Anti-entanglement dust brush control optimization method for intelligent sweeping machine

Abstract

This invention discloses an optimization method for anti-tangling dust brush control in intelligent sweeping robots, relating to the field of intelligent control technology. The method includes: analyzing the roller brush humidity information and the first roller brush shaft rotation speed through a debris removal rate prediction network to obtain a first predicted debris removal rate, wherein the first roller brush shaft rotation speed is randomly assigned based on a rated range of roller brush shaft rotation speeds; when the first predicted debris removal rate is greater than or equal to the convergence removal rate, the first roller brush shaft rotation speed is set as the target roller brush shaft rotation speed, wherein the convergence removal rate is a user-preset removal rate threshold, and the convergence removal rate is ≥0.6; and anti-tangling dust brush control is performed based on the target roller brush shaft rotation speed. This solves the technical problem in existing intelligent sweeping robots where the roller brush easily tangles with debris during cleaning, leading to reduced cleaning efficiency. By predicting the debris removal rate and dynamically adjusting the roller brush shaft rotation speed, intelligent anti-tangling dust brush control is achieved, resulting in improved cleaning efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and more specifically to an optimization method for controlling anti-tangling dust brushes in intelligent sweeping machines. Background Technology

[0002] With the rapid advancement of technology and the continuous improvement of people's living standards, smart home products are integrating into our daily lives at an unprecedented pace, becoming an indispensable part of modern families. As a leader in the smart home cleaning field, the intelligent robotic vacuum cleaner has won the favor of many consumers due to its convenient operation, efficient cleaning capabilities, and intelligent features. However, traditional intelligent robotic vacuum cleaners often face a thorny problem during cleaning: the roller brush easily gets tangled with small debris such as hair and lint. This not only significantly reduces cleaning efficiency and increases energy consumption but may also lead to machine malfunctions, requiring frequent manual intervention from the user, seriously affecting the user experience. Summary of the Invention

[0003] This application provides an optimization method for anti-tangling dust brush control in intelligent sweeping machines, which solves the technical problem in the prior art that the roller brush of intelligent sweeping machines is prone to tangling with debris during the sweeping process, resulting in reduced sweeping efficiency.

[0004] In view of the above problems, this application provides an optimization method for anti-tangling dust brush control for intelligent sweeping machines.

[0005] This application provides a method for optimizing the control of anti-tangling dust brushes in intelligent sweeping robots, the method comprising:

[0006] The first predicted cleaning rate of debris is obtained by analyzing the humidity information of the roller brush and the rotation speed of the first roller brush shaft through a debris cleaning rate prediction network. The rotation speed of the first roller brush shaft is randomly assigned based on the rated range of the roller brush shaft rotation speed. When the first predicted cleaning rate of debris is greater than or equal to the convergence cleaning rate, the rotation speed of the first roller brush shaft is set as the target roller brush shaft rotation speed. The convergence cleaning rate is a user-preset cleaning rate threshold, and the convergence cleaning rate is ≥0.6. Anti-tangling dust brush control is performed according to the target roller brush shaft rotation speed.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] First, a debris removal rate prediction network analyzes the roller brush humidity information and the first roller brush shaft speed to obtain the first predicted debris removal rate. The first roller brush shaft speed is randomly assigned based on the rated range of roller brush shaft speeds. Then, when the first predicted debris removal rate is greater than or equal to the convergence removal rate, the first roller brush shaft speed is set as the target roller brush shaft speed. The convergence removal rate is a user-preset removal rate threshold, ≥0.6. Finally, anti-tangling dust brush control is implemented based on the target roller brush shaft speed. This solves the technical problem in existing intelligent sweeping robots where the roller brush easily gets tangled with debris during cleaning, leading to reduced cleaning efficiency. By predicting the debris removal rate and dynamically adjusting the roller brush shaft speed, intelligent anti-tangling dust brush control is achieved, resulting in improved cleaning efficiency. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of the anti-tangling dust brush control optimization method for an intelligent sweeping robot provided in an embodiment of this application;

[0011] Figure 2 This is a flowchart illustrating the process of matching the debris cleaning rate prediction network in the anti-tangling dust brush control optimization method for intelligent sweeping machines provided in this application embodiment.

