Swimming pool cleaning robot system
By employing dual-modal perception technology combining vision and ultrasound, the pool cleaning robot system achieves high-precision automatic recharging in complex environments, solving the problems of inaccurate and abnormal recharging in existing technologies, and possessing adaptive and self-learning capabilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-07
- Publication Date
- 2026-03-13
AI Technical Summary
Existing pool cleaning robots lack accuracy and the ability to detect abnormal conditions during automatic recharging, especially at night or in turbid water conditions, which can easily lead to recharging failure.
Employing dual-modal perception technology combining vision and ultrasound, along with an image acquisition system and a sound wave acquisition device, the robot monitors the optical visual characteristics and ultrasonic signals of the charging pile in real time. Through a collaborative strategy of visual coarse alignment and ultrasonic fine docking, the robot achieves automatic recharging.
It improves the accuracy and reliability of automatic recharging, can successfully connect to charging piles in various complex environments, has self-learning and fault handling capabilities, and achieves high-precision recharging in all weather conditions.
Smart Images
Figure CN121654271A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of swimming pool cleaning equipment technology, and in particular to a swimming pool cleaning robot system. Background Technology
[0002] Most pool cleaning robots on the market currently rely on visual recognition to return to their charging docks. However, this method is prone to recognition failure at night, in obstructed conditions, or in murky water. Some systems incorporate ultrasonic guidance, but the propagation path of ultrasonic waves in water is easily interfered with and has limited accuracy, which may cause errors in returning to the dock or even prevent the robot from successfully aligning with the charging dock. In addition, most current robots lack the ability to sense abnormal conditions such as charging dock failure or power outages, and are prone to getting stuck in an abnormal state of charging failure or waiting in place.
[0003] In summary, existing pool robots suffer from inaccurate and abnormal automatic recharging issues. Summary of the Invention
[0004] Therefore, the technical problem to be solved by the present invention is to overcome the problems of inaccurate and abnormal automatic recharging of pool robots in the prior art.
[0005] To address the aforementioned technical problems, this invention provides a swimming pool cleaning robot system, comprising: A charging station, wherein at least part of the charging station is located below the water surface; The main body of the pool cleaning robot is equipped with an image acquisition system, a sound wave acquisition device, a walking system, and a water pump system. The control unit for the pool cleaning robot includes: The real-time data acquisition module is used to acquire image data containing the optical visual features of the charging pile and ultrasonic signals emitted by the charging pile in real time, which are monitored by the image acquisition system and the sound wave acquisition device when the pool cleaning robot is below the charging pile and begins to climb the wall during the return process to the charging pile. The charging docking module is used to determine the movement path based on real-time ultrasonic signals and / or real-time image data, start the walking system and water pump system to perform movement, and make dynamic path adjustments according to real-time ultrasonic signals and / or real-time image data to complete the charging docking.
[0006] Preferably, during the return process to the charging station, when the pool cleaning robot is below the charging station and begins to climb the wall, acquiring image data containing the optical visual features of the charging station and ultrasonic signals emitted by the charging station, which are monitored in real time by the image acquisition system and the sound wave acquisition device, includes: Upon receiving or detecting a low battery signal or a user-issued recharge command, the system switches from operating mode to recharge mode. The ultrasonic signals emitted by the charging pile are acquired in real time using an acoustic wave acquisition device, and image data containing the optical visual features of the charging pile are acquired in real time using an image acquisition system on the top or front of the robot.
[0007] Preferably, the step of determining the motion path based on real-time ultrasonic signals and / or real-time image data, activating the walking system and water pump system to perform the motion, and dynamically adjusting the path according to real-time ultrasonic signals and / or real-time image data to complete the charging docking includes: The effectiveness of the real-time ultrasonic signal is analyzed. If the signal is effective, the ultrasonic re-pile mode is entered; otherwise, the visual re-pile mode is switched. Based on the initial pile return mode decision, an initial movement path is output, the walking system and water pump system are started to execute the initial movement path, and a real-time pile return mode decision is made based on real-time ultrasonic signals and / or real-time image data. Based on the real-time pile return mode decision, dynamic path adjustment is performed to complete the charging docking.
