Automatic blackboard erasing device and method

Abstract

The invention discloses an automatic blackboard groove cleaning device and method. The automatic blackboard groove cleaning device comprises a main machine, a power transmission system, a supporting walking mechanism and a multi-mode control module. A battery is arranged in the host, and a cleaning sponge is arranged below the host and is embedded into a blackboard groove; the power transmission system comprises a motor, a gear transmission set and a driving wheel, the top of a motor shell is connected with the host through a connecting frame, the output end is connected with the driving wheel through the gear transmission set, and the driving wheel is attached to the outer side wall of the blackboard groove after installation. The supporting walking mechanism comprises a supporting arm and an unpowered auxiliary wheel, one end of the supporting arm is connected with the bottom of the motor shell, the other end of the supporting arm is connected with the unpowered auxiliary wheel, and the unpowered auxiliary wheel is attached to the wall below the blackboard groove after installation. The multi-mode control module is integrated on the host and comprises a timing control unit, a manual control unit and a foreign body sensing unit; during use, the motor drives the cleaning sponge to wipe in the groove back and forth. Manual low-efficiency operation is replaced, multi-mode intelligent cleaning is achieved, the cleaning precision and effect are improved, and the energy consumption and scene adaptability is optimized.

Description

Technical Field

[0001] This invention relates to the field of cleaning equipment technology, specifically to an automatic blackboard erasing device and method. Background Technology

[0002] In educational settings such as schools and training institutions, as well as in places where traditional chalk is used for writing, such as conference rooms and lecture halls, the blackboard tray serves as a major accumulation area for chalk dust, eraser scraps, and other debris. Cleaning and maintaining it is an indispensable part of daily management. However, for a long time, cleaning the blackboard tray has relied primarily on manual labor. Manually wiping the tray requires cleaning section by section, which, especially in frequently used areas, consumes a significant amount of time and manpower daily. During the short breaks after class, cleaning staff often cannot thoroughly clean the entire area, leading to the accumulation of debris and affecting the cleanliness of the teaching environment.

[0003] To address the aforementioned issues, some automatic cleaning devices have emerged on the market, but existing technologies still have significant limitations: Limited functionality and low level of intelligence: Most devices only have timed cleaning functions and cannot flexibly adjust the cleaning strategy according to the actual pollution of the blackboard tray. This results in problems of "over-cleaning" or "under-cleaning", which wastes energy and makes it difficult to guarantee the cleaning effect.

[0004] Lack of precise sensing ability: It cannot accurately identify the density distribution and material type of foreign objects in the blackboard groove, and uses a single speed and mode during the cleaning process, which is not effective for cleaning wet chalk dust.

[0005] Poor adaptability and complex maintenance: Some devices have unreasonable structural designs, which can easily interfere with the blackboard slot after installation. They are unstable during movement, and it is difficult to replace the cleaning parts after they are worn out, resulting in high maintenance costs.

[0006] Therefore, there is an urgent need for a blackboard groove cleaning device and method that can achieve automated and intelligent cleaning while taking into account cleaning effectiveness and efficiency. Summary of the Invention

[0007] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an automatic blackboard erasing device and method to replace inefficient manual operation, achieve multi-mode intelligent cleaning, improve cleaning accuracy and effect, optimize energy consumption and scene adaptability, and reduce maintenance costs.

[0008] To solve the above problems, the technical solution adopted by the present invention is as follows: This invention provides an automatic blackboard erasing tray device, comprising: Main unit: Built-in battery, with a cleaning sponge installed below, which is then embedded in the blackboard slot after installation; Power transmission system: includes motor, gear transmission assembly and drive wheel. The top of the motor housing is connected to the main unit through a connecting frame. The output end is connected to the drive wheel through the gear transmission assembly. After installation, the drive wheel fits against the outer wall of the blackboard groove. Support walking mechanism: includes support arm and non-powered auxiliary wheel. One end of the support arm is connected to the bottom of the motor housing, and the other end is connected to the non-powered auxiliary wheel. After installation, the non-powered auxiliary wheel is attached to the wall below the blackboard groove. Multi-mode control module: integrated into the host, containing three types of control units: Timer control unit: Built-in timing program can automatically trigger the motor to start at a preset time to complete the blackboard groove erasure; Manual control unit: The motor is started via a physical button, allowing users to operate it manually as needed; Foreign object detection unit: Equipped with an ultrasonic sensor, it can detect foreign objects in the blackboard groove in real time. After detecting foreign objects, it can start the motor to wipe them in a targeted manner according to a preset strategy. In use, the motor drives the drive wheel to reciprocate linearly along the outer wall of the blackboard groove via a gear transmission group, which in turn drives the cleaning sponge to reciprocate synchronously within the groove, thus achieving automated wiping.

[0009] The present invention also provides an automatic blackboard erasing tray method, which uses the above-mentioned device and includes at least one of the following operating modes: Automatic timed wiping mode: The timed control unit automatically sends a start signal at a preset time to drive the cleaning sponge to wipe the blackboard grooves back and forth; Manually triggered wiping mode: Press the physical button to send a start signal. After receiving the signal, the multi-mode control module starts the motor, which drives the cleaning sponge to reciprocate. The machine will automatically stop after the set time is reached. Foreign object sensing wiping mode: The ultrasonic sensor monitors the blackboard groove in real time. When a foreign object is detected, a signal is sent to the multi-mode control module, which starts the motor to drive the cleaning sponge to wipe the area with the foreign object. The machine stops automatically after the object is removed.

[0010] Preferably, when the timer control unit automatically sends a start signal at a preset time, it includes: Five minutes before the scheduled start time, the entire blackboard slot is scanned using an ultrasonic sensor. If the amount of foreign matter accumulated in all areas is below the preset threshold, the start time is canceled; if the amount of foreign matter accumulated in any area is above the preset threshold, a start signal is sent.

[0011] Preferably, when driving the cleaning sponge to reciprocate and wipe the blackboard groove, a dynamic adjustment strategy for the movement speed and stroke is implemented, including: Zoned adaptive speed control: Based on the intensity of ultrasonic echo signals, the blackboard groove is divided into high, medium and low density foreign matter zones; the high density zone controls the cleaning sponge to reciprocate at a low speed of 1.5-2cm / s, the medium density zone wipes at a medium speed of 3-4cm / s, and the low density zone wipes at a fast speed of 6-8cm / s. Material-sensing intelligent speed regulation: By analyzing the ultrasonic echo frequency characteristics through a foreign material identification model, dry chalk dust, wet chalk dust, and eraser debris are distinguished; for wet chalk dust, the speed is reduced to below 1cm / s and two reverse strokes are added; for dry chalk dust and eraser debris, the speed is adjusted in different zones.