[0012] Figure 3 This is a schematic diagram of the structure of the anti-tangling dust brush provided in an embodiment of this application.

[0013] Figure labeling: 1. Brush roller; 2. Brush roller shaft. Detailed Implementation

[0014] This application provides an optimization method for controlling the anti-tangling dust brush of a smart sweeper, which solves the technical problem in the prior art where the roller brush of a smart sweeper easily gets tangled with debris during the sweeping process, resulting in reduced sweeping efficiency.

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0016] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0017] Examples, such as Figure 1 As shown in the embodiment of this application, an optimization method for controlling the anti-tangling dust brush of a smart sweeping robot is provided, wherein the method includes:

[0018] The first debris removal rate is obtained by analyzing the brush humidity information and the first brush shaft speed through a debris removal rate prediction network. The first brush shaft speed is obtained by randomly assigning a value based on the rated range of the brush shaft speed.

[0019] The debris removal rate prediction network is typically trained using machine learning or deep learning techniques to analyze and predict the debris removal efficiency of a roller brush under different conditions. Roller brush humidity can affect the physical properties of the roller brush, such as friction and adhesion between the brush and the ground and debris, thus impacting the cleaning effect. For example, excessive humidity may cause too much water to adhere to the roller brush surface, reducing its ability to adsorb small debris; while excessively low humidity may increase the electrostatic effect, making it easier for lightweight debris such as hair to entangle on the roller brush. The first roller brush shaft speed is randomly assigned based on the rated range of roller brush shaft speeds. The rated range of roller brush shaft speeds is usually determined according to equipment design and performance requirements, limiting the speed range at which the roller brush can rotate safely and effectively under normal operating conditions. By inputting the roller brush humidity information and the first roller brush shaft speed into the debris removal rate prediction network for analysis, the first predicted debris removal rate is output. The first predicted debris removal rate represents the predicted debris removal effect of the roller brush under the current roller brush humidity and speed conditions.

[0020] Furthermore, by analyzing the brush humidity information and the first brush shaft rotation speed through a debris removal rate prediction network, a first predicted debris removal rate is obtained, which includes:

[0021] Receive anti-tangling dust brush model information; receive roller brush humidity information through a humidity sensor installed in the inner layer of the anti-tangling dust brush; match a debris removal rate prediction network according to the anti-tangling dust brush model information; obtain the rated range of roller brush shaft speed according to the anti-tangling dust brush model information.

[0022] Before obtaining the first predicted debris removal rate, it is necessary to identify the model of the currently installed anti-tangling dust brush. Different models of dust brushes may have different design characteristics, materials and structures, which will affect the humidity sensitivity, speed adaptability and debris removal effect of the roller brush.

[0023] A humidity sensor installed inside the anti-tangling dust brush monitors and obtains the humidity information of the brush surface in real time. Different humidity levels affect the physical properties of the brush, such as adhesion and friction.

[0024] Different dust brush models require different debris removal rate prediction networks to accurately predict their debris removal effect. Based on the received dust brush model information, a suitable debris removal rate prediction network is selected from the pre-stored network model library for matching.

[0025] Each model of anti-tangling dust brush has its own specific rated range of roller brush shaft speed. The rated range is determined based on the dust brush's design parameters, material properties, and test results, aiming to ensure that the roller brush can operate safely and effectively under normal working conditions. According to the dust brush model information, the corresponding rated range of roller brush shaft speed is retrieved from the database.

