Electric Rice Cooker Control System Based on Electrooculography

By using an electrooculography (EOG)-based rice cooker control system, combined with a voting classifier algorithm and a multi-module design, the adaptability and usability issues of traditional EOG control systems have been solved, enabling convenient operation and efficient control of the smart rice cooker.

CN114755946BActive Publication Date: 2024-11-26SOUTH CHINA UNIV OF TECH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202210268969.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-18
Publication Date
2024-11-26
Estimated Expiration
2042-03-18

AI Technical Summary

Technical Problem

Traditional electrooculography (EOG) control systems are not robust to different users and cannot meet the convenient use needs of people with disabilities such as deafness and muteness. Furthermore, existing smart rice cookers lack adaptive interaction solutions.

Method used

An electrooculogram (EOG)-based rice cooker control system was adopted. The system utilizes an EOG acquisition module, an algorithm module, a communication module, and a control module, combined with a voting classifier algorithm for blink detection and function analysis. An operation interface for training and usage modes was designed, which includes "start" and "undo" function buttons to achieve accurate transmission of user function commands.

Benefits of technology

It improves the system's adaptability and robustness, making it suitable for different users, simplifies the operation process, enhances the user experience for people with disabilities such as the deaf and mute, and has broad application prospects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114755946B_ABST
    Figure CN114755946B_ABST
Patent Text Reader

Abstract

The invention discloses an electric rice cooker control system based on electrooculography, comprising: an electrooculography acquisition module, comprising an acquisition module and an operation interface module, the acquisition module acquires electrooculography signals and sends them to an algorithm module, the operation interface module displays different contents to interact with users, first, buttons set according to the functions of the electric rice cooker flash in a set order, second, prompts for confirming the execution of the electric rice cooker functions and prompts for completing the electric rice cooker functions; the algorithm module uses the acquired electrooculography signals to parse instructions through a classification algorithm and sends them to a communication module; the communication module receives instructions from the algorithm module and sends them to a control module and an operation interface module, and sends completion information of the control module to the operation interface module; the control module controls the electric rice cooker to execute functions according to the instructions of the communication module, and feeds back information to the communication module after the functions are completed. The invention enables users with disabilities such as the deaf and dumb to effectively control and use the electric rice cooker.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of human-computer interaction, in particular to an electrooculographic-based rice cooker control system. Background Art

[0002] Traditional electrooculogram control, blink judgment, and data analysis are designed based on cumbersome prior knowledge. The threshold design of the electrooculogram peak is complex and has poor robustness for different users, which may limit its application in large-scale samples. The use of training and testing methods for characteristic signals to judge blinks enhances the applicability of the system to different users.

[0003] Traditional artificial intelligence rice cookers mainly use technologies such as voice interaction to control the rice cooker, but cannot take into account the use problems of people with disabilities such as the deaf and mute. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the current artificial intelligence rice cookers, systems and interaction methods do not take into account the convenient use of users with disabilities such as the deaf and dumb. An electric cooker control system based on electrooculography is proposed, which can be used by users with disabilities such as the deaf and dumb to effectively control and use the rice cooker.

[0005] To achieve the above object, the technical solution provided by the present invention is: an electric cooker control system based on electrooculography, comprising:

[0006] The electrooculogram acquisition module is divided into two parts, one is an acquisition module, and the other is an operation interface module; the electrooculogram signal data acquired by the acquisition module is sent to the algorithm module; the operation interface of the operation interface module displays the following contents: buttons set according to the functions of the rice cooker, the buttons flash in the set order, and prompts for confirming the execution of the rice cooker function and completing the rice cooker function;

[0007] The algorithm module uses the collected electrooculogram data to perform blink detection through a classification algorithm, determines whether to select a button displayed on the operation interface and which button is selected based on the result of the blink detection, further analyzes the function required by the user according to the set rules, obtains the corresponding instruction based on the function, and sends the instruction to the communication module;

[0008] The communication module receives various instructions sent by the algorithm module and transmits them to the control module before the electric cooker performs a function. After the electric cooker completes a function, the communication module sends the information of the function completion to the operation interface module.

[0009] The control module controls the rice cooker to complete the corresponding function according to the instruction received by the communication module, and transmits the information of the function completion to the communication module.

[0010] Furthermore, the acquisition module needs to place electrodes around the user's eyes, the electrodes are connected to a signal amplifier, and then the signal amplifier is connected to the algorithm module, and the collected electrooculogram signal data will be sent to the algorithm module.

[0011] Furthermore, in the operation interface module, there are three different display modes: training mode, use mode and prompt mode;

[0012] In the training mode, the operation interface only displays one button, which flashes continuously. In the first case, the user is informed in advance that the button flashes and then blinks. In the second case, the user is informed in advance that the button flashes and then does not blink.

[0013] In the use mode, the operation interface is arranged with 7 buttons: "Porridge", "Hot Rice", "Keep Warm", "Porridge", "Steam", "Start" and "Cancel". Among them, "Porridge", "Hot Rice", "Keep Warm", "Porridge" and "Steam" are function buttons, and "Start" and "Cancel" are decision buttons. The 7 buttons flash according to the set time interval and set order. Users are told that if they want to trigger the function of a button, they need to blink immediately after the button flashes.

[0014] Prompt mode prompts the user to execute and complete functions; after the algorithm module determines that a certain function is required by the user, the operation interface displays that the function is successfully selected and executed; after the function is completed, the operation interface displays that the function is successfully completed.