[0012] Preferably, when dividing the blackboard groove into high, medium, and low density foreign object areas, it includes: Before each division, the clean reference area of ​​the blackboard groove is scanned using an ultrasonic sensor to obtain the reference intensity of the echo signal; the echo intensity and attenuation rate of each area are extracted, and the area is divided according to the following criteria: High-density area: echo intensity > 80% of the reference, and attenuation rate ≥ 0.5dB / ms; Medium density region: echo intensity is 40%-80% of the baseline, and attenuation rate is 0.2-0.5 dB / ms; Low-density region: echo intensity < 40% of reference, and attenuation rate < 0.2dB / ms.

[0013] Preferably, when analyzing the frequency characteristics of ultrasonic echoes to distinguish between dry chalk dust, wet chalk dust, and eraser debris, the method includes: The ultrasonic sensor is controlled to emit a probe wave at 40kHz, and the raw frequency data of the echo is simultaneously acquired at a sampling frequency of ≥100kHz to capture the frequency shift caused by the material. Adaptive wavelet threshold filtering is applied to the echo data to extract three core feature parameters: frequency peak, fluctuation amplitude, and frequency offset. The three core feature parameters are input into the foreign object material recognition model and matched with the three built-in standard templates: Dry chalk dust: peak frequency 38-39kHz, fluctuation range <5%, frequency offset <1kHz; Wet chalk dust: peak frequency 35-37kHz, fluctuation range 10%-15%, frequency offset 2-3kHz; Board rubbing debris: The frequency peak is randomly distributed in the range of 36-40kHz, with fluctuation amplitude >20% and frequency offset >3kHz; If the matching degree is ≥85%, the material type is output directly; if it is 60%-85%, secondary sampling is started; if it is <60%, it is determined that there are no foreign objects.

[0014] Preferably, when performing adaptive wavelet threshold filtering on echo data, the following steps are included: Echo data is processed at multiple scales using a three-level wavelet decomposition structure: Low-frequency approximation layer: Soft threshold filtering is used to suppress low-frequency noise; First high-frequency detail layer: Hard threshold filtering is used to preserve the high-frequency feature differences between chalk dust and eraser debris; The second high-frequency detail layer: Adaptive sliding window threshold filtering is used to remove random impulse noise, and the window size is dynamically adjusted according to the signal energy; the noise-reduced data is generated by wavelet reconstruction.

[0015] Preferably, when performing adaptive wavelet threshold filtering on the echo data, the method further includes: For the echo data in the edge region of the blackboard groove, edge enhancement filtering is used: Automatically reduce the wavelet threshold in edge regions by 25% to prevent edge reflection signals from being misjudged as noise; The Sobel operator is introduced to perform secondary processing on the filtered data, thereby enhancing the abrupt change points of edge frequency characteristics.

[0016] Preferably, when monitoring the blackboard groove in real time using an ultrasonic sensor, the following is included: The blackboard groove is divided into several equally long detection units along its length. The ultrasonic sensors are controlled to conduct directional detection sequentially along a preset path. A signal processing time is reserved after each detection unit is completed, so as to achieve full coverage without blind spots.

[0017] Preferably, when the motor drives the cleaning sponge to reciprocate wiping the area containing foreign objects, the process includes: If the amount of foreign matter accumulated in all detection units is below the preset threshold, the wiping operation will not be performed. If the amount of foreign matter accumulated in any detection unit exceeds the preset threshold, the strategy of dynamically adjusting the reciprocating speed and wiping stroke of the cleaning sponge is executed to ensure that stubborn chalk dust is completely removed. If the amount of foreign matter accumulated in all detection units is below the preset threshold, wiping is not performed; if the amount of foreign matter accumulated in any detection unit is above the preset threshold, the dynamic adjustment strategy for movement speed and stroke is executed.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: Replaces inefficient manual operations: It solves the pain point of traditional blackboard trays relying on manual wiping section by section. Especially in high-frequency use scenarios such as schools and training institutions, it can effectively reduce the time spent cleaning after class, significantly reduce the workload of cleaning staff, and improve the cleanliness and management efficiency of teaching venues.

[0019] Multiple modes cover multiple scenario requirements: Automatic timed mode: Supports automatic start at preset time periods (such as after class each day) to achieve unattended routine cleaning and adapt to the needs of fixed work and rest scenarios.

[0020] Manual trigger mode: One-click activation via physical button, capable of responding to sudden large-scale dirt accumulation (such as large spills and piles of chalk dust) with flexible and efficient response.

[0021] Foreign object detection mode: The ultrasonic sensor monitors in real time and automatically starts targeted wiping when a foreign object is detected, achieving precise response and avoiding the accumulation and hardening of stains.

[0022] Zone-based adaptive speed adjustment: The ultrasonic echo intensity is used to precisely divide the foreign matter areas into high-density, medium-density, and low-density areas, and a differentiated wiping speed of 1.5-8cm / s is matched for different areas: the low-speed reciprocating motion in the high-density area ensures the removal of stubborn stains, while the fast wiping in the low-density area improves efficiency, balancing cleaning effect and energy consumption.

[0023] Material-sensing precision cleaning: Based on the ultrasonic frequency characteristics, it identifies three types of foreign objects: dry chalk dust, wet chalk dust, and eraser debris. For sticky stains such as wet chalk dust, it automatically reduces the speed to below 1cm / s and increases the reverse stroke, completely solving the problem of traditional devices not cleaning sticky stains thoroughly; for dry dust and debris, it maintains a high-efficiency speed to avoid dust flying.

[0024] Full-area blind-spot-free detection: Segmented scanning combined with edge enhancement filtering technology solves the industry pain point that signals in the edge area of ​​the blackboard groove are easily misjudged, achieving accurate detection with full-area coverage and eliminating edge omissions.

[0025] On-demand startup reduces unnecessary operation: The blackboard slot is pre-scanned 5 minutes before the scheduled startup, and the startup is canceled if the amount of foreign matter accumulation is below the threshold; the foreign matter sensing mode only works when stains are detected, which reduces the unnecessary operation time by about 60% compared with traditional timer devices, and significantly reduces energy consumption.

[0026] High adaptability to different scenarios: The structural design of the drive wheels fitting into the outer wall of the blackboard slot and the auxiliary wheels supporting the wall ensures that there is no interference with the blackboard slot after installation. It is compatible with more than 90% of traditional blackboard slot sizes and does not require modification of the existing site.

[0027] Modular and easy to maintain: The main unit, power system and control module adopt independent modular design. The cleaning sponge can be replaced by hand after it wears out. When core components such as motors and sensors fail, they can be quickly disassembled and repaired. The maintenance operation threshold is low and the cost is low.

[0028] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0029] Figure 1 This is a diagram of an automatic blackboard erasing slot device according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the automatic sending of the start signal according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating the strategy for dynamically adjusting the reciprocating speed and wiping stroke of the cleaning sponge according to an embodiment of the present invention. Figure 4 This is a flowchart of the adaptive wavelet threshold filtering process according to an embodiment of the present invention.