[0026] Furthermore, the humidity information of the roller brush is received by a humidity sensor installed inside the anti-tangling dust brush, including:

[0027] The humidity sensor includes a first humidity sensor, a second humidity sensor, and a third humidity sensor. The first humidity sensor collects first roller brush humidity monitoring time-series information; the second humidity sensor collects second roller brush humidity monitoring time-series information; and the third humidity sensor collects third roller brush humidity monitoring time-series information. Each of the first, second, and third roller brush humidity monitoring time-series information includes at least five time points, with a collection time interval of 0.5 seconds or more and 0.2 seconds or more. The first, second, and third roller brush humidity monitoring time-series information are evaluated for central tendency (CNT) to obtain a first roller brush humidity CNT, a second roller brush humidity CNT, and a third roller brush humidity CNT. The average of the first, second, and third roller brush humidity CNTs is calculated to obtain the roller brush humidity information.

[0028] The humidity sensor system includes a first humidity sensor, a second humidity sensor, and a third humidity sensor. These three humidity sensors are located at different positions within the inner layer of the roller brush to ensure comprehensive and accurate monitoring of humidity changes on the roller brush surface. The first humidity sensor collects timing information on the humidity of the first roller brush. This information is recorded at specific time intervals (0.5 seconds ≥ acquisition interval ≥ 0.2 seconds), including at least five data points to reflect the change in roller brush humidity over time. Similarly, the second and third humidity sensors collect timing information on the humidity of the second and third roller brushes, respectively, ensuring that the humidity information of the roller brush is obtained from multiple perspectives.

[0029] The collected humidity monitoring time series information of the first, second, and third roller brushes was evaluated by central tendency. By calculating the central tendency (such as the median) in each time series information, more stable and reliable central tendency values ​​of the first, second, and third roller brushes were obtained. The mean of the central tendency values ​​of the first, second, and third roller brushes was calculated to obtain the roller brush humidity information, which reflects the overall level of humidity on the roller brush surface.

[0030] Furthermore, such as Figure 2 As shown, based on the anti-tangling dust brush model information, a debris removal rate prediction network is matched, including:

[0031] Based on the anti-tangling dust brush model information, an anti-tangling dust brush training dataset is matched, wherein the anti-tangling dust brush training dataset includes a roller brush humidity training dataset, a roller brush shaft speed training dataset, and a debris cleaning rate label dataset; based on the debris cleaning rate label dataset, the roller brush humidity training dataset and the roller brush shaft speed training dataset are supervised to train a first debris cleaning rate prediction base network; based on the debris cleaning rate label dataset, the roller brush humidity training dataset and the roller brush shaft speed training dataset are supervised to train a second debris cleaning rate prediction base network; until the roller brush humidity training dataset is supervised based on the debris cleaning rate label dataset... Based on the training dataset of the roller brush shaft speed, train the Nth debris cleaning rate prediction base network; construct a fully connected layer to calculate the weighted average of the first debris cleaning rate prediction base network, the second debris cleaning rate prediction base network, up to the Nth debris cleaning rate prediction base network, to obtain the debris cleaning rate prediction network; the fusion weight is the ratio of the test accuracy of each debris cleaning rate prediction base network from the first debris cleaning rate prediction base network to the Nth debris cleaning rate prediction base network to the sum of the N test accuracies; associate and store the anti-tangling dust brush model information with the debris cleaning rate prediction network.

[0032] Specifically, based on the anti-tangling dust brush model information, the corresponding anti-tangling dust brush training dataset is retrieved and matched from the database. This training dataset includes a roller brush humidity training dataset, a roller brush shaft speed training dataset, and a debris cleaning rate identification dataset. Using the debris cleaning rate identification dataset as supervision information, the roller brush humidity training dataset and the roller brush shaft speed training dataset are supervised respectively, training N debris cleaning rate prediction base networks. Each debris cleaning rate prediction base network is trained based on different initialization parameters, optimization algorithms, or network structures, aiming to capture the complex relationship between roller brush humidity and speed and debris cleaning rate from different perspectives. A fully connected layer is constructed to statistically analyze the prediction results of all debris cleaning rate prediction base networks to improve the accuracy and robustness of the predictions. The connection layer calculates the weighted average of the prediction results of all debris removal rate prediction base networks. The weight of each debris removal rate prediction base network is determined based on its test accuracy. That is, the weight of each debris removal rate prediction base network is equal to the ratio of its test accuracy to the sum of the test accuracies of all debris removal rate prediction base networks. After fusion, a debris removal rate prediction network is obtained. The debris removal rate prediction network can more comprehensively consider the impact of factors such as brush humidity and rotation speed on the debris removal rate, thus providing more accurate and reliable prediction results. The anti-tangling dust brush model information is associated and stored with the trained debris removal rate prediction network so that the corresponding debris removal rate prediction network can be quickly matched according to the currently installed dust brush model, realizing real-time debris removal rate prediction and anti-tangling dust brush control.

