embodiment can correspond to physical components. For example, the control unit can be a processing module such as a CPU, GPU, or FPGA. The specific physical component can be any component or combination of components with the above functions. The specific method depends on the actual scenario and is not limited here.

[0074] The aforementioned control device uses a visual sensor to collect environmental detection information of the operating surface during the operation of the window cleaning robot. Based on this information, it determines the distance parameter between the robot and the edge of the operating surface. If the distance parameter meets the requirements, the robot is controlled to perform a turning operation, and its ball-head photoelectric sensor is activated. Furthermore, by collecting the trigger status signal of the ball-head photoelectric sensor, a risk status signal is generated. This risk status signal controls the robot to perform a posture adjustment operation to prevent it from falling off the operating surface. This solution utilizes a visual sensor for non-contact distance detection, enabling proactive turning and avoidance before reaching the edge. The contact ball-head photoelectric sensor is activated in advance as backup protection, achieving dual protection of proactive prediction and passive contact, reducing the risk of the robot dangling or falling due to delayed protection.

[0075] Specific limitations regarding the control device of the window cleaning robot can be found in the limitations regarding the control method of the window cleaning robot mentioned above, and will not be repeated here. Each unit module in the control device of the aforementioned window cleaning robot can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0076] Another embodiment of this application also proposes a computing device, see [link to relevant documentation] Figure 5 As shown, an exemplary embodiment of this specification also provides a computing device, including: a memory and a processor, the memory storing a computer program, the processor executing the computer program to perform the steps in the control method of the window cleaning robot according to various embodiments of this specification described in the above embodiments.

[0077] The internal structure of the computing device can be as follows: Figure 5 As shown, the computing device includes a processor, memory, network interface, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it follows the steps of the control method for a window cleaning robot according to various embodiments of this specification, as described in the above embodiments.

[0078] The processor may include the main processor, as well as baseband chips, modems, etc.

[0079] The memory stores a program that executes the technical solution of this invention, and may also store an operating system and other critical business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0080] The processor can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0081] Input devices may include devices that receive data and information input by the user, such as keyboards, mice, cameras, scanners, light pens, voice input devices, touch screens, pedometers, or gravity sensors.

[0082] Output devices may include devices that allow information to be output to the user, such as displays, printers, speakers, etc.

[0083] The communication interface may include any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0084] The processor executes the program stored in the memory and calls other devices, which can be used to implement the various steps of any of the window cleaning robot control methods provided in the above embodiments of this application.

[0085] The computing device may also include a display component and a voice component. The display component may be a liquid crystal display screen or an e-ink display screen. The input device of the computing device may be a touch layer covering the display component, or a button, trackball or touchpad set on the casing of the computing device, or an external keyboard, touchpad or mouse, etc.

[0086] Those skilled in the art will understand that Figure 5The structures shown are merely block diagrams of some structures related to the solutions in this specification and do not constitute a limitation on the computing devices on which the solutions in this specification are applied. Specific computing devices may include more or fewer components than those shown in the figures, or combine certain components, or have different component arrangements.

[0087] In addition to the methods and devices described above, the control method for the window cleaning robot provided in the embodiments of this specification can also be a computer program product, which includes a computer program that, when run by a processor, causes the processor to perform the steps in the control method for the window cleaning robot according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0088] The computer program product described herein can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments described herein. These programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0089] Furthermore, embodiments of this specification also provide a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor of the steps in the control methods of the window cleaning robot according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0090] Those skilled in the art will understand that all or part of the processes in the methods of 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 the embodiments provided in this specification can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may 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 data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0091] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0092] The embodiments described above are merely illustrative of several implementation methods outlined in this specification. While the descriptions are specific and detailed, they should not be construed as limiting the scope of the solutions provided in this specification. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this specification, and these all fall within the scope of protection of this specification. Therefore, the scope of protection for this patent should be determined by the appended claims.