[0030] The following are the reference numerals: 1. Main unit; 11. Timer control unit; 12. Manual control unit; 13. Foreign object sensing unit; 2. Cleaning sponge; 3. Motor; 4. Gear transmission group; 5. Drive wheel; 6. Connecting frame; 7. Support arm; 8. Non-powered auxiliary wheel. Detailed Implementation

[0031] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0032] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0033] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the scope of this application and its application or use.

[0034] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0035] Example 1, see Figure 1 The present invention provides an automatic blackboard erasing slot device as shown in the figure. Figure 1 The automatic blackboard erasing slot device shown includes: a main unit 1, a power transmission system, a support and walking mechanism, and a multi-mode control module.

[0036] The main unit 1 has a built-in battery, and a cleaning sponge 2 is installed below it. After installation, the cleaning sponge 2 is embedded in the blackboard slot. The power transmission system is electrically connected to the host 1, including a motor 3, a gear transmission group 4 and a drive wheel 5; the top of the motor 3 housing is fixedly connected to the host 1 through a connecting bracket 6, the output end is connected to one end of the gear transmission group 4, and the other end of the gear transmission group 4 is connected to the drive wheel 5. After installation, the drive wheel 5 fits against the outer wall of the blackboard groove. The supporting walking mechanism includes a support arm 7 and a non-powered auxiliary wheel 8. One end of the support arm 7 is connected to the bottom of the motor 3 housing, and the other end is connected to the non-powered auxiliary wheel 8. After installation, the non-powered auxiliary wheel 8 is attached to the wall below the blackboard groove to support the whole device and reduce moving friction, ensuring the stability of the reciprocating linear motion of the drive wheel 5 and the cleaning sponge 2. The multi-mode control module is integrated into the main unit 1 and includes three control units: a timing control unit 11, a manual control unit 12, and a foreign object sensing unit 13, wherein: The timing control unit 11 has a built-in timing program that can automatically trigger the motor 3 to start at a preset time node (such as when class ends) to complete the blackboard sill erasing operation; The manual control unit 12 triggers the motor 3 to start via a physical button, allowing users to manually start the wiping operation as needed. The foreign object sensing unit 13 is equipped with an ultrasonic sensor to detect foreign objects in the blackboard groove in real time. After detecting foreign objects, the motor 3 is started to wipe them according to a preset strategy.

[0037] When this device is in use, the motor 3 drives the drive wheel 5 through the gear transmission group 4, so that the drive wheel 5 makes a reciprocating linear motion on the outer wall of the blackboard groove, which in turn drives the cleaning sponge 2 to reciprocate linear motion in the blackboard groove, thereby realizing automated wiping.

[0038] In this embodiment of the invention, it should be further explained that the automatic blackboard trough erasing device is an automated device that integrates mechanical transmission, intelligent control and cleaning functions. It mainly consists of four parts: main unit 1, power transmission system, support walking mechanism and multi-mode control module. Through the coordinated work of each module, the automatic and intelligent cleaning of the blackboard trough is realized, replacing manual wiping and improving cleaning efficiency and convenience.

[0039] The main unit 1 serves as the core carrier and energy supply center of the device, with a built-in rechargeable battery providing power support for the entire device's operation. Below, a cleaning sponge 2 is installed. After installation, the sponge embeds itself inside the blackboard slot, directly contacting chalk dust, debris, and other foreign objects within the slot, achieving physical wiping and cleaning through reciprocating motion.

[0040] Powertrain System: Power source: Micro motor 3 provides the power output for the operation of the device.

[0041] Transmission mechanism: The output end of motor 3 is connected to gear transmission group 4. Through the meshing transmission of gears, the rotational power of motor 3 is converted into the linear motion power of drive wheel 5. At the same time, the deceleration characteristics of gear transmission group 4 are used to ensure the stability of drive wheel 5 movement and torque output.

[0042] Actuating component: The drive wheel 5 is installed against the outer wall of the blackboard groove. Driven by the motor 3 and the gear transmission group 4, it moves back and forth linearly along the outer wall of the blackboard groove, thereby driving the cleaning sponge 2 to move synchronously through the main unit 1.

[0043] Support walking mechanism: Support arm 7: One end is connected to the bottom of the motor 3 housing, and the other end is connected to the non-powered auxiliary wheel 8, which serves to fix the auxiliary wheel and transmit the support force.

[0044] Non-powered auxiliary wheel 8: After installation, it fits against the wall below the blackboard slot and rolls passively with the drive wheel 5 when the device moves. On the one hand, it provides support for the whole device and ensures that the drive wheel 5 and the blackboard slot wall, and the cleaning sponge 2 and the blackboard slot are in close contact. On the other hand, it converts sliding friction into rolling friction, reduces the moving resistance of the device, and ensures the stability of the reciprocating linear motion of the drive wheel 5 and the cleaning sponge 2.

[0045] Multi-mode control module: Integrated within the main unit 1, it achieves diverse cleaning modes through three types of control units: The time control unit 11 has a built-in timer program. Users can preset cleaning time nodes (such as after class). When the time is up, the motor 3 will be automatically triggered to start and drive the device to complete the blackboard wiping operation, realizing unattended timed cleaning.

[0046] Manual control unit 12: The motor 3 is started by triggering a physical button, allowing users to manually start the wiping operation at any time according to actual needs, flexibly responding to sudden cleaning needs.

[0047] Foreign object sensing unit 13: Equipped with an ultrasonic sensor, it uses the principle of ultrasonic reflection to detect foreign objects in the blackboard groove in real time. When a foreign object is detected, the sensor feeds back the signal to the control module, which then starts the motor 3 according to a preset strategy, driving the cleaning sponge 2 to wipe the area with the foreign object in a targeted manner, achieving on-demand cleaning and avoiding ineffective operation.

[0048] Example 2: The present invention provides an automatic blackboard erasing slot method, which includes at least one of the following operating modes: Automatic timed wiping mode: The timed control unit 11 of the multi-mode control module automatically sends a start signal at a preset time, which starts the motor 3 to drive the cleaning sponge 2 to wipe the blackboard groove repeatedly. Manual trigger wiping mode: Press the physical button of the manual control unit 12 to send a start signal. After receiving the signal, the multi-mode control module starts the motor 3, which drives the cleaning sponge 2 to reciprocate. The machine will automatically stop after the set wiping time is reached. Foreign object sensing wiping mode: The ultrasonic sensor monitors foreign objects in the blackboard groove in real time. When a foreign object is detected, the signal is fed back to the multi-mode control module, which starts the motor 3 to drive the cleaning sponge 2 to wipe the foreign object area back and forth. The machine stops automatically after the foreign object is removed.

