Washing machine control method

Abstract

The invention provides a washing machine control method which comprises the following steps: acquiring a gesture track to generate an instruction set, matching a washing program by using weight and spectral data, receiving a gesture correction instruction to adjust parameters, acquiring water consumption and power consumption data in real time and mapping a chart, and generating an offline scheme by using a local program when a network is abnormal. Visually impaired users are guided through vibration of the tactile navigation bar, and dehydration time is adjusted through a neural network model. Program intelligent matching and parameter self-adaption are achieved through gesture clustering and fuzzy rule adjustment, an offline scheme is evaluated and optimized through historical programs, visual impaired experience is improved through multi-dimensional vibration feedback, water is dynamically controlled through a neural network model, and the problems of poor adaptability and the like are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent household appliance control, in particular to a washing machine control method. BACKGROUND

[0002] With the rapid development of Internet of Things technology and artificial intelligence algorithms, the field of intelligent household appliances gradually realizes the upgrade from device interconnection to behavior perception. Although the existing washing machine control method introduces sensor data collection and basic program matching functions, there are still technical bottlenecks in user interaction personalization, washing parameter dynamic adjustment and special scene adaptability. For example, traditional gesture recognition relies on fixed action templates and lacks self-learning ability for user operation habits; washing program matching is only based on a single weight parameter and does not fully integrate multi-dimensional analysis of material characteristics; the program reuse strategy in offline mode lacks a parameter compensation mechanism, resulting in unstable washing effect in complex scenarios.

[0003] The traditional washing machine control method has the following defects: first, the human-computer interaction mode is single, and visually impaired users cannot complete the operation through the visual interface, and the tactile feedback design has not formed a systematic priority coding; second, the program adaptation accuracy is insufficient, the fuzzy logic rule base lacks a dynamic update mechanism, and cannot adapt to the material differences and dirt level changes of different batches of clothes; third, the energy consumption visualization degree is low, only the cumulative data can be displayed, and real-time dynamic mapping and abnormal warning of water and electricity consumption have not been realized; fourth, the drying degree judgment relies on fixed threshold, without training intelligent model combined with historical data, the accuracy of dehydration time adjustment needs to be improved.

[0004] Therefore, it is necessary to design a washing machine control method that integrates user operation habit self-learning, multi-sensor data fusion analysis and intelligent algorithm optimization, to solve the problems of poor interaction adaptability, insufficient program matching accuracy, inconvenience of special user operation and extensive energy management in the prior art, and to improve the intelligent control level of the washing machine in different use scenarios through gesture trajectory clustering, fuzzy rule weight adjustment, offline parameter compensation and neural network drying prediction and other innovative mechanisms. SUMMARY

[0005] In view of this, the present application provides a washing machine control method, which aims to solve the problem of how to realize personalized learning of user operation habits through time series analysis of gesture trajectory and dynamic time warping clustering algorithm, improve washing program matching accuracy by combining multi-sensor data fusion and deviation trigger weight adjustment mechanism of fuzzy rule base, optimize offline mode parameter compensation strategy by using historical program effectiveness evaluation model, enhance the operation experience of visually impaired users by multi-dimensional vibration feedback coding, and dynamically control the dehydration time by using a neural network model.

[0006] In one aspect, the present application provides a washing machine control method, comprising:

[0007] A user operation habit database is established, the user's operation gesture trajectory is collected by an image collection device arranged on the operation interface of the washing machine, and a gesture control instruction set is generated;

[0008] Based on the weight sensor and the spectrum recognition device arranged in the washing machine, the weight data and the material characteristic parameters of the clothes to be washed are collected, and the washing program is matched through a fuzzy logic algorithm;

[0009] The correction instruction input by the user through the gesture is received, the user's parameter adjustment demand for the washing program is analyzed according to the gesture control instruction set, and a corrected washing parameter combination is generated;

[0010] Through the flow sensor arranged at the water inlet of the washing machine and the current sensor arranged at the motor, the water consumption and the power consumption data in the washing process are collected in real time, and are mapped to the transparent display area of the operation interface in the form of a preset dynamic chart;

[0011] When the network connection is detected to be abnormal, the offline control mode is switched to, a default running scheme is generated based on the last three effective washing program parameters stored locally, and after the network is restored, whether the historical operation data is synchronized to the cloud server is prompted to the user;

[0012] A tactile navigation bar is arranged on the operation interface of the washing machine, the function area of a visually impaired user is guided through vibration feedback of different frequencies, and the vibration frequency of the tactile navigation bar is associated with the operation priority of the function area.