[0008] Preferably, the step of performing an effectiveness analysis on the real-time ultrasonic signal, and entering the ultrasonic re-tracking mode if the signal is valid, and switching to the visual re-tracking mode otherwise, includes: Determine whether the real-time ultrasonic signal has a signal of the expected frequency. If it does, analyze whether the signal strength of the real-time ultrasonic signal is higher than the first preset threshold, whether the signal-to-noise ratio is higher than the second preset threshold, and whether the signal stability is within the preset tolerance range. If all conditions are met, the real-time ultrasonic signal is determined to be valid. If the real-time ultrasonic signal is valid, enter ultrasonic pile return mode; otherwise, switch to visual pile return mode.
[0009] Preferably, the initial motion path output based on the initial pile-back mode decision includes: When the initial charging station return mode is determined to be ultrasonic charging station return mode, the direction angle and distance of the charging station relative to the robot are estimated based on the characteristics of ultrasonic signals, and an initial motion path is generated to make the robot move towards the charging station. When the initial charging station return mode is determined to be visual charging station return mode, the optical visual features of the charging station in the image data are identified. Based on the pixel position and size of the optical visual features of the charging station in the image, the relative direction and distance of the charging station are estimated, and an initial motion path is generated to correct the directional deviation and align the robot with the charging station.
[0010] Preferably, when the real-time charging station return mode decision is the visual charging station return mode, if there is no charging station information in the real-time image data, a preset action is executed to autonomously find the location of the charging station. The preset action includes, but is not limited to: random walking, cross walking, walking along the pool wall, climbing the wall to find the charging station. When the charging station information appears in the real-time image data, the action of moving towards the charging station is executed.
[0011] Preferably, the dynamic path adjustment based on real-time backtracking mode decision includes: When the real-time pile return mode is determined to be ultrasonic pile return mode, dynamic path adjustment is performed based on real-time ultrasonic signals, and dynamic path correction is performed based on real-time image data. When the real-time backfilling mode decision is set to visual backfilling mode, dynamic path adjustment is performed based on real-time image data, and dynamic path correction is performed based on real-time ultrasonic signals.
[0012] Preferably, the dynamic path adjustment based on real-time backtracking mode decision-making further includes: If the real-time docking mode decision is set to visual docking mode, and the distance between the robot and the charging pile meets the preset range, then switch to ultrasonic docking mode for precise docking.
[0013] Preferably, in the visual docking mode, and when the distance between the robot and the charging station meets a preset range, switching to the ultrasonic docking mode for precise docking includes: In visual charging docking mode, the system identifies the optical visual features of the charging dock in the image data. When the optical visual features of the charging dock enter the preset area in the center of the image, it switches to ultrasonic charging docking mode. Based on the arrival time difference or signal intensity gradient of the ultrasonic signal, the system estimates the direction angle and distance of the charging dock relative to the robot through triangulation or orientation algorithms, and accurately docks the robot's charging contacts with the charging dock.
[0014] Preferably, the charging docking includes: After successful physical docking, a charging handshake is performed to confirm charging. In the event of charging failure, the robot is controlled to detach from the charging station and re-enter the recharging mode.
[0015] Preferably, the step of performing a charging handshake confirmation after successful physical docking, and controlling the robot to detach from the charging station and re-enter the recharging mode in the event of charging failure, includes: After successful physical docking, the charging contacts communicate with the charging pile to detect the charging current and voltage. If no charging current is detected within a preset time or if charging is unexpectedly interrupted, the charging is deemed to have failed, and the robot is controlled to detach from the charging station and re-enter the recharge mode.
[0016] Preferably, the pool cleaning robot control device further includes: The parameter adjustment module is used to adjust the internal parameters of the ultrasonic positioning algorithm based on the error value by comparing the location of the charging pile sensed by the image acquisition system with the angle of the charging pile sensed by the acoustic acquisition device.
[0017] Preferably, the pool cleaning robot system further includes: a cleaning module and an integrated battery, and the charging pile includes a power interface, an infrared / LED lighting device and a controllable ultrasonic transmitter.