[0049] In this embodiment of the invention, it is necessary to further explain that the method uses a multi-mode control module as the core control unit, and drives the cleaning sponge 2 to complete the automated cleaning of the blackboard groove by integrating three triggering mechanisms: timed triggering, manual triggering, and foreign object sensing. The principles of each mode are as follows: Automatic timed wiping mode: The timed control unit 11 built into the multi-mode control module actively triggers according to the preset time nodes; after the timed control unit 11 reaches the preset time, it automatically sends a start signal to the motor 3; after receiving the signal, the motor 3 drives the cleaning sponge 2 to reciprocate in the blackboard groove to complete the wiping; no manual intervention is required, which is suitable for regular cleaning needs at fixed time periods, such as after class every day, before leaving get off work, etc.

[0050] Manually triggered wiping mode: This mode is manually triggered by a physical button and is a flexible cleaning mode that starts on demand. After pressing the button, the manual control unit 12 transmits the start signal to the multi-mode control module. The control module sends a command to the motor 3 to drive the cleaning sponge 2 to reciprocate. When the wiping time reaches the preset value, the control module automatically sends a stop signal, and the motor 3 stops working. It responds quickly and can deal with sudden blackboard tray dirt at any time.

[0051] Foreign Object Sensing Wiping Mode: Passively triggered by real-time monitoring data from the ultrasonic sensor, this is an intelligent and precise cleaning mode. The ultrasonic sensor continuously scans the environment inside the blackboard slot, and when it detects foreign objects (such as chalk dust accumulation or fallen debris), it immediately feeds back the detection signal to the multi-mode control module. The control module starts motor 3, which drives the cleaning sponge 2 to perform directional reciprocating wiping on the foreign object area to complete the cleaning. Features: Highly targeted, it only starts when cleaning is needed, saving energy and improving cleaning efficiency.

[0052] See Figure 2 The flowchart for automatically sending the start signal, in one possible embodiment, includes the following steps when the timing control unit 11 of the multi-mode control module automatically sends the start signal at a preset time: Five minutes before sending the start signal, the timing control unit 11 performs a full-area rapid scan of the blackboard groove using the ultrasonic sensor of the foreign object sensing unit 13. If the amount of foreign object accumulation in all areas is lower than the preset threshold, the start signal sending operation is temporarily canceled, and the start signal is sent again at the next preset time node. If the amount of foreign matter accumulated in any area exceeds a preset threshold, a start signal is sent.

[0053] In this embodiment of the invention, it is necessary to further explain that the timing control unit 11, as the core scheduling module, generates start signal triggering instructions according to a preset time period (such as a fixed time period or interval each day), which is the time triggering basis of the entire mechanism and ensures that the equipment can enter the working preparation state in a regular manner.

[0054] A 5-minute pre-detection window is set before the official start signal is sent to activate the ultrasonic sensor for foreign object detection. The 5-minute buffer period provides sufficient detection time without excessively delaying the device's response to the requirements.

[0055] Differentiated logic is executed based on foreign object detection results to achieve "on-demand startup": Active cancellation when there is no demand: If the amount of foreign matter accumulated in all areas is below the preset threshold, it means that the blackboard tray is clean. Starting the equipment at this time is invalid. Therefore, the start signal is temporarily cancelled and the detection process is restarted at the next preset time node. Instant triggering when needed: As long as the amount of foreign matter accumulated in any area exceeds the preset threshold, it is determined that the equipment operation is necessary, and a start signal is immediately sent to trigger the cleaning of the blackboard groove.

[0056] See Figure 3 The flowchart illustrates the strategy for dynamically adjusting the reciprocating speed and wiping stroke of the cleaning sponge 2. In one possible embodiment, when the starter motor 3 drives the cleaning sponge 2 to reciprocate within the blackboard groove, the strategy for dynamically adjusting the reciprocating speed and wiping stroke of the cleaning sponge 2 is executed, including: Regional adaptive speed control: Based on the difference in echo signal intensity of ultrasonic sensors, the blackboard groove is divided into high-density foreign object area, medium-density foreign object area and low-density foreign object area; for high-density area, the cleaning sponge 2 is controlled to perform low-speed reciprocating motion at 1.5-2cm / s, for medium-density area, it is wiped at medium speed at 3-4cm / s, and for low-density area, it is wiped at fast speed at 6-8cm / s, so as to achieve the optimal balance between cleaning efficiency and energy consumption; Material-sensing intelligent speed adjustment: Utilizing the foreign object material recognition model built into the multi-mode control module, the type of foreign object (dry chalk dust, wet chalk dust, eraser debris) is distinguished by the frequency characteristics of the ultrasonic echo. For sticky foreign objects such as wet chalk dust, the wiping speed is automatically reduced to below 1cm / s, and two reverse wiping strokes are added to enhance the adsorption of the cleaning sponge 2 by extending the contact time. For dry chalk dust and eraser debris, a fixed speed is maintained by regional adaptive speed control to avoid excessive wiping that causes dust to fly.

[0057] In this embodiment of the invention, it is necessary to further explain that the cleaning strategy achieves the optimal balance between the cleaning effect, efficiency, and energy consumption of the blackboard tray through a closed-loop logic of sensor perception + intelligent algorithm decision-making + dynamic adjustment of the actuator. The core principle is as follows: (1) Regional adaptive speed control principle Differential speed regulation based on foreign object density sensing mechanism using ultrasonic sensors: Density zoning logic: An ultrasonic sensor emits a signal into the blackboard groove. The density of foreign matter in different areas will cause differences in the intensity of the echo signal: high-density foreign matter areas (such as areas with a large amount of accumulated chalk dust) will reflect ultrasonic waves more strongly, while low-density areas will have weaker signals. By analyzing the echo intensity threshold, the blackboard groove is divided into three foreign matter areas: high-density, medium-density, and low-density.

[0058] Speed ​​matching logic: Match the appropriate speed to the cleaning difficulty of areas with different densities. High-density foreign matter area: Wipe at a low speed of 1.5-2cm / s to extend the contact time between the sponge and the foreign matter, ensuring that stubborn deposits are fully absorbed and avoiding missed areas; Medium-density foreign matter area: Wipe at a medium speed of 3-4cm / s to balance cleaning effectiveness and efficiency; Low-density foreign object area: Wipe quickly at 6-8cm / s, significantly shortening cleaning time and reducing motor energy consumption while ensuring basic cleanliness.

[0059] Balancing goals: By adopting a differentiated strategy of "slow in difficult areas and fast in easy areas," we can avoid both the inefficiency caused by low speed across the entire area and the incomplete cleaning caused by high speed across the entire area, thus achieving the optimal balance between cleaning efficiency and energy consumption.

[0060] (2) Material-sensing intelligent speed regulation principle Targeted speed adjustment is achieved through a material identification mechanism based on the frequency characteristics of ultrasonic echoes. Material identification logic: The multi-mode control module has a built-in foreign object material identification model that distinguishes the type of foreign object by analyzing the frequency characteristics of ultrasonic echoes. Different materials of foreign objects (dry chalk dust, wet chalk dust, eraser debris) have different ultrasonic reflection frequencies: wet chalk dust has a more stable echo frequency due to its high viscosity and aggregated form; dry chalk dust is a loose particle with large fluctuations in echo frequency; eraser debris is a fibrous foreign object with a specific waveform in its echo frequency.