[0013] When it is detected that the washing program is ended, the drying degree of the clothes is judged through the humidity sensor and the temperature sensor arranged in the washing machine, and the dehydration time is prolonged or shortened, and the judgment logic is based on a neural network model trained by historical drying data.

[0014] Further, the gesture control instruction set is generated, including: time sequence segmentation of the collected gesture trajectory, continuous trajectory is divided into basic action units according to the trajectory direction change point, each basic action unit contains starting coordinates, ending coordinates and direction change rate; a motion unit feature vector library is established, and the dynamic time warping algorithm is used to calculate the trajectory similarity, and the trajectories with a similarity coefficient greater than a preset threshold are clustered into the same instruction category;

[0015] The dynamic time warping algorithm is used to calculate the trajectory similarity, including: constructing a trajectory time sequence matrix, the row dimension is the time point, and the column dimension is the coordinate component; the Euclidean distance of the matrix element is calculated as the local similarity, and the dynamic programming is used to accumulate the path similarity; the similarity coefficient is defined as the ratio of the total distance of the warping path to the shortest path distance, and when the ratio is less than 0.8, it is determined that the trajectories are similar.

[0016] Further, the washing program matching through the fuzzy logic algorithm comprises:

[0017] A fuzzy rule base is established, the rule antecedent is a combination of weight data interval and material characteristic parameters, and the rule consequent is a parameter combination of washing time, water level height and washing rotation speed;

[0018] The fuzzy reasoning result is de-fuzzied to generate basic program parameters;

[0019] The updating mechanism of the fuzzy rule base is that when the actual running parameters of the same rule in the historical washing data deviate from the reasoning parameters by more than 10%, the rule weight adjustment is triggered, and the adjustment amplitude is the deviation value multiplied by the square root of the historical execution times.

[0020] Further, the parameter adjustment demand of the user for the washing program according to the gesture control instruction set comprises:

[0021] The motion mode of the gesture trajectory is recognized, and the motion mode comprises linear sliding, curve rotation and point pressing;

[0022] A mode-parameter mapping relationship is established, linear sliding corresponds to linear adjustment parameters, curve rotation corresponds to cyclic switching parameters, and point pressing corresponds to parameter reset.

[0023] Further, the default running scheme based on the last three effective washing program parameters stored locally comprises:

[0024] A historical program effectiveness evaluation model is established, the evaluation indexes comprise program completion rate, parameter adjustment times and washing effect score, and the washing effect score is obtained through user feedback data statistics;

[0025] The program matching priority is set, the historical program of the same material type is preferentially matched, and the material type matching degree is calculated through the cosine similarity of spectral characteristic parameters.

[0026] Further, the program completion rate is the ratio of the actual execution time to the preset time;

[0027] The program with more than three times of parameter adjustment times is marked as an unstable program, and the matching priority thereof is reduced. Further, the function area positioning of the visual impairment user through vibration feedback of different frequencies comprises:

[0028] An operation heat map of the function area is established, the heat distribution is generated according to the historical operation dwell time and click times of the user, and the intensity of the vibration feedback is set according to the heat value;

[0029] A dynamic attenuation strategy of the vibration feedback is set, when the user continuously touches the same area, the vibration frequency gradually increases, and after leaving, the frequency gradually attenuates;

[0030] The operation priority is associated with a vibration feature, the vibration feature including frequency, amplitude and duration, a high-priority area adopts a high-frequency short vibration mode, a secondary priority adopts a medium-frequency long vibration mode, and a low priority adopts a low-frequency intermittent vibration mode.