[0018] The technical solution of the present invention has the following advantages compared with the prior art: The swimming pool cleaning robot system described in this invention, by integrating visual and ultrasonic dual-modal perception technologies, endows the robot with strong environmental adaptability. Based on real-time analysis of ultrasonic signal strength, the robot intelligently judges the charging station's operating status and signal reliability. Therefore, in well-lit and clear water conditions, it prioritizes ultrasonic guidance for recharging, which has strong anti-interference capabilities. At night, in murky water, or when ultrasonic signals fail, it seamlessly switches to visual navigation mode, ensuring the success rate and reliability of the recharging process in various complex swimming pool environments. Furthermore, when approaching the charging station, the system innovatively employs a collaborative strategy of "visual coarse alignment + ultrasonic fine docking." Visual recognition is used for general positioning, followed by ultrasonic fine-tuning at the millimeter level, greatly improving the docking accuracy and success rate of the charging interface. In addition, the system possesses intelligent fault handling and self-learning capabilities. It can automatically retreat and restart the recharging process after a charging failure, and can also use successful visual positioning data to perform online correction of ultrasonic parameters, continuously optimizing subsequent recharging performance. Ultimately, this achieves a highly robust, high-precision, and all-weather fully automated recharging solution. Attached Figure Description
[0019] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein: Figure 1 This is a schematic diagram of the structure of a swimming pool cleaning robot system provided by the present invention. Detailed Implementation
[0020] The core of this invention is to provide a swimming pool cleaning robot system that effectively improves the accuracy of automatic recharging and realizes anomaly detection and post-processing.
[0021] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0022] Please refer to Figure 1. Figure 1 This is a schematic diagram of the structure of a swimming pool cleaning robot system provided by the present invention; specifically including: A charging station, wherein at least part of the charging station is located below the water surface; The main body of the pool cleaning robot is equipped with an image acquisition system, a sound wave acquisition device, a walking system, and a water pump system. The control unit for the pool cleaning robot includes: The real-time data acquisition module is used to acquire image data containing the optical visual features of the charging pile and ultrasonic signals emitted by the charging pile in real time, which are monitored by the image acquisition system and the sound wave acquisition device when the pool cleaning robot is below the charging pile and begins to climb the wall during the return process to the charging pile. The charging docking module is used to determine the movement path based on real-time ultrasonic signals and / or real-time image data, start the walking system and water pump system to perform movement, and make dynamic path adjustments according to real-time ultrasonic signals and / or real-time image data to complete the charging docking.
[0023] It should be noted that at least part of the charging station is located below the water surface. The charging station includes a sound wave emitting device, which is located below the water surface because sound wave transmission is better below the water surface. It should be noted that the image acquisition system can simultaneously acquire partial image information of the charging pile and image information of the pool wall. This is because visually distinguishing the edge information of the charging pile to identify its angle, or visually recognizing visual markings on the charging pile to identify its angle, are both possible solutions.
[0024] It should be noted that the ultrasonic wave acquisition device of the pool robot and the ultrasonic wave transmission device of the charging pile are not obstructed by the charging pile or other components of the robot. In this way, the ultrasonic signal received by the robot propagates in a straight line without reflection or diffraction, which can obtain more effective ultrasonic wave characteristics.
[0025] Based on the above embodiments, this embodiment provides a detailed description of the real-time data acquisition module: In some embodiments, during the return to the charging station, when the pool cleaning robot is below the charging station and begins to climb the wall, acquiring image data containing the optical visual features of the charging station and ultrasonic signals emitted by the charging station, which are monitored in real time by an image acquisition system and an acoustic wave acquisition device, includes: Upon receiving or detecting a low battery signal or a user-issued recharge command, the system switches from operating mode to recharge mode. The ultrasonic signals emitted by the charging pile are acquired in real time using an acoustic wave acquisition device, and image data containing the optical visual features of the charging pile are acquired in real time using an image acquisition system on the top or front of the robot.
[0026] It's important to note that the robot is not always in a recharge-ready state; it only actively switches from "working mode" (usually cleaning mode) to "recharge mode" after receiving a clear trigger signal. The trigger signal comes from two sources: Automatic system response (low battery signal): When the robot's power management system detects that the battery level is below a preset threshold, it automatically generates a recharge request to ensure that the robot can autonomously return to its charging station before running out of power. User command: Users can manually issue a recharge command via remote control, mobile app, or buttons on the robot to actively recall it.