[0061] Differentiated cleaning strategy: For sticky foreign objects such as wet chalk dust: the wiping speed is automatically reduced to below 1cm / s, and two reverse wiping strokes are added. By extending the contact time between the sponge and the foreign object, the sponge pores can fully absorb the sticky foreign object, avoiding foreign object residue due to wiping too quickly. For dry chalk dust and eraser debris: maintain a fixed speed for each area with adaptive speed control to ensure cleaning efficiency while avoiding excessive wiping that causes dust to fly, and reduce unnecessary energy consumption.

[0062] This embodiment uses an ultrasonic sensor to collect data in real time → an algorithm model to analyze the density and material of foreign objects → output speed and stroke adjustment commands → motor 3 drives the sponge to perform corresponding actions in a cycle that dynamically adapts to the complex cleaning scenarios in the blackboard groove. While ensuring thorough cleaning, it maximizes cleaning efficiency and reduces energy consumption and the risk of secondary pollution.

[0063] In one possible embodiment, dividing the blackboard groove into a high-density foreign matter area, a medium-density foreign matter area, and a low-density foreign matter area includes: Before each area division, the ultrasonic sensor scans the clean reference area of ​​the blackboard slot (such as an unused blank slot segment) to obtain the reference intensity of the echo signal under the current environment; the intensity and attenuation rate of the echo signal in each area of ​​the blackboard slot are extracted, and the blackboard slot is divided into three types of areas, with the following judgment criteria: High-density foreign object region: echo intensity is higher than 80% of the reference intensity, and attenuation rate ≥ 0.5dB / ms; Medium-density foreign object region: echo intensity is 40%-80% of the reference intensity, and attenuation rate is between 0.2-0.5dB / ms; Low-density foreign object region: The echo intensity is less than 40% of the reference intensity, and the attenuation rate is <0.2dB / ms.

[0064] In this embodiment of the invention, it is necessary to further explain that, based on the propagation and reflection characteristics of ultrasonic waves, this embodiment achieves accurate classification of foreign object density areas within the blackboard groove by quantitatively analyzing the core parameters of the echo signal. The principle can be divided into three core steps: benchmark calibration, signal feature analysis, and area determination. (1) Reference signal calibration: a reference standard to eliminate environmental interference. Before each region division, a clean reference region (unused blank slot segment) is scanned to obtain the reference intensity of the echo signal under the current environment. The principle is as follows: When ultrasound propagates in a uniform medium (clean blackboard groove) free of foreign objects, the laws governing reflection, refraction, and energy attenuation are stable. The reference intensity represents the reference value of ultrasound reflection energy when there is "no foreign object interference".

[0065] This step can offset the impact of environmental variables (such as sound speed fluctuations caused by changes in temperature and humidity, and sensitivity drift of sensors over long-term use) on the detection results, ensuring the objectivity and consistency of subsequent area determination.

[0066] (2) Signal feature extraction: Quantization mapping of foreign matter density The two core parameters extracted from each region of the blackboard groove are intensity and attenuation rate. Their physical meaning is as follows: Echo intensity: Reflects the energy reflected by ultrasound waves after encountering a foreign object. The higher the density of the foreign object and the denser the particles, the stronger its ability to reflect ultrasound waves, and the higher the energy (intensity) of the echo signal; conversely, the fewer the foreign objects, the more ultrasound waves penetrate or are absorbed, and the lower the echo intensity.

[0067] Attenuation rate: This reflects how quickly ultrasonic waves lose energy during their propagation path. High-density foreign objects rapidly consume ultrasonic energy through scattering and absorption, resulting in a rapid decrease in signal strength over time (high attenuation rate); low-density foreign objects consume ultrasonic energy less effectively, resulting in a low attenuation rate.

[0068] (3) Region classification determination: density grading based on signal features By using a combined threshold of echo intensity and attenuation rate, the blackboard grooves are divided into three regions, corresponding to different foreign object density scenarios, as shown in Table 1: Table 1

[0069] This method essentially utilizes the physical principle of the difference in propagation characteristics of ultrasound in different media to transform the invisible foreign object density into quantifiable electrical signal parameters, thereby achieving non-contact and precise regional density grading and providing data support for intelligent cleaning of blackboard slots (such as enhanced cleaning for high-density areas).

[0070] In one possible embodiment, distinguishing the type of foreign object by the frequency characteristics of the ultrasonic echo includes: Signal acquisition and synchronization: Control the ultrasonic sensor to emit a probe wave into the blackboard groove at a fixed transmission frequency of 40kHz, and simultaneously acquire the raw frequency data of the reflected echo at a sampling frequency of not less than 100kHz to ensure that the slight frequency shift caused by the foreign material is captured. Feature denoising and extraction: Adaptive wavelet threshold filtering is applied to the echo frequency data to filter out interference signals such as environmental vibration and equipment noise, and three core feature parameters of the echo frequency are extracted: peak value, fluctuation amplitude and frequency offset. Material template matching: Input the three types of core feature parameters into the foreign object material recognition model, and perform similarity matching calculations with the three types of standard material feature templates built into the model: Dry chalk dust template: The peak frequency is stable at 38-39kHz, the fluctuation range is <5%, and the frequency deviation is <1kHz; Wet chalk stencil: the peak frequency drops to 35-37kHz, the fluctuation range is 10%-15%, and the frequency offset is 2-3kHz; Plate-wound debris template: Frequency peaks are randomly distributed between 36-40kHz, with fluctuation amplitude >20% and frequency offset >3kHz; Recognition result determination: If the matching degree between the feature parameters and a certain type of template is ≥85%, the corresponding foreign material type is directly output; If the matching degree is in the range of 60%-85%, initiate the secondary sampling verification process; If the matching degree is less than 60%, it is determined that there are no foreign objects.

[0071] In this embodiment of the invention, it is necessary to further explain that the accurate identification of foreign objects in the blackboard groove is achieved based on the material differences in ultrasonic reflection characteristics. The core principle is that different materials have significant differences in the reflection, absorption, and scattering of ultrasonic waves, resulting in a regular distribution of the frequency characteristics (peak value, fluctuation amplitude, and offset) of the echo signal. The type determination is completed through signal analysis and template matching. The specific principle is analyzed in four stages: (1) Signal acquisition and synchronization The core principle is that when ultrasonic waves encounter objects of different densities, hardness, and humidity during propagation, the frequency of the echo will shift due to differences in reflection, refraction, and energy absorption. The system uses a fixed transmission frequency of 40kHz because ultrasonic waves attenuate less during air propagation at this frequency and have a high degree of recognition for reflected signals from lightweight foreign objects such as chalk dust and eraser debris.