[0033] Furthermore, based on the anti-tangling dust brush model information, a matching anti-tangling dust brush training dataset is performed, including:

[0034] Based on the anti-tangling dust brush model information, an anti-tangling dust brush test dataset is obtained. Each anti-tangling dust brush test dataset includes test roller brush humidity, test roller brush shaft speed, and multiple debris removal rate records. Mode analysis is performed on the multiple debris removal rate records to obtain debris removal rate identification data. The test roller brush humidity is set as roller brush humidity training data, and the test roller brush shaft speed is set as roller brush shaft speed training data.

[0035] Specifically, based on the anti-tangling dust brush model information, the corresponding anti-tangling dust brush test dataset is retrieved from the database. This dataset contains multiple test cases, each with a specific set of test conditions (such as test brush humidity and test brush shaft speed) and records of debris removal rates obtained from multiple experiments under those conditions. Mode analysis is performed on these debris removal rate records. The mode is the most frequently occurring value in a set of data, representing the central trend or typical value of that set. Assuming that the most frequently occurring value among the debris removal rate data obtained from multiple experiments best represents the debris removal effect under those test conditions, this mode value is used as the debris removal rate identifier for that test case. This identifier will be used during training as supervisory information to guide the network in learning the relationship between brush humidity, brush shaft speed, and debris removal rate. The test brush humidity data from the test cases is set as the brush humidity training data, and the test brush shaft speed data is set as the brush shaft speed training data. These data, together with the debris removal rate identifier data, constitute the anti-tangling dust brush training dataset.

[0036] When the predicted cleaning rate of the first debris is greater than or equal to the convergence cleaning rate, the rotation speed of the first roller brush shaft is set to the target roller brush shaft rotation speed, wherein the convergence cleaning rate is a user-preset cleaning rate threshold, and the convergence cleaning rate is ≥0.6; anti-tangling dust brush control is performed according to the target roller brush shaft rotation speed.

[0037] In the anti-tangling dust brush control logic of the intelligent robotic vacuum cleaner, after the first predicted debris cleaning rate is calculated by the debris cleaning rate prediction network, it is compared with the user-preset convergence cleaning rate. The convergence cleaning rate is the user-preset cleaning rate threshold, representing the minimum standard of debris cleaning effect expected by the user. When the first predicted debris cleaning rate is greater than or equal to the convergence cleaning rate, the current roller brush shaft speed (i.e., the first roller brush shaft speed) is considered to be sufficiently efficient and can meet the user's cleaning needs. Therefore, the current roller brush shaft speed is set as the target roller brush shaft speed, and subsequent cleaning work is maintained at this level.

[0038] Furthermore, it also includes:

[0039] When the first predicted debris removal rate is less than the convergent removal rate, the second roller brush shaft speed is obtained; the roller brush humidity information and the second roller brush shaft speed are analyzed by the debris removal rate prediction network to obtain the second predicted debris removal rate, wherein the second roller brush shaft speed is randomly assigned based on the rated range of roller brush shaft speed; when the second predicted debris removal rate is greater than or equal to the convergent removal rate, the second roller brush shaft speed is set as the target roller brush shaft speed, wherein the convergent removal rate is a user-preset removal rate threshold, and the convergent removal rate is ≥0.6; anti-tangling dust brush control is performed according to the target roller brush shaft speed.