[0031] Further, the judgment logic based on the neural network model trained by historical drying data includes:

[0032] Collect humidity, temperature, washing time and drying result in historical washing data to construct a training data set;

[0033] A three-layer fully connected neural network is constructed, the input layer is normalized data of humidity, temperature and washing time, the number of hidden layer neurons is determined by grid search, and the output layer is a drying degree probability value;

[0034] The real-time collected humidity, temperature and washing time are input into the neural network model, when the output drying degree probability value is greater than a preset threshold, it is determined that drying is completed, otherwise, the extended dehydration time is calculated according to the difference between the probability value and the preset threshold.

[0035] Compared with the prior art, the beneficial effects of the present application are that the washing machine control method of the present application can realize personalized interaction and meet different user needs by establishing a user operation habit database and a gesture control instruction set. The washing program adaptability can be improved by collecting data with multiple sensors and matching the washing program with a fuzzy logic algorithm. Real-time acquisition of water consumption and power consumption data and visualization facilitate users to master the energy consumption situation. The offline control mode and historical data synchronization function in network exception ensure the stability and data integrity of use. The tactile navigation bar facilitates the operation of visually impaired users. The dehydration time is adjusted based on the neural network model to judge the drying degree of clothes, which can improve the washing effect, and the intelligent, humanized and energy-saving level of the washing machine is improved as a whole. BRIEF DESCRIPTION OF DRAWINGS

[0036] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are for purposes of illustration only and are not considered a limitation of the present application. Moreover, like reference numerals are used to designate identical components throughout the specification. In the drawings:

[0037] Figure 1 The washing machine control method flow chart of the embodiment system of the present application; DETAILED DESCRIPTION

[0038] Exemplary embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms without being 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 application to those skilled in the art. It should be noted that the embodiments of the present application and features in the embodiments can be combined with each other without conflict unless otherwise stated. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0039] Referring to Figure 1 As shown in some embodiments of the present application, a washing machine control method includes:

[0040] S100, a user operation habit database is established, the operation gesture trajectory of the user is collected by an image collection device arranged on the operation interface of the washing machine, and a gesture control instruction set is generated;

[0041] S200, the weight data and material characteristic parameters of the clothes to be washed are collected based on a weight sensor and a spectrum recognition device arranged inside the washing machine, and a washing program is matched by a fuzzy logic algorithm;

[0042] S300, a correction instruction input by the user through a gesture is received, the parameter adjustment requirement of the user for the washing program is analyzed according to the gesture control instruction set, and a corrected washing parameter combination is generated;

[0043] S400, water consumption and power consumption data in the washing process are collected in real time by a flow sensor arranged at the water inlet of the washing machine and a current sensor arranged at the motor, and are mapped to a transparent display area of the operation interface in the form of a preset dynamic chart;

[0044] S500, when a network connection exception is detected, an offline control mode is switched to, a default running scheme is generated based on the last three effective washing program parameters stored locally, and after the network is restored, the user is prompted whether to synchronize the historical operation data to a cloud server;

[0045] S600, a tactile navigation bar is arranged on the operation interface of the washing machine, and the positioning function area of the visually impaired user is guided by vibration feedback of different frequencies, and the vibration frequency of the tactile navigation bar is associated with the operation priority of the function area;

[0046] S700, when it is detected that the washing program is ended, the drying degree of the clothes is judged by a humidity sensor and a temperature sensor arranged inside the washing machine, and the dehydration time is extended or shortened, and the judgment logic is based on a neural network model trained by historical drying data.

[0047] Specifically, the image acquisition device continuously operates to capture various gesture actions of the user on the operation interface in real time. After the gesture trajectory is acquired, time sequence segmentation processing is performed on the gesture trajectory, and the continuous trajectory is finely divided into basic action units according to the direction change points of the trajectory, and each basic action unit accurately records the starting coordinates, the ending coordinates and the direction change rate. Subsequently, a motion unit feature vector library is established, and a dynamic time warping algorithm is used to calculate the trajectory similarity. A trajectory time sequence matrix is constructed, the row dimension of the matrix represents the time point, and the column dimension represents the coordinate component. The Euclidean distance of the matrix element is calculated as the local similarity, and the dynamic programming is used to accumulate the path similarity. The trajectories with a similarity coefficient greater than a preset threshold are clustered into the same instruction category.