[0072] Sampling logic: Echoes are acquired at a sampling frequency of ≥100kHz (satisfying the Nyquist sampling theorem, the sampling frequency is ≥2 times the signal frequency) to ensure complete capture of subtle frequency changes in the echo signal, providing raw data support for subsequent feature extraction.

[0073] (2) Feature denoising and extraction Noise reduction principle: Adaptive wavelet threshold filtering can accurately distinguish between environmental vibration, equipment noise and effective echo signals by analyzing the local features of the signal, filtering out high-frequency random interference and retaining the frequency characteristics related to the material.

[0074] Feature selection logic: The three types of core features extracted directly reflect the material's interaction with ultrasound: Peak frequency: The difference in acoustic impedance of different materials leads to different degrees of shift in the main frequency of the echo (for example, wet chalk dust has a high humidity and density, which makes it more absorbent of ultrasonic energy, resulting in a lower peak frequency of the echo). Fluctuation amplitude: The uniformity of the material determines the stability of the echo signal (e.g., irregular shape of the board eraser debris, loose material, and chaotic reflected waves with large fluctuation amplitude). Frequency offset: The difference between the transmitted frequency and the echo peak value, which directly reflects the material's ability to absorb and scatter ultrasonic energy.

[0075] (3) Material template matching Template construction logic: The three built-in standard templates are material feature benchmarks obtained through training with a large number of samples. Dry chalk dust: It has a uniform texture and low density, and is stable in reflecting ultrasonic waves. Therefore, its frequency peak is close to the transmission frequency (38-39kHz), with small fluctuation amplitude (<5%) and offset <1kHz. Wet chalk dust: Increased humidity leads to increased density and viscosity, enhanced energy absorption, and a drop in echo peak to 35-37kHz. The fluctuation amplitude increases to 10%-15% due to uneven humidity, and the offset reaches 2-3kHz. Board eraser debris: The material is mixed (sponge, rubber fiber, etc.), the surface is rough, the reflected wave has no fixed pattern, the peak value is randomly distributed in 36-40kHz, the fluctuation amplitude is >20%, and the offset is >3kHz.

[0076] Matching principle: The cosine similarity or Euclidean distance between real-time features and template features is calculated using machine learning algorithms to quantify feature overlap and achieve preliminary mapping of material type.

[0077] (4) Judgment of recognition results Threshold setting logic: Matching accuracy ≥ 85%: Features and templates are highly consistent, material types are clearly defined, and results are output directly; 60%-85%: Features may be blurred due to foreign matter mixing or environmental interference. Data reliability can be improved by secondary sampling verification (increasing the number of samplings and optimizing filtering parameters). <60%: The features show no obvious material regularity, indicating no foreign objects or unknown interference, thus avoiding misjudgment.

[0078] Overall logical closed loop: Through the process of "physical signal acquisition → feature extraction and noise reduction → intelligent template matching → multi-threshold result judgment", the physical interaction law between ultrasonic waves and materials is transformed into quantifiable digital features, ultimately achieving accurate and efficient identification of three types of foreign objects in the blackboard groove.

[0079] See Figure 4 The flowchart of adaptive wavelet threshold filtering, in one possible embodiment, includes the following steps when performing adaptive wavelet threshold filtering on echo frequency data: A three-level wavelet decomposition structure is used to perform multi-scale decomposition processing on the echo frequency data: Low-frequency approximation layer: Soft threshold filtering is used to suppress low-frequency environmental noise such as air conditioners and fans; First high-frequency detail layer: Hard threshold filtering is used to preserve the high-frequency feature differences between chalk dust and eraser debris; The second high-frequency detail layer: Adaptive sliding window threshold filtering is used to remove random impulse noise, and the window size is dynamically adjusted according to the signal energy; the denoised data is generated by wavelet reconstruction.

[0080] In this embodiment of the invention, it is necessary to further explain that the signal is decomposed into different frequency levels through wavelet decomposition. Differential threshold filtering strategies are matched based on the noise characteristics and useful signal features of each level, ultimately reconstructing a clean target signal. The specific principle is as follows: (1) Three-layer wavelet decomposition Wavelet decomposition breaks down the original echo frequency data (which mixes useful signals with various types of noise) into one low-frequency approximation layer and two high-frequency detail layers through wavelet basis function transformation. Low-frequency approximation layer: corresponds to the core trend and low-frequency components of the signal, contains the basic characteristics of the echo signal, and is superimposed with continuous low-frequency environmental noise such as air conditioners and fans; High-frequency detail layer: corresponds to the abrupt changes and high-frequency characteristics of the signal, including the high-frequency feature differences of target objects such as chalk dust and eraser debris, as well as high-frequency interference such as random impulse noise.

[0081] By using multi-scale decomposition, noise and useful signals are separated at different frequency dimensions, providing a foundation for precise noise reduction.

[0082] (2) Hierarchical threshold filtering Three threshold filtering strategies are adopted to address the signal and noise characteristics at different levels: Low-frequency approximation layer: Soft threshold filtering suppresses low-frequency ambient noise. The core logic of soft threshold filtering is to set the absolute value of coefficients less than a set threshold to 0, and to retain coefficients greater than the threshold after subtracting the threshold value. This method can smooth low-frequency signals, effectively filter continuous low-frequency interference from sources like air conditioners and fans, while avoiding excessive loss of the low-frequency fundamental trend of the echo signal.

[0083] First high-frequency detail layer: Hard thresholding filter preserves the target's high-frequency features. The logic of hard thresholding is to set only the components whose absolute values ​​are less than the threshold to 0, while retaining all coefficients greater than the threshold. Since chalk dust and eraser debris exhibit high-frequency abrupt changes in signal characteristics, hard thresholding can preserve these sharp feature differences to the greatest extent possible, avoiding signal ambiguity caused by soft thresholding and ensuring accurate differentiation of the target objects in subsequent processing.

[0084] Second high-frequency detail layer: Adaptive sliding window threshold filtering to remove random impulse noise. Random impulse noise is a sudden, large-amplitude fluctuation in interference, and a fixed threshold cannot adapt to its randomness. The core of adaptive sliding window thresholding is to dynamically adjust the threshold based on the real-time signal energy within the window: when the signal energy within the window suddenly increases (most likely impulse noise), the threshold is automatically increased to filter out the interference; when the energy is stable, the threshold is decreased to preserve useful details. This dynamic adjustment strategy can accurately identify and filter random impulse noise while protecting useful high-frequency signal details.

[0085] (3) Wavelet reconstruction The low-frequency approximation coefficients and two high-frequency detail coefficients, after threshold filtering, are reconstructed through inverse wavelet transform and recombined into a single time-domain / frequency-domain signal. This results in echo frequency data that removes low-frequency environmental noise and random impulse noise while fully preserving the characteristics of chalk dust and eraser debris.

[0086] This hierarchical adaptive processing method solves the problem of "difficulty in balancing noise reduction and feature preservation" in traditional single threshold filtering, and achieves a balance between accurate noise suppression and maximum preservation of target features.