[0027] It should be noted that upon entering recharging mode, the robot immediately activates two sets of sensor systems in parallel: The system uses acoustic wave acquisition devices (such as hydrophone arrays) to capture ultrasonic signals of specific frequencies (e.g., 40-60kHz) continuously emitted by the charging pile in real time. This system is mainly used to detect the location and distance of the charging pile and to make a preliminary judgment on its working status.
[0028] The system uses an image acquisition system (such as a high-definition camera) mounted on the top or front of the robot to capture real-time video streams of the environment, extracting image data containing optical visual features of the charging stations (such as LED light strips, infrared patterns, and specific markings). This system is primarily used to identify the appearance and location of the charging stations.
[0029] Based on the above embodiments, this embodiment provides a detailed description of the charging docking module: In some embodiments, determining the motion path based on real-time ultrasonic signals and / or real-time image data, activating the walking system and water pump system to perform motion, and dynamically adjusting the path according to real-time ultrasonic signals and / or real-time image data to complete the charging docking includes: The effectiveness of the real-time ultrasonic signal is analyzed. If the signal is effective, the ultrasonic re-pile mode is entered; otherwise, the visual re-pile mode is switched. Based on the initial pile return mode decision, an initial movement path is output, the walking system and water pump system are started to execute the initial movement path, and a real-time pile return mode decision is made based on real-time ultrasonic signals and / or real-time image data. Based on the real-time pile return mode decision, dynamic path adjustment is performed to complete the charging docking.
[0030] In some embodiments, performing an effectiveness analysis on the real-time ultrasonic signal, and if effective, entering the ultrasonic re-tracking mode; otherwise, switching to the visual re-tracking mode, includes: Determine whether the real-time ultrasonic signal has a signal of the expected frequency. If it does, analyze whether the signal strength of the real-time ultrasonic signal is higher than the first preset threshold, whether the signal-to-noise ratio is higher than the second preset threshold, and whether the signal stability is within the preset tolerance range. If all conditions are met, the real-time ultrasonic signal is determined to be valid. If the real-time ultrasonic signal is valid, enter ultrasonic pile return mode; otherwise, switch to visual pile return mode.
[0031] It should be noted that the raw signal received by the sound wave acquisition device is first subjected to a Fast Fourier Transform (FFT) or bandpass filter to analyze its frequency components. The purpose is to look for the presence of a signal at the specific frequency (e.g., 40-60kHz) claimed to be emitted by the charging pile in the background noise. This is a preliminary screening; if even the expected frequency signal is not present, the signal is immediately determined to be invalid.
[0032] It should be noted that measuring the amplitude (intensity) of the received signal and determining whether it exceeds the first preset threshold is crucial. If the signal strength is too low, it indicates that the robot may be too far from the charging station or that the signal propagation path is severely obstructed. Even if the robot can identify the direction, the error may be large, making it insufficient to reliably guide the robot back to the charging station. Calculating the ratio of signal power to noise power and determining whether it exceeds the second preset threshold is also important. A high signal-to-noise ratio is fundamental for reliable demodulation and signal analysis. A low signal-to-noise ratio means that the signal is submerged in environmental noise, resulting in extremely inaccurate direction estimation and potential robot misjudgment. Within a certain time window, monitoring the fluctuation of signal strength or frequency is essential to determine whether the fluctuation range is within the preset tolerance range. Complex underwater environments (such as surface fluctuations, bubbles, and other disturbances) can cause ultrasonic signals to be reflected and scattered, resulting in drastic changes in the received signal strength. An unstable signal is also unreliable.
[0033] It should be noted that if all three conditions are met, the real-time ultrasonic signal is deemed valid, and the control system decides to enter the ultrasonic pile return mode. If any of the above conditions are not met, the real-time ultrasonic signal is deemed invalid, and the control system automatically switches to the visual pile return mode.
[0034] In some embodiments, based on the initial pile-back mode decision, the output initial motion path includes: When the initial charging station return mode is determined to be ultrasonic charging station return mode, the direction angle and distance of the charging station relative to the robot are estimated based on the characteristics of ultrasonic signals, and an initial motion path is generated to make the robot move towards the charging station. When the initial charging station return mode is determined to be visual charging station return mode, the optical visual features of the charging station in the image data are identified. Based on the pixel position and size of the optical visual features of the charging station in the image, the relative direction and distance of the charging station are estimated, and an initial motion path is generated to correct the directional deviation and align the robot with the charging station.