[0040] When the predicted cleaning rate of the first debris is less than the user-preset convergence cleaning rate (i.e., the cleaning effect does not meet expectations), the second roller brush shaft speed can be randomly adjusted according to the rated range of the roller brush shaft speed. The roller brush humidity information and the second roller brush shaft speed are input into the debris cleaning rate prediction network for analysis, and the second debris predicted cleaning rate is output. If the second debris predicted cleaning rate is greater than or equal to the convergence cleaning rate (i.e., the user-preset cleaning rate threshold, such as 0.6), then the current second roller brush shaft speed is considered a feasible choice because it can meet the user's cleaning needs. Once it is determined that the predicted cleaning rate at the second roller brush shaft speed meets the requirements, the second roller brush shaft speed is set as the target roller brush shaft speed, and the roller brush of the sweeper is controlled according to the target roller brush shaft speed.

[0041] Furthermore, it also includes:

[0042] When the predicted debris removal rate of the kth brush shaft speed is less than the convergence removal rate, the k brush shaft speeds are clustered according to the brush shaft speed threshold to obtain multiple sets of brush shaft speeds. The maximum predicted debris removal rate is extracted from each of the multiple sets of brush shaft speeds to obtain multiple maximum brush shaft speeds. From the multiple maximum brush shaft speeds, three reference brush shaft speeds with relatively large predicted debris removal rates are extracted: a first reference brush shaft speed, a second reference brush shaft speed, and a third reference brush shaft speed. The minimum predicted debris removal rate is extracted from each of the multiple sets of brush shaft speeds to obtain multiple minimum brush shaft speeds. According to the brush shaft speed adjustment step range, with the first reference brush shaft speed, the second reference brush shaft speed, and the third reference brush shaft speed as adjustment targets, the multiple minimum brush shaft speeds are iterated through and randomly adjusted several times to obtain brush shaft speed expansion values. When the expanded brush shaft speed value has a brush shaft speed that satisfies the convergence removal rate, it is set as the target brush shaft speed; otherwise, the expanded brush shaft speed value is iteratively updated.

[0043] When multiple brush shaft speeds (denoted as the k-th brush shaft speed) are continuously tried and it is found that the corresponding debris prediction cleaning rate (the k-th debris prediction cleaning rate) is less than the user-preset convergence cleaning rate, K-means clustering is used to cluster the k brush shaft speeds according to the brush shaft speed threshold to obtain multiple sets of brush shaft speeds. The brush shaft speed threshold is determined based on the rated range and historical data of the brush shaft speed.

[0044] For the multiple sets of roller brush shaft speeds obtained after clustering, the maximum predicted debris removal rate and its corresponding roller brush shaft speed (denoted as multiple maximum roller brush shaft speeds) are calculated and extracted for each group. These maximum roller brush shaft speeds represent the speeds with the best cleaning effect in each group. From the multiple maximum roller brush shaft speeds, the top three roller brush shaft speeds with the highest predicted debris removal rates are further extracted and used as the first, second, and third benchmark roller brush shaft speeds, respectively. At the same time, the minimum predicted debris removal rate and its corresponding roller brush shaft speed (denoted as multiple minimum roller brush shaft speeds) are extracted from each group of roller brush shaft speeds. These minimum roller brush shaft speeds represent the cleaning effect in each group. The worst-case rotational speed is determined. The adjustment step size is set according to a preset roller brush shaft rotational speed range. Using the first, second, and third reference roller brush shaft rotational speeds as adjustment targets, multiple minimum roller brush shaft rotational speeds are randomly adjusted several times to find rotational speeds with potentially higher cleaning efficiency around the reference speeds, thus obtaining an expanded value for the roller brush shaft rotational speed. It is then checked whether a roller brush shaft rotational speed that meets the convergence cleaning rate exists within the expanded value range. If it does, this speed is set as the target roller brush shaft rotational speed, and preparations are made for anti-tangling dust brush control. If it does not exist, the expanded value for the roller brush shaft rotational speed is iteratively updated by adjusting the step size, increasing the number of random adjustments, or changing other parameters to expand the search range.