Claims

1. A control method for a window cleaning robot, characterized in that, include: During the operation of the window cleaning robot, environmental detection information of the surface on which it operates is collected through visual sensors; The distance parameter between the window cleaning robot and the edge of the operating surface is determined based on the environmental detection information; If the distance parameter meets the distance condition, the window cleaning robot is controlled to perform a turning operation, and the ball head through-beam sensor configured on the window cleaning robot is activated; By collecting the trigger status signal of the ball-head through-beam sensor, the output of the risk status signal is determined. The risk status signal is used to control the window cleaning robot to perform a posture adjustment operation, which is used to prevent the window cleaning robot from falling off the running surface.

2. The method according to claim 1, characterized in that, The step of determining the distance parameter between the window cleaning robot and the edge of the operating surface based on the environmental detection information includes: The detection parameters of the window cleaning robot at the edge of the operating surface are determined based on the environmental detection information. Obtain the operating map of the window cleaning robot; The map parameters are determined based on the operating map and the location information corresponding to the window cleaning robot. The detection parameters are verified based on the map parameters to determine the distance parameters.

3. The method according to claim 1, characterized in that, If the distance parameter meets the distance condition, the window cleaning robot is controlled to perform a turning operation, and the ball-head through-beam sensor configured on the window cleaning robot is activated, including: Determine the distance threshold indicated by the distance condition; When the distance parameter approaches the distance threshold, the window cleaning robot is controlled to perform a deceleration operation; When the distance parameter is equal to or less than the distance threshold, the distance parameter is indicated to meet the distance condition, the window cleaning robot is controlled to perform a turning operation, and the ball head photoelectric sensor configured on the window cleaning robot is activated.

4. The method according to claim 3, characterized in that, Determining the distance threshold indicated by the distance condition includes: Determine the preset threshold value indicated by the distance condition; The thickness information of the operating surface is obtained based on a vision sensor; An adjustment factor is configured based on the thickness information, wherein the thickness information is negatively correlated with the adjustment factor; The preset threshold is adjusted by the adjustment factor to obtain the distance threshold.

5. The method according to claim 1, characterized in that, The step of determining the output of a risk status signal by acquiring the trigger status signal of the ball-head through-beam sensor includes: Acquire the trigger status signal of the ball head through-beam sensor; In response to the switching process of the trigger state signal, the attitude angle signal is acquired by the inertial measurement unit. The ball head through-beam sensor includes an elastic contact rod and an infrared through-beam tube. The switching process is triggered when the elastic contact rod is pushed by an obstacle. After the elastic contact rod is pushed, it blocks the optical path between the infrared through-beam tubes. The optical path between the infrared through-beam tubes is opened after the elastic contact rod passes the obstacle. The attitude angle signal is used to indicate the tilt angle of the window cleaning robot. If the tilt angle of the machine exceeds a preset angle threshold, a risk status signal is output.

6. The method according to any one of claims 1-5, characterized in that, The method further includes: Perform time-series statistics on the distance parameters to obtain information on their changes; If the change information indicates that the window cleaning robot is moving away from the surface edge of the operating surface, then the ball head through-beam sensor is turned off and the drive parameters of the window cleaning robot are adjusted.

7. The method according to any one of claims 1-5, characterized in that, The method further includes: In response to the triggering of the distance condition, the window cleaning robot performs path planning according to the edge distribution based on the operation map configured by the robot to obtain edge paths and non-edge paths; The window cleaning robot is controlled to clean along the edge path; After the edge path is completed, the cleaning task corresponding to the non-edge path is executed. The driving parameters of the edge path are different from those of the non-edge path.

8. A control device for a window cleaning robot, characterized in that, include: The acquisition unit is used to collect environmental detection information of the operating surface through a vision sensor during the operation of the window cleaning robot; The control unit is used to determine the distance parameter between the window cleaning robot and the edge of the operating surface based on the environmental detection information. The control unit is further configured to control the window cleaning robot to perform a turning operation and activate the ball head through-beam sensor configured on the window cleaning robot if the distance parameter meets the distance condition. The control unit is also used to determine the output of a risk status signal by acquiring the trigger status signal of the ball head through-beam sensor. The risk status signal is used to control the window cleaning robot to perform a posture adjustment operation, which is used to prevent the window cleaning robot from falling off the running surface.

9. A cleaning device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the control method of the window cleaning robot according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions for causing a computer to perform the method according to any one of claims 1 to 7.

Images