[0035] It should be noted that by analyzing the minute time differences in the arrival of ultrasonic signals at different receivers or the intensity gradients of received signals in different directions, triangulation (calculating distance and angle through time differences) or orientation algorithms (such as beamforming, calculating the direction of arrival through signal phase differences) are used to calculate the approximate direction angle and distance of the charging station relative to the robot's current coordinate system. Based on this orientation information, a simple directional path is generated, for example: "Move a distance D in the direction of angle θ, with the current point as the origin." This path is usually a straight line or a large-curvature arc, with the aim of aligning the robot's approximate front end with the charging station.
[0036] It should be noted that the image data is processed to identify the optical visual features of the charging pile (such as specific luminous patterns, LED array shapes, or QR codes). The relative orientation (left / right, deflection angle) of the charging pile is calculated based on the pixel positions of these features in the image (e.g., the number of horizontal and vertical pixels offset from the image center). Simultaneously, a rough distance to the charging pile is estimated based on the size of the features (e.g., the pixel area occupied by the pattern). Based on this pose estimation, a correction path is generated, for example: "Rotate α degrees to the right to eliminate center offset, then advance β meters." The core purpose of this path is to correct the relative pose deviation between the robot and the charging pile, allowing it to gradually align with the charging pile from any angle.
[0037] In some embodiments, when the real-time charging station return mode decision is a visual charging station return mode, if there is no charging station information in the real-time image data, a preset action is performed to autonomously find the location of the charging station. The preset action includes, but is not limited to: random walking, cross walking, walking along the pool wall, climbing the wall to find the charging station. When the charging station information appears in the real-time image data, the action of moving towards the charging station is performed.
[0038] In some embodiments, activating the walking system and water pump system to execute an initial movement path, and making a real-time pile-back mode decision based on real-time ultrasonic signals and / or real-time image data includes: The locomotion system (such as a water pump, tires, or tracks) receives and begins executing the generated initial motion path. During movement, the image acquisition system and acoustic acquisition device do not stop working but continuously and in real-time acquire the latest environmental data (real-time ultrasonic signals and real-time image data) and continuously perform signal validity analysis (i.e., repeating the logic of "analyzing the validity of real-time ultrasonic signals; if valid, entering ultrasonic backtracking mode; otherwise, switching to visual backtracking mode"). For example, when moving in visual mode, if a stable and valid ultrasonic signal is suddenly received, the system may immediately switch back to the better ultrasonic mode; conversely, if the signal suddenly fails due to interference in ultrasonic mode, the system will decisively switch to visual mode.
[0039] It should be noted that even if the mode remains unchanged, the system will continuously update its estimation of the charging station's location based on the latest sensor data and dynamically adjust its movement path accordingly. For example, in ultrasonic mode, it accurately calculates the remaining distance based on the continuous increase in signal strength and controls deceleration. In visual mode, it fine-tunes its direction of travel by observing feature points gradually moving towards the image center.
[0040] In some embodiments, dynamic path adjustment based on real-time back-piling mode decision-making includes: When the real-time pile return mode is determined to be ultrasonic pile return mode, dynamic path adjustment is performed based on real-time ultrasonic signals, and dynamic path correction is performed based on real-time image data. When the real-time backfilling mode decision is set to visual backfilling mode, dynamic path adjustment is performed based on real-time image data, and dynamic path correction is performed based on real-time ultrasonic signals.
[0041] It should be noted that when the real-time docking mode decision is ultrasonic docking mode, dynamic path adjustment is performed based on real-time ultrasonic signals. This dynamic path adjustment includes real-time correction of the robot's travel direction based on the intensity change trend, directional change characteristics, and signal continuity of the ultrasonic signals, ensuring that the robot approaches the charging pile along the ultrasonic signal guidance direction. Simultaneously, dynamic path correction is performed based on real-time image data. This dynamic path correction includes fine-tuning the robot's posture based on the positional offset, angle deviation, and feature clarity changes of the charging pile's visual features in the field of view in the image, to compensate for positioning errors of the ultrasonic signals at close range or in complex environments, thereby improving docking accuracy. It should be noted that when the real-time charging pile return mode decision is the visual charging pile return mode, dynamic path adjustment is performed based on real-time image data. The dynamic path adjustment includes real-time planning and correction of the robot's travel path based on the recognition results of the visual features of the charging pile in the image, the motion trajectory of the feature points in the image plane and the changes in their relative positions, ensuring that the robot moves towards the visual center area of the charging pile. At the same time, dynamic path correction is performed based on real-time ultrasonic signals. The dynamic path correction includes using the existence, directional stability and change trend of ultrasonic signals to compensate for and correct deviations caused by visual recognition errors or occlusions, thereby enhancing the robustness and reliability of path execution.