[0045] Furthermore, such as Figure 3 As shown, an anti-tangling dust brush includes a roller brush 1, a roller brush shaft 2, and a roller brush scraper, and further includes:

[0046] A plurality of bristle arrays, wherein the bristles of any one of the bristle arrays are installed on the surface of the roller brush shaft at a preset interval, and the bristles of any one of the bristle arrays are arranged on the surface of the roller brush shaft according to a preset curve, and the ends of the bristles of the plurality of bristle arrays that are not in contact with the roller brush shaft abut against the inner surface of the roller brush cylinder; an adhesive member, wherein the adhesive member is deployed at a preset position on the roller brush shovel; wherein, when the roller brush shaft rotates, the roller brush contacts the adhesive member, and the impurities are removed by the contact pressure of the plurality of bristle arrays.

[0047] exist Figure 3The anti-tangling dust brush design shown includes a roller brush 1, a roller brush shaft 2, and a roller brush scraper. This design effectively improves cleaning performance and anti-tangling capabilities. The anti-tangling dust brush has several bristle arrays mounted on the roller brush shaft 2. The bristles of each array are precisely installed on the surface of the roller brush shaft at preset intervals. This precise installation ensures that the bristles are evenly stressed during rotation and effectively clean the floor or other surfaces requiring cleaning. Furthermore, the bristles in each array are not simply arranged in a straight line, but rather in a preset curve on the surface of the roller brush shaft. This curved arrangement offers several advantages: it increases the contact area between the bristles and the floor, thus improving cleaning efficiency; the curved bristles generate a more complex motion trajectory during rotation, helping to remove or entangle hard-to-clean debris (such as hair and fibers); and it reduces interference between bristles, preventing jamming or tangling during use. The ends of all the bristles in the brush array that are not in contact with the roller shaft abut against the inner surface of the roller cylinder. This design not only ensures the stability of the bristles, but also allows the roller to maintain a certain rigidity when rotating, preventing deformation or damage caused by uneven force.

[0048] Adhesive components are deployed at preset positions on the roller brush shovel. The specific position and number of these adhesive components can be adjusted according to actual needs to ensure that they can effectively contact the roller brush and play their role.

[0049] As the roller brush shaft rotates, the roller brush comes into contact with the adhesive component, and the pressure from the bristle array removes debris adhering to the roller brush. The adhesive component is designed with a certain degree of adhesion, enabling it to adhere to and remove small particles of debris, thus promoting the shedding of the debris.

[0050] In summary, this anti-tangling dust brush not only improves cleaning efficiency and effectiveness but also significantly reduces the risk of bristle tangling and clogging. The curved arrangement and secure installation of the bristle array ensure uniformity and force in cleaning; while the introduction of adhesive components further enhances cleaning power and maintains the cleanliness of the roller brush.

[0051] Furthermore, a smart sweeping machine has the aforementioned anti-tangling dust brush.

[0052] The intelligent sweeping robot is equipped with an anti-tangling dust brush.

[0053] In summary, the embodiments of this application have at least the following technical effects:

[0054] First, a debris removal rate prediction network analyzes the roller brush humidity information and the first roller brush shaft speed to obtain the first predicted debris removal rate. The first roller brush shaft speed is randomly assigned based on the rated range of roller brush shaft speeds. Then, when the first predicted debris removal rate is greater than or equal to the convergence removal rate, the first roller brush shaft speed is set as the target roller brush shaft speed. The convergence removal rate is a user-preset removal rate threshold, ≥0.6. Finally, anti-tangling dust brush control is implemented based on the target roller brush shaft speed. This solves the technical problem in existing intelligent sweeping robots where the roller brush easily gets tangled with debris during cleaning, leading to reduced cleaning efficiency. By predicting the debris removal rate and dynamically adjusting the roller brush shaft speed, intelligent anti-tangling dust brush control is achieved, resulting in improved cleaning efficiency.