[0087] In one possible embodiment, when performing adaptive wavelet threshold filtering on the echo frequency data, the method further includes: For the echo data in the edge region of the blackboard groove, an edge-enhancing filtering strategy is adopted, which specifically includes: Automatically reduce the wavelet threshold in edge regions by 25% to prevent edge reflection signals from being misjudged as noise; Edge detection operators (such as the Sobel operator) are introduced to perform secondary processing on the filtered echo data, thereby enhancing the frequency characteristic abrupt changes in the edge region.

[0088] In this embodiment of the invention, it is necessary to further explain that the echo signal in the edge region of the blackboard groove exhibits strong abrupt changes (the edge reflection signal differs significantly from the background signal), and conventional threshold processing easily misjudges it as noise filtering. Therefore, an edge-enhanced filtering strategy is required, the specific principle of which is as follows: (1) The principle of reducing the wavelet threshold in the edge region by 25% Limitations of conventional thresholds: Global or universal adaptive thresholds are usually calculated based on the overall noise level and lack specificity for strong abrupt changes in edge regions. If the threshold is too high, the wavelet coefficients of edge signals may be misjudged as noise and set to zero.

[0089] The effect of threshold reduction: Reducing the threshold in the edge region by 25% can expand the range of effective signals that are retained, ensuring that the wavelet coefficients of the edge reflection signal (the amplitude is usually greater than the noise but may be lower than the global threshold) are not mistakenly filtered out.

[0090] Adaptive logic: By pre-identifying edge regions (such as gradient detection, region growing, etc.), the threshold is adjusted only for the local region to avoid noise residue caused by the reduction of the global threshold.

[0091] (2) Introducing the principle of quadratic processing using the Sobel operator The core function of the Sobel operator is a gradient-based edge detection operator that identifies edge regions with abrupt changes in grayscale by calculating the rate of change of grayscale values ​​of pixels in an image (or signal). Its mathematical principle involves using a 3×3 convolution kernel to perform convolution operations in both the horizontal and vertical directions, and then calculating the gradient magnitude by summing the squares and taking the square root.

[0092] The purpose of secondary processing is to enhance the frequency characteristic abrupt changes in edge regions by using the Sobel operator. While edge signals are preserved after adaptive threshold filtering, they may be weakened due to threshold shrinkage. This process highlights edge contours and further improves signal discernibility.

[0093] Process complementarity: Adaptive threshold filtering is responsible for "noise reduction and edge preservation", while the Sobel operator is responsible for "edge enhancement". The combination of the two achieves a complete process from "noise reduction" to "enhancement", ensuring that the echo signal in the edge area of ​​the blackboard groove is clearly distinguishable.

[0094] This strategy solves the "filtering wave" problem of traditional wavelet threshold filtering when processing strong abrupt signals by combining local threshold adjustment and secondary enhancement through edge detection, achieving the following: It effectively removes background noise while fully preserving the reflected signals from the edges of the blackboard groove.

[0095] Enhance edge features to improve the accuracy of subsequent signal analysis (such as edge localization and size measurement).

[0096] The adaptive mechanism ensures the algorithm's robustness to different scenarios without requiring manual intervention to adjust parameters.

[0097] In one possible embodiment, when detecting foreign objects in the blackboard groove in real time using an ultrasonic sensor, the process includes: A segmented dynamic scanning mechanism is adopted: the blackboard groove is divided into several equally long detection units along its length (e.g., each unit is 10cm). The ultrasonic sensor is controlled to perform directional detection on each detection unit in sequence according to a preset path. After the detection of each detection unit is completed, a signal processing time is reserved to achieve full coverage of the blackboard groove without blind spots.

[0098] In this embodiment of the invention, it is necessary to further explain that dividing the continuous blackboard groove into several equally long detection units (e.g., one unit every 10cm) along the length direction is essentially decomposing the long, linear detection area into independent, discrete gridded detection units.

[0099] Principle and Logic: The blackboard groove has a long and narrow structure. If a whole-body scanning method is used, the ultrasonic signal is prone to reflection, superposition, or attenuation at both ends and corners of the groove, resulting in a decrease in detection accuracy in edge areas. By dividing the area into units, each unit becomes an independent detection "target," which can focus on the signal characteristics of a local area and avoid signal interference from long-distance scanning.

[0100] This provides a foundation for accurate positioning of foreign objects, allowing direct association with specific units when a foreign object is detected, supporting a "targeted wiping" cleaning strategy.

[0101] Directional design: The sensor emits vertically or at a small angle to each unit, ensuring that the sound wave energy is concentrated on the unit area, avoiding side lobe signal interference, and improving the recognition of the echo signal.

[0102] After each detection unit is completed, a signal processing time is reserved. The core purpose is to provide a time window for the analysis and calculation of the echo signal. After the sensor receives the echo, multi-dimensional signal analysis needs to be performed: The distance between the foreign object and the sensor is calculated based on the echo time difference to confirm the location of the foreign object within the unit; Analyze echo intensity and attenuation rate to determine the density accumulation of foreign matter; Based on frequency characteristics, the material of the foreign object (dry / wet chalk dust, eraser debris) can be preliminarily identified.

[0103] The necessity of allowing sufficient time: If continuous detection occurs without intervals, the signal processing of the previous unit will not be completed, and the echo signal of the next unit will be superimposed with interference, leading to data confusion and increased detection errors. Intervals ensure that the detection of the next unit can only begin after the data processing of the previous unit is completed, realizing a pipeline-like operation of "detection-processing-re-detection".

[0104] In one possible embodiment, when the starter motor 3 drives the cleaning sponge 2 to reciprocate wiping the area containing foreign matter, the process includes: If the amount of foreign matter accumulated in all detection units is below the preset threshold, the wiping operation will not be performed. If the amount of foreign matter accumulated in any detection unit exceeds the preset threshold, a strategy of dynamically adjusting the reciprocating speed and wiping stroke of the cleaning sponge 2 is implemented to ensure that stubborn chalk dust is completely removed.

[0105] In this embodiment of the invention, it should be further explained that when the amount of foreign matter accumulation reported by all detection units is lower than a preset threshold, it is determined that the foreign matter in the current area is within an acceptable range, and the system does not start the wiping operation, thereby avoiding ineffective operation and reducing equipment energy consumption and component wear.

[0106] If the amount of foreign matter accumulated in any detection unit exceeds the threshold, the motor 3 will be immediately triggered to drive the cleaning sponge 2 to start wiping, and at the same time, the dynamic adjustment strategy will be activated: regional adaptive speed control and material-sensing intelligent speed adjustment.

[0107] Compared to traditional timed cleaning and fixed parameter cleaning methods, this principle achieves precision and efficiency in cleaning operations through closed-loop control of "detection-judgment-execution-adjustment": it avoids the waste of resources by "over-cleaning" and can optimize cleaning strategies for stubborn foreign objects.