[0042] Based on the above embodiments, in some embodiments, when the distance between the robot and the charging pile meets a preset range in visual docking mode, the system switches to ultrasonic docking mode for precise docking, specifically including: In visual charging docking mode, the system identifies the optical visual features of the charging dock in the image data. When the optical visual features of the charging dock enter the preset area in the center of the image, it switches to ultrasonic charging docking mode. Based on the arrival time difference or signal intensity gradient of the ultrasonic signal, the system estimates the direction angle and distance of the charging dock relative to the robot through triangulation or orientation algorithms, and accurately docks the robot's charging contacts with the charging dock.
[0043] It should be noted that when the robot moves towards the estimated charging station location in visual charging station homing mode, it performs the following judgments for each frame of the image: confirming that the optical visual features of the charging station still exist in the current image and have been successfully identified, calculating the pixel coordinates of the feature in the image, and determining whether it falls within a predefined central area of the image (e.g., a circular or rectangular area in the center of the image, occupying about 10%-20% of the entire field of view); after confirming entry, it officially switches from visual charging station homing mode to ultrasonic charging station homing mode, and the robot's main navigation sensor changes from a camera to an ultrasonic receiver, while the vision system is downgraded to auxiliary monitoring or completely stops working, because at extremely close distances, the ultrasonic system begins to exert its advantages of shorter wavelength and strong directionality; again, through triangulation or orientation algorithms (such as beamforming), it calculates the extremely precise orientation and distance of the charging station interface relative to the robot's charging contacts, with an accuracy that can reach the millimeter level.
[0044] Based on the above embodiments, in some embodiments, the charging docking includes, after successful physical docking, performing a charging handshake confirmation; and in the event of charging failure, controlling the robot to detach from the charging station and re-enter the recharging mode, specifically including: After successful physical docking, the charging contacts communicate with the charging pile to detect the charging current and voltage. If no charging current is detected within a preset time or if charging is unexpectedly interrupted, the charging is deemed to have failed, and the robot is controlled to detach from the charging station and re-enter the recharge mode.
[0045] It should be noted that after the physical connection is established, the correct connection and good contact are confirmed by applying a specific identification voltage to the charging circuit or sending a digital handshake signal. After a successful handshake, the charging station begins to output charging voltage and current. If the charging current reaches the expected threshold range and the voltage stabilizes within a preset time (e.g., 2-5 seconds), charging is considered successful. If the connection fails, the robot's walking system moves backward a preset distance (e.g., 20-50 cm) and re-enters the recharging mode. After the backward movement is complete, the robot's state machine re-triggers the entire recharging process (i.e., starting from step S101). It will again use sensors to find the charging station, re-determine the mode, plan the path, and attempt docking. This process can be repeated several times until charging is successful or a final error is reported. Based on the above embodiments, in some embodiments, the pool cleaning robot control device further includes: The parameter adjustment module is used to adjust the internal parameters of the ultrasonic positioning algorithm based on the error value by comparing the location of the charging pile sensed by the image acquisition system with the angle of the charging pile sensed by the acoustic acquisition device.
[0046] This process is usually triggered the moment when the visual charging station mode successfully completes precise positioning. At this time, it is confirmed that the charging station feature is located in the center of the image, providing a position estimate that is considered very accurate and can be used as a "true value" or "reference value". Simultaneously, the robot position or target orientation estimated based on the current ultrasonic signal at this moment is recorded. Compare the two estimates and calculate the error between them. This error may include angular deviation and distance deviation.