[0055] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0056] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0057] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An anti-entanglement dust brush control optimization method for an intelligent sweeper, characterized in that: The method comprises: The roller brush humidity information and the first roller brush shaft speed are analyzed by a debris cleaning rate prediction network to obtain a first debris predicted cleaning rate, wherein the first roller brush shaft speed is obtained by randomly assigning a value based on a rated range of the roller brush shaft speed; When the first debris predicted cleaning rate is greater than or equal to the convergence cleaning rate, the first roller brush shaft speed is set as the target roller brush shaft speed, wherein the convergence cleaning rate is a user-preset cleaning rate threshold, and the convergence cleaning rate is ≥0.6; Performing anti-entanglement dust brush control according to the target roller brush shaft speed; The debris removal rate prediction network construction step includes: According to the anti-entanglement dust brush model information, matching the anti-entanglement dust brush training data set, wherein the anti-entanglement dust brush training data set includes a roller brush humidity training data set, a roller brush shaft speed training data set and a debris cleaning rate identification data set; According to the debris cleaning rate identification data set, the roller brush humidity training data set and the roller brush shaft speed training data set are supervised to train a first debris cleaning rate prediction base network; According to the debris cleaning rate identification data set, the roller brush humidity training data set and the roller brush shaft speed training data set are supervised to train a second debris cleaning rate prediction base network; Until the Nth debris cleaning rate prediction base network is trained according to the debris cleaning rate identification data set, the roller brush humidity training data set and the roller brush shaft speed training data set are supervised; Constructing a fully connected layer for counting the weighted average of the first debris clearing rate prediction base network, the second debris clearing rate prediction base network, and the Nth debris clearing rate prediction base network to obtain the debris clearing rate prediction network, wherein the fusion weight is the ratio of the test accuracy of each debris clearing rate prediction base network of the first debris clearing rate prediction base network, the second debris clearing rate prediction base network, and the Nth debris clearing rate prediction base network to the sum of the N test accuracy rates; Associating and storing the anti-entanglement dust brush model information with the debris cleaning rate prediction network; According to the anti-entanglement dust brush model information, matching the anti-entanglement dust brush training data set includes: According to the anti-entanglement dust brush model information, an anti-entanglement dust brush test data set is obtained, wherein any anti-entanglement dust brush test data includes a test roller brush humidity, a test roller brush shaft speed and a plurality of debris cleaning rate record data; Performing mode analysis on the plurality of debris removal rate record data to obtain debris removal rate identification data; The test roller brush humidity is set as roller brush humidity training data, and the test roller brush shaft speed is set as roller brush shaft speed training data.

2. The anti-entanglement dust brush control optimization method for an intelligent sweeping machine according to claim 1, characterized in that: The roller brush humidity information and the first roller brush shaft speed are analyzed through the debris cleaning rate prediction network to obtain the first debris predicted cleaning rate, which includes: Receive anti-entanglement dust brush model information; Receive the roller brush humidity information through a humidity sensor installed on the inner layer of the anti-winding dust brush; According to the anti-entanglement dust brush model information, matching the debris cleaning rate prediction network; According to the anti-entanglement dust brush model information, the rated range of the roller brush shaft speed is obtained.