[0048] It can be understood that establishing the user operation habit database and generating the gesture control instruction set can enable the user to operate the washing machine through simple gesture operation, and improve the convenience and interest of operation. The dynamic time warping algorithm ensures the accuracy of the similarity judgment of the gesture trajectory, improves the precision of the instruction recognition, and enables the user's gesture instruction to be more accurately analyzed and executed.

[0049] Specifically, the weight sensor measures the weight of the laundry to be washed in real time, and the spectrum recognition device accurately analyzes the material of the laundry to obtain the material characteristic parameters. A fuzzy rule library is established, the rule antecedent is set as the combination of the weight data interval and the material characteristic parameters, and the rule consequent is the combination of the washing time, the water level height, the washing speed and the like. The fuzzy inference result is de-fuzzified to generate the basic program parameters. At the same time, the fuzzy rule library has an updating mechanism. When the actual running parameters of the same rule in the historical washing data deviate from the inference parameters by more than a certain percentage, the rule weight adjustment is triggered, and the adjustment amplitude is determined according to the deviation value and the historical execution times.

[0050] It can be understood that the use of the weight sensor and the spectrum recognition device to collect laundry information, combined with the fuzzy logic algorithm to match the washing program, can accurately select the appropriate washing program according to the actual situation of the laundry, and improve the washing effect. The updating mechanism of the fuzzy rule library can enable the washing machine to continuously optimize the washing program according to the historical washing conditions, adapt to the washing needs of different types of laundry, and improve the intelligence of the washing machine.

[0051] Specifically, when the user inputs a correction instruction through a gesture, the system first identifies the motion mode of the gesture trajectory, such as straight line sliding, curve rotating and point pressing. Then a mode-parameter mapping relationship is established, the straight line sliding corresponds to the linear adjustment parameter, the curve rotating corresponds to the cycle switching parameter, and the point pressing corresponds to the parameter reset. According to this mapping relationship, the user's parameter adjustment demand for the washing program is analyzed, and the corrected washing parameter combination is generated.

[0052] It can be understood that allowing the user to input the correction instruction through the gesture can meet the user's demand for personalization adjustment of the washing program. The clear mode-parameter mapping relationship enables the user's adjustment intention to be accurately understood and executed, improves the user's control ability over the washing process, and enhances the user experience.

[0053] Specifically, the flow sensor monitors the water flow in real time during the water inlet process of the washing machine, and the current sensor monitors the current change of the motor in real time, so as to obtain the water consumption and power consumption data in the washing process. These data are displayed in the form of a preset dynamic chart in the transparent display area of the operation interface, and the dynamic chart changes accordingly with the real-time changes of the water consumption and power consumption.

[0054] It can be understood that real-time acquisition and display of the water consumption and power consumption data can enable the user to intuitively understand the energy consumption in the washing process, and enhance the user's energy-saving awareness. Displaying the data in the form of a dynamic chart makes the information more intuitive and vivid, and facilitates the user to view and understand.

[0055] Specifically, the washing machine monitors the network connection state in real time, and automatically switches to an offline control mode when detecting network connection abnormity. A historical program effectiveness evaluation model is established, and the evaluation indexes include program completion rate, parameter adjustment times, and washing effect score, etc. The washing effect score is calculated through user feedback data. The program matching priority is set, and the historical program with the same material type is preferentially matched, and the material type matching degree is calculated through the similarity of spectral characteristic parameters. The default running scheme is generated based on the last three effective washing program parameters stored locally. When the network is restored, a prompt box is popped up to ask the user whether to synchronize the historical operation data to the cloud server.

[0056] It can be understood that the offline control mode and the default running scheme ensure that the washing machine can still operate normally in the case of network abnormity, and do not affect the user's washing demand. The historical program effectiveness evaluation model and the program matching priority setting can make the default running scheme as much as possible to meet the user's washing demand. After the network is restored, the user is prompted to synchronize the data, which is convenient for the user to view and manage the washing history record on different devices.