[0047] Based on the calculated error value, specific algorithms (such as filtering algorithms, least squares methods, etc.) are used to adjust the internal parameters of the ultrasonic positioning algorithm in reverse. These parameters may include: wave velocity calibration value, sensor offset compensation, and algorithm gain or filtering coefficients. Some weighting values or filtering parameters in the positioning algorithm can be adjusted to optimize its output.
[0048] This invention provides a swimming pool cleaning robot, comprising a robot body and a charging station, wherein the robot body includes: Fuselage: The main structural component that carries all parts, and usually has a watertight design.
[0049] Locomotive system: The actuators that provide mobility, such as tracks, tires, or water pump propulsion, are responsible for executing movement commands issued by the control device.
[0050] Cleaning module: The core working unit, such as brush plate, filter screen, vacuum suction port, etc., is used to complete the pool cleaning task.
[0051] Integrated battery: The power source that powers the entire robot, and its low battery state is the main condition for triggering automatic recharging.
[0052] Image acquisition system: Installed on the top or front of the robot, usually a waterproof camera.
[0053] Acoustic wave acquisition device: A sensor used to receive ultrasonic signals emitted by the charging pile, such as an ultrasonic microphone or receiver array; Such as the control device for the pool cleaning robot mentioned above.
[0054] In some embodiments, the charging station includes: Power interface: Connects to the municipal power grid or low-voltage power supply to provide charging power for the robot.
[0055] Optical devices, such as LED light strips, infrared emitters, or other visual markers, are used to provide guidance features for the robot's image acquisition system.
[0056] Ultrasonic signal transmitting device: A transmitter that emits ultrasonic waves at a specific frequency (such as 40-60kHz) to provide acoustic guidance signals for the robot's sound wave acquisition device.
[0057] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.
Claims
1. A swimming pool cleaning robot system, characterized in that, include: A charging station, wherein at least part of the charging station is located below the water surface; The main body of the pool cleaning robot is equipped with an image acquisition system, a sound wave acquisition device, a walking system, and a water pump system. The control unit for the pool cleaning robot includes: The real-time data acquisition module is used to acquire image data containing the optical visual features of the charging pile and ultrasonic signals emitted by the charging pile in real time, which are monitored by the image acquisition system and the sound wave acquisition device when the pool cleaning robot is below the charging pile and begins to climb the wall during the return process to the charging pile. The charging docking module is used to determine the movement path based on real-time ultrasonic signals and / or real-time image data, start the walking system and water pump system to perform movement, and make dynamic path adjustments according to real-time ultrasonic signals and / or real-time image data to complete the charging docking.
2. The swimming pool cleaning robot system according to claim 1, characterized in that, During the return process to the charging station, when the pool cleaning robot is below the charging station and begins to climb the wall, the acquisition of image data containing the optical visual features of the charging station and ultrasonic signals emitted by the charging station, monitored in real time by an image acquisition system and an acoustic wave acquisition device, includes: Upon receiving or detecting a low battery signal or a user-issued recharge command, the system switches from operating mode to recharge mode. The ultrasonic signals emitted by the charging pile are acquired in real time using an acoustic wave acquisition device, and image data containing the optical visual features of the charging pile are acquired in real time using an image acquisition system on the top or front of the robot.
3. The swimming pool cleaning robot system according to claim 1, characterized in that, The process of determining the motion path based on real-time ultrasonic signals and / or real-time image data, activating the walking system and water pump system to execute the motion, and dynamically adjusting the path according to real-time ultrasonic signals and / or real-time image data to complete the charging docking includes: The effectiveness of the real-time ultrasonic signal is analyzed. If the signal is effective, the ultrasonic re-pile mode is entered; otherwise, the visual re-pile mode is switched. Based on the initial pile return mode decision, an initial movement path is output, the walking system and water pump system are started to execute the initial movement path, and a real-time pile return mode decision is made based on real-time ultrasonic signals and / or real-time image data. Based on the real-time pile return mode decision, dynamic path adjustment is performed to complete the charging docking.