[0108] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automatic blackboard erasing slot device, characterized by, Comprise: Host: built-in battery, install the cleaning sponge below, install embedded in the blackboard slot; Power transmission system: contains motor, gear transmission group and drive wheel, the top of the motor shell is connected to the host through the connecting frame, the output end is connected to the drive wheel through the gear transmission group, and the drive wheel is attached to the outer side wall of the blackboard slot after installation; Support walking mechanism: contains support arm and non-powered auxiliary wheel, one end of the support arm is connected to the bottom of the motor shell, and the other end is connected to the non-powered auxiliary wheel, and the non-powered auxiliary wheel is attached to the wall surface below the blackboard slot after installation; Multi-mode control module: integrated in the host, containing three types of control units: Timing control unit: built-in timing program, can automatically trigger the motor to start at the preset time, complete the blackboard slot cleaning; Manual control unit: trigger the motor to start through the physical button, support users to manually operate as needed; Foreign matter sensing unit: equipped with ultrasonic sensor, real-time detection of foreign matter in the blackboard slot, after detecting foreign matter, start the motor according to the preset strategy for targeted cleaning; When in use, the motor drives the drive wheel to move linearly along the outer side wall of the blackboard slot through the gear transmission group, driving the cleaning sponge to move synchronously in the slot, realizing automatic cleaning.

2. A method of automatically erasing a blackboard slot, characterized by, The device of claim 1 comprises at least one of the following operating modes: Timing automatic cleaning mode: through the timing control unit, automatically send start signal at the preset time, drive the cleaning sponge to reciprocatingly clean the blackboard slot; Manual trigger cleaning mode: press the physical button to send start signal, after receiving by the multi-mode control module, start the motor, drive the cleaning sponge to reciprocate, and automatically stop after reaching the set time; Foreign matter sensing cleaning mode: through the ultrasonic sensor, real-time monitor the blackboard slot, after detecting foreign matter, feedback signal to the multi-mode control module, start the motor to drive the cleaning sponge to reciprocate in the foreign matter area, and automatically stop after cleaning.

3. The method of claim 2, wherein, When the timing control unit automatically sends start signal at the preset time, it includes: 5 minutes before timing start, through the ultrasonic sensor, scan the blackboard slot, if the foreign matter accumulation of all areas is lower than the preset threshold, cancel this start; if the foreign matter accumulation of any area is higher than the preset threshold, send start signal.

4. The method of claim 2, wherein, When driving the cleaning sponge to reciprocate to clean the blackboard slot, execute dynamic adjustment of motion speed and stroke strategy, including: Subarea adaptive speed regulation: based on ultrasonic echo signal strength, divide the blackboard slot into high, medium and low density foreign matter areas; control the cleaning sponge to reciprocate at low speed of 1.5-2 cm / s in high density area, medium speed of 3-4 cm / s in medium density area, and high speed of 6-8 cm / s in low density area; Material perception intelligent speed regulation: through foreign matter material identification model, analyze ultrasonic echo frequency characteristics to distinguish dry chalk dust, wet chalk dust and eraser debris; for wet chalk dust, reduce the speed to below 1 cm / s and increase 2 reverse strokes; for dry chalk dust and eraser debris, maintain subarea speed regulation speed.

5. The method of claim 4, wherein, When dividing the blackboard slot into high, medium and low density foreign matter areas, it includes: Before each division, scan the clean reference area of the blackboard slot through the ultrasonic sensor to obtain the reference strength of the echo signal; extract the echo strength and attenuation rate of each area, and divide according to the following standards: High density area: echo strength > 80% of reference, and attenuation rate ≥ 0.5 dB / ms; Medium density area: echo intensity is 40%-80% of the reference, and the decay rate is 0.2-0.5 dB / ms; Low density area: echo intensity <40% of the reference, and the decay rate <0.2 dB / ms.

6. The method of claim 5, wherein, When distinguishing dry chalk dust, wet chalk dust and eraser debris by analyzing the frequency characteristics of ultrasonic echoes, it includes: Control the ultrasonic sensor to emit a detection wave at 40 kHz, and simultaneously collect raw frequency data of the echo at a sampling frequency of ≥100 kHz to capture the frequency shift caused by the material; Adaptive wavelet threshold filtering of echo data, extraction of three types of core feature parameters: frequency peak, fluctuation amplitude and frequency shift; Input the three types of core feature parameters into the foreign matter material identification model, and match with the built-in three types of standard templates: Dry chalk dust: frequency peak 38-39 kHz, fluctuation amplitude <5%, frequency shift <1 kHz; Wet chalk dust: frequency peak 35-37 kHz, fluctuation amplitude 10%-15%, frequency shift 2-3 kHz; Eraser debris: frequency peak 36-40 kHz randomly distributed, fluctuation amplitude >20%, frequency shift >3 kHz; If the matching degree is ≥85%, the material type is directly output; if it is 60%-85%, secondary sampling is started; if it is <60%, it is determined that there is no foreign matter.

7. The method of claim 6, wherein, When performing adaptive wavelet threshold filtering on echo data, it includes: Multi-scale processing of echo data using a three-layer wavelet decomposition structure: Low-frequency approximation layer: use soft threshold filtering to suppress low-frequency noise; First high-frequency detail layer: use hard threshold filtering to preserve the high-frequency feature difference between chalk dust and eraser debris; Second high-frequency detail layer: use adaptive sliding window threshold filtering to filter out random impulse noise, and the window size is dynamically adjusted according to the signal energy; the denoised data is generated by wavelet reconstruction.

8. The method of claim 7, wherein, When performing adaptive wavelet threshold filtering on echo data, it also includes: For echo data in the edge area of the blackboard groove, use edge enhancement filtering: Automatically reduce the edge area wavelet threshold by 25% to avoid misjudgment of edge reflection signals as noise; Introduce Sobel operator for secondary processing of filtered data to strengthen the edge frequency feature mutation point.

9. The method of claim 4, wherein, When monitoring the blackboard groove in real time through the ultrasonic sensor, it includes: Divide the blackboard groove into several equal-length detection units, control the ultrasonic sensor to detect in a predetermined path, and reserve a signal processing time length after completing each detection unit to achieve full-area coverage without blind area.

10. The method of claim 9, wherein, When starting the motor to drive the cleaning sponge to reciprocally wipe the foreign matter area, it includes: If the foreign matter accumulation of all detection units is below the preset threshold, do not perform wiping operation; If the foreign matter accumulation of any detection unit is higher than the preset threshold, execute the dynamic adjustment of the reciprocating motion speed and wiping stroke strategy of the cleaning sponge to ensure that stubborn chalk dust is completely removed; If the foreign matter accumulation of all detection units is below the preset threshold, do not perform wiping; if the foreign matter accumulation of any detection unit is higher than the preset threshold, execute the dynamic adjustment of the motion speed and stroke strategy.

Images