[0057] Specifically, a tactile navigation bar is installed on the operation interface of the washing machine, and an operation heat map of the function area is established. The heat distribution is generated according to the user's historical operation dwell time and click times, and the intensity of vibration feedback is set according to the heat value. The dynamic attenuation strategy of vibration feedback is set, and when the user continuously touches the same area, the vibration frequency gradually increases, and the frequency gradually decays after leaving. The operation priority is associated with the vibration characteristics, such as using a high-frequency short vibration mode for high-priority areas, a medium-frequency long vibration mode for secondary priority areas, and a low-frequency intermittent vibration mode for low-priority areas.

[0058] It can be understood that setting the tactile navigation bar and guiding the visually impaired user to locate the function area through vibration feedback of different frequencies embodies the humanized design of the washing machine, providing convenience for the visually impaired user and enabling them to operate the washing machine independently. The operation of the heat map and the setting of the dynamic attenuation strategy enable the visually impaired user to more accurately perceive the function area, improving the accuracy and efficiency of operation.

[0059] Specifically, after the washing program ends, the humidity sensor and the temperature sensor collect the humidity and temperature data of the clothes in real time. The humidity, temperature, washing time and drying result in the historical washing data are collected to construct a training data set. A three-layer fully connected neural network is constructed, the input layer is the normalized data of humidity, temperature and washing time, the number of hidden layer neurons is determined by grid search, and the output layer is the drying degree probability value. The real-time collected humidity, temperature and washing time are input into the neural network model, when the output drying degree probability value is greater than the preset threshold, it is determined that the clothes are dry, otherwise the dehydration time is extended according to the difference between the probability value and the preset threshold.

[0060] It can be understood that using the humidity sensor and the temperature sensor in combination with the neural network model to determine the drying degree of the clothes and adjust the dehydration time can realize accurate control of the drying degree of the clothes, avoid the situation of over-dewatering or insufficient dehydration of the clothes, improve the quality of the clothes after washing, and save energy at the same time.

[0061] It should be noted that:

[0062] In the specification provided herein, a large number of specific details are described. However, it can be understood that the embodiments of the present application can be practiced without these specific details. In some examples, well-known structures and technologies are not shown in detail in order not to obscure the understanding of the present specification.

[0063] In addition, those skilled in the art can understand that although some embodiments described herein include certain features included in other embodiments but not others, the combination of features of different embodiments means within the scope of the present application and forms different embodiments.

[0064] The above is only the preferred specific implementation of the present application, but the protection scope of the present application is not limited to this. Any changes or replacements easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A washing machine control method, characterized in that: include: Establish a user operation habit database, collect the user's operation gesture trajectory through the image acquisition device set on the washing machine operation interface, and generate a gesture control instruction set; Based on the weight sensor and spectrum recognition device installed inside the washing machine, the weight data and material characteristic parameters of the clothes to be washed are collected, and the washing program is matched through the fuzzy logic algorithm; receiving a correction instruction input by a user through gestures, parsing the user's parameter adjustment requirements for the washing program according to the gesture control instruction set, and generating a corrected washing parameter combination; The flow sensor installed at the water inlet of the washing machine and the current sensor installed at the motor collect water consumption and power consumption data in real time during the washing process, and map them to the transparent display area of ​​the operation interface in the form of preset dynamic graphs; When a network connection anomaly is detected, the system switches to offline control mode, generates a default operation plan based on the last three valid washing program parameters stored locally, and prompts the user whether to synchronize historical operation data to the cloud server after the network is restored; A tactile navigation bar is set up on the washing machine operation interface to guide visually impaired users to locate functional areas through vibration feedback of different frequencies. The vibration frequency of the tactile navigation bar is associated with the operation priority of the functional area; When the end of the washing program is detected, the humidity sensor and temperature sensor installed inside the washing machine are used to judge the dryness of the clothes and extend or shorten the dehydration time. The judgment logic is based on a neural network model trained with historical drying data.