4. The swimming pool cleaning robot system according to claim 3, characterized in that, The process of analyzing the effectiveness of real-time ultrasonic signals, and entering ultrasonic re-tracking mode if effective, and switching to visual re-tracking mode otherwise, includes: Determine whether the real-time ultrasonic signal has a signal of the expected frequency. If it does, analyze whether the signal strength of the real-time ultrasonic signal is higher than the first preset threshold, whether the signal-to-noise ratio is higher than the second preset threshold, and whether the signal stability is within the preset tolerance range. If all conditions are met, the real-time ultrasonic signal is determined to be valid. If the real-time ultrasonic signal is valid, enter ultrasonic pile return mode; otherwise, switch to visual pile return mode.
5. The swimming pool cleaning robot system according to claim 3, characterized in that, The initial motion path output based on the initial pile-back mode decision includes: When the initial charging station return mode is determined to be ultrasonic charging station return mode, the direction angle and distance of the charging station relative to the robot are estimated based on the characteristics of ultrasonic signals, and an initial motion path is generated to make the robot move towards the charging station. When the initial charging station return mode is determined to be visual charging station return mode, the optical visual features of the charging station in the image data are identified. Based on the pixel position and size of the optical visual features of the charging station in the image, the relative direction and distance of the charging station are estimated, and an initial motion path is generated to correct the directional deviation and align the robot with the charging station.
6. The swimming pool cleaning robot system according to claim 5, characterized in that, When the real-time charging station return mode is determined to be the visual charging station return mode, if there is no charging station information in the real-time image data, a preset action is executed to autonomously find the location of the charging station. The preset actions include, but are not limited to: random walking, cross walking, walking along the pool wall, and climbing the wall to find the charging station. When the charging station information appears in the real-time image data, the action of moving towards the charging station is executed.
7. The swimming pool cleaning robot system according to claim 3, characterized in that, The dynamic path adjustment based on real-time back-piling mode decision-making includes: When the real-time pile return mode is determined to be ultrasonic pile return mode, dynamic path adjustment is performed based on real-time ultrasonic signals, and dynamic path correction is performed based on real-time image data. When the real-time backfilling mode decision is set to visual backfilling mode, dynamic path adjustment is performed based on real-time image data, and dynamic path correction is performed based on real-time ultrasonic signals.
8. The swimming pool cleaning robot system according to claim 7, characterized in that, The dynamic path adjustment based on real-time back-piling mode decision-making also includes: If the real-time docking mode decision is set to visual docking mode, and the distance between the robot and the charging pile meets the preset range, then switch to ultrasonic docking mode for precise docking.
9. The swimming pool cleaning robot system according to claim 8, characterized in that, In the visual docking mode, and provided the distance between the robot and the charging station meets a preset range, switching to the ultrasonic docking mode for precise docking includes: In visual charging docking mode, the system identifies the optical visual features of the charging dock in the image data. When the optical visual features of the charging dock enter the preset area in the center of the image, it switches to ultrasonic charging docking mode. Based on the arrival time difference or signal intensity gradient of the ultrasonic signal, the system estimates the direction angle and distance of the charging dock relative to the robot through triangulation or orientation algorithms, and accurately docks the robot's charging contacts with the charging dock.
10. The swimming pool cleaning robot system according to claim 1, characterized in that, The charging docking includes: After successful physical docking, a charging handshake is performed to confirm charging. In the event of charging failure, the robot is controlled to detach from the charging station and re-enter the recharging mode.
11. The swimming pool cleaning robot system according to claim 10, characterized in that, After successful physical docking, a charging handshake is performed to confirm the connection. In the event of charging failure, the robot is controlled to detach from the charging station and re-enter the recharging mode. This includes: After successful physical docking, the charging contacts communicate with the charging pile to detect the charging current and voltage. If no charging current is detected within a preset time or if charging is unexpectedly interrupted, the charging is deemed to have failed, and the robot is controlled to detach from the charging station and re-enter the recharge mode.
12. The swimming pool cleaning robot system according to claim 1, characterized in that, The control device for the pool cleaning robot also includes: The parameter adjustment module is used to adjust the internal parameters of the ultrasonic positioning algorithm based on the error value by comparing the location of the charging pile sensed by the image acquisition system with the angle of the charging pile sensed by the acoustic acquisition device.
13. The swimming pool cleaning robot system according to claim 1, characterized in that, Also includes: The charging station includes a cleaning module, an integrated battery, a power interface, an infrared / LED lighting device, and a controllable ultrasonic transmitter.