3. The anti-entanglement dust brush control optimization method for an intelligent sweeping machine according to claim 2, characterized in that: Receiving the roller brush humidity information by means of a humidity sensor installed on the inner layer of the anti-winding dust brush, including: The humidity sensor includes a first humidity sensor, a second humidity sensor and a third humidity sensor; Collecting first roller brush humidity monitoring timing information through the first humidity sensor; Collecting the second roller brush humidity monitoring timing information through the second humidity sensor; Collecting the third roller brush humidity monitoring timing information through the third humidity sensor; The first roller brush humidity monitoring timing information, the second roller brush humidity monitoring timing information and the third roller brush humidity monitoring timing information include at least five moments, and the collection time interval is 0.5 seconds or more and 0.2 seconds or more; Performing concentration value evaluation on the first roller brush humidity monitoring timing information, the second roller brush humidity monitoring timing information and the third roller brush humidity monitoring timing information respectively to obtain a first roller brush humidity concentration value, a second roller brush humidity concentration value and a third roller brush humidity concentration value; The roller brush humidity information is obtained by performing mean calculation on the first roller brush humidity concentration value, the second roller brush humidity concentration value and the third roller brush humidity concentration value.

4. The anti-entanglement dust brush control optimization method for an intelligent sweeping machine according to claim 1, characterized in that: Also includes: When the first debris predicted cleaning rate is less than the convergence cleaning rate, obtaining a second roller brush shaft speed, the convergence cleaning rate being a user-preset cleaning rate threshold; The roller brush humidity information and the second roller brush shaft speed are analyzed by the debris cleaning rate prediction network to obtain a second debris predicted cleaning rate, wherein the second roller brush shaft speed is obtained by randomly assigning a value based on a rated range of the roller brush shaft speed; When the second debris predicted cleaning rate is greater than or equal to the convergence cleaning rate, the second roller brush shaft speed is set as the target roller brush shaft speed, wherein the convergence cleaning rate is a user-preset cleaning rate threshold, and the convergence cleaning rate is ≥0.6; The anti-entanglement dust brush is controlled according to the target roller brush shaft speed.

5. The anti-entanglement dust brush control optimization method for an intelligent sweeping machine according to claim 4, characterized in that: Also includes: When the kth predicted debris cleaning rate of the kth roller brush shaft speed is less than the converged cleaning rate, clustering the k roller brush shaft speeds according to the roller brush shaft speed threshold to obtain multiple groups of roller brush shaft speeds; Extracting the maximum value of the predicted debris cleaning rate for the plurality of groups of roller brush shaft speeds respectively, and obtaining a plurality of maximum roller brush shaft speeds; Extracting, from the plurality of maximum roller brush shaft speeds, a first reference roller brush shaft speed, a second reference roller brush shaft speed, and a third reference roller brush shaft speed, three with relatively large predicted debris cleaning rates; Extracting the minimum value of the predicted debris cleaning rate from the multiple groups of roller brush shaft speeds to obtain multiple minimum roller brush shaft speeds; According to the roller brush shaft speed adjustment step interval, taking the first reference roller brush shaft speed, the second reference roller brush shaft speed and the third reference roller brush shaft speed as adjustment targets, traversing the multiple minimum roller brush shaft speeds and performing random adjustments several times respectively, to obtain a roller brush shaft speed expansion value; When the roller brush shaft speed expansion value has a roller brush shaft speed that satisfies the convergence cleaning rate, it is set as the target roller brush shaft speed; Otherwise, the roller brush shaft speed expansion value is iteratively updated.

6. An intelligent sweeping machine, characterized in that: It includes an anti-entanglement dust brush, which is used to implement the anti-entanglement dust brush control optimization method for an intelligent sweeping machine as described in any one of claims 1 to 5; The anti-entanglement dust brush comprises: Roller brush, roller brush shaft and roller brush shovel; A plurality of bristle arrays, the bristles of any of the bristle arrays are installed on the surface of the roller brush shaft according to a preset interval, the bristles of any of the bristle arrays are arranged on the surface of the roller brush shaft according to a preset curve, and the ends of the bristles of the plurality of bristle arrays that are not in contact with the roller brush shaft abut against the inner surface of the roller brush cylinder; An adhesive member, the adhesive member being disposed at a preset position of the roller brush shovel; When the roller brush shaft rotates, the roller brush contacts the adhesive component, and the debris is removed by the resistance pressure of the plurality of bristle arrays.

Images