2. A washing machine control method according to claim 1, characterized in that: The method of generating a gesture control instruction set includes: segmenting the collected gesture trajectory into time series, dividing the continuous trajectory into basic action units according to the trajectory direction change points, each basic action unit including the starting coordinate, the ending coordinate and the direction change rate; establishing an action unit feature vector library, and calculating the trajectory similarity using a dynamic time warping algorithm, clustering trajectories with a similarity coefficient greater than a preset threshold into the same instruction category; The dynamic time warping algorithm is used to calculate trajectory similarity, including: constructing a trajectory time series matrix with row dimensions as time points and column dimensions as coordinate components; calculating the Euclidean distance of matrix elements as local similarity, and using dynamic programming to accumulate path similarity; the similarity coefficient is defined as the ratio of the total distance of the warped path to the shortest path distance. When the ratio is less than 0.8, the trajectory is determined to be similar.

3. A washing machine control method according to claim 1, characterized in that: The matching of washing programs by fuzzy logic algorithm includes: A fuzzy rule base is established, where the rule antecedents are the combination of weight data interval and material characteristic parameters, and the rule consequences are the combination of washing time, water level, and washing speed parameters. Defuzzify the fuzzy reasoning results and generate basic program parameters; The updating mechanism of the fuzzy rule base is: when the deviation between the actual operation parameters and the inference parameters of the same rule in the historical scrubbing data exceeds 10%, the rule weight adjustment is triggered, and the adjustment range is the deviation value multiplied by the square root of the historical execution times.

4. A washing machine control method according to claim 1, characterized in that: The step of analyzing the user's demand for adjusting the parameters of the washing program according to the gesture control instruction set includes: Identify the motion pattern of the gesture trajectory, wherein the motion pattern includes linear sliding, curved rotation, and tapping; Establish a mode-parameter mapping relationship. Linear sliding corresponds to linear adjustment of parameters, curved rotation corresponds to cyclic switching of parameters, and tapping corresponds to reset of parameters.

5. A washing machine control method according to claim 1, characterized in that: The generation of a default operation plan based on the three most recent valid washing program parameters stored locally includes: Establish a historical program effectiveness evaluation model, with evaluation indicators including program completion rate, number of parameter adjustments, and washing effect score, which is obtained through user feedback data statistics; Set the program matching priority to give priority to matching historical programs with the same material type. The material type matching degree is calculated by the cosine similarity of the spectral feature parameters.

6. A washing machine control method according to claim 5, characterized in that: The program completion rate is the ratio of actual execution time to preset time; A program whose parameter has been adjusted more than three times is marked as an unstable program, and its matching priority is lowered.

7. A washing machine control method according to claim 1, characterized in that: The method of guiding the visually impaired user to locate the functional area through vibration feedback of different frequencies includes: Establish an operation heat map for the functional area, generate a heat distribution based on the user's historical operation dwell time and click count, and set the intensity of the vibration feedback based on the heat value; Set a dynamic attenuation strategy for vibration feedback. When the user continues to touch the same area, the vibration frequency gradually increases, and the frequency gradually decreases after the user leaves the area. The operation priority is associated with vibration characteristics, which include frequency, amplitude and duration. The high-priority area adopts a high-frequency short vibration mode, the second-priority area adopts a medium-frequency long vibration mode, and the low-priority area adopts a low-frequency intermittent vibration mode.

8. A washing machine control method according to claim 1, characterized in that: The judgment logic is based on a neural network model trained with historical drying data and includes: Collect humidity, temperature, washing time, and drying results from historical washing data to build a training dataset; A three-layer fully connected neural network was constructed. The input layer was the normalized data of humidity, temperature, and washing time. The number of neurons in the hidden layer was determined by grid search. The output layer was the probability value of dryness. The real-time collected humidity, temperature and washing time are input into the neural network model. When the output dryness probability value is greater than the preset threshold, it is judged to be dry. Otherwise, the dehydration time is extended according to the difference between the probability value and the preset threshold.

Images