[0015] Further, the algorithm module receives the electrooculogram signal data, and performs blink detection on the electrooculogram signal data during each round of button blinking in combination with the order of button blinking on the operation interface; the blink detection first determines the number of blinking actions performed by the user after the button blinks; if there is zero blinking action or blinking actions are performed after multiple button blinks, no button is selected, and the operation interface continues with the next round of button blinking; if only one button blinks and a blink action is performed, it is further determined which button the blinking action is after, so as to determine which button the user has triggered; specifically, in the blink detection, the electrooculogram signal data after the button blinks is classified to determine whether a blink action is performed after the button is clicked;

[0016] The specific content of the classification algorithm is divided into two parts: training and use. The training is divided into two parts: data preprocessing and model building. The use is divided into two parts: data preprocessing and classification using the model.

[0017] In the data preprocessing part of the training, the operation interface module is in training mode. The electrooculogram signal data collected in Case 1 and Case 2 are labeled as blinking and no blinking. The system takes the button flashing time point as the starting point and intercepts a fixed length of electrooculogram signal segment; first, the signal segment is band-pass filtered as a whole to remove the interference of high-frequency electromyography and low-frequency head shaking, and then the difference is performed to obtain the differential signal segment. The waveform feature parameters that need to be extracted include: the maximum value a of the filtered signal segment max , the maximum value d of the differential signal segment max , the minimum value d of the differential signal segment min The time point of the maximum and minimum values ​​of the differential signal segment is t p and t v , the mean value a of the filtered signal segment ave , waveform span t d Given by the formula:

[0018] t d =|t p -t v |

[0019] a max d max d min ,t p ,t v 、a ave As 6 feature data;

[0020] The construction of the trained model, χ represents the feature space, Y represents the label set, -1 represents no blinking, 1 represents blinking, R n represents the real number domain, N represents the total amount of data, and the i-th eigenvector The i-th label y i ∈Y={-1,+1}, there exists 1≤i≤N; each sample point consists of features and labels, and the training data set is expressed as:

[0021] T={(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x N ,y N )}

[0022] The classification model is constructed using the training data set T. Due to the small number of input features, a voting classifier G(x) model is used to improve the classification effect. The basic classifiers used are: K-nearest neighbor classifier g 1 (x), Linear Discriminant Analysis classifier g 2 (x), support vector machine classifier g 3 (x); g m(x) represents the mth basic classifier, a m is the linear combination of g m (x) corresponding to the coefficient, and Use features and labels to train the basic classifier g separately 1 (x),g 2 (x),g 3 (x); The voting classifier G(x) is essentially a boosting method, and the final classifier is a linear combination of the basic classifiers:

[0023]

[0024] Calculate g m The classification error rate e on the training data of (x) m :

[0025]

[0026] Among them, the coefficient

[0027] According to the above formula, we get the voting classifier G(x);

[0028] For the data preprocessing part used, the operation interface module is in the use mode, and after the button flashes, the same data processing method as the training is adopted;

[0029] The classification part of the model used is a max d max d min ,t p ,t v 、a ave These 6 feature data are put into the trained voting classifier, and the classification output is used as the result of blink detection;

[0030] When two consecutive selected buttons are respectively a function button and a decision button, a judgment is made: the function button is selected first, and then the cancel button is selected, and it is considered that the previously selected function button is canceled; the function button is selected first, and then the start button is selected, and it is considered that the function corresponding to the selected function button is determined to be executed. The algorithm module sends the instruction corresponding to the function execution to the communication module, and displays the successful selection of this function in the operation interface module in the electrooculogram acquisition module.

[0031] Furthermore, the communication module is a Bluetooth interface capable of two-way communication. Before the instruction is issued, it specifically receives the instruction information sent by the algorithm module, and controls the operation of the rice cooker according to the instruction information. After the function corresponding to the instruction is completed, the function completion information is sent to the operation interface module.

[0032] Further, the control module receives the instruction sent by the communication module, and then controls the rice cooker to execute the function corresponding to the instruction, and sets a timer and performs temperature detection at the same time, wherein the time of the timer is the completion time corresponding to the set function; when the timer timing ends and the temperature is greater than the completion temperature corresponding to the set function, it is judged that the function is completed, and then the information of the completion of the function is fed back to the communication module.

[0033] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0034] 1. The present invention adopts a voting classifier algorithm to perform blink detection, which is more adaptable and robust than the traditional method of directly setting a threshold to detect the peak. For different users or different time periods of the same user, the training model can be updated according to the user's EEG signal to achieve better results.

[0035] 2. The present invention designs a communication module, which is responsible for the communication between the algorithm module and the control module, transmits the instructions output by the algorithm module to the control module, and transmits the information on whether the control module completes the instructions to the electrooculogram acquisition module, thereby communicating multiple modules and improving the availability of the entire system.

[0036] 3. Compared with the algorithms used in other EEG control systems, the algorithm used in the present invention has a simpler model, smaller parameters and calculation amount, and faster speed. Therefore, it has a wide range of use space in actual application scenarios and has broad application prospects.

[0037] 4. When designing the button selection of the electric rice cooker, the present invention sets two function buttons of "start" and "cancel", which can prevent accidental button touching due to blinking to a certain extent, thereby improving operating efficiency.

[0038] 5. The present invention provides a new idea for intelligent control of electric rice cookers, which uses a physiological electrical signal to control the use of electrical appliances, and has a very broad prospect in the practical application of smart homes. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the system of the present invention.

[0040] Figure 2 A schematic diagram of the screen button layout for the operation display module.

[0041] Figure 3 This is a flowchart of the blink detection process. DETAILED DESCRIPTION

[0042] The present invention is further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0043] See also Figure 1As shown, this embodiment discloses a rice cooker control system based on electrooculography, including the following functional modules:

[0044] The electrooculogram acquisition module is divided into two parts, one is an acquisition module, and the other is an operation interface module; the electrooculogram signal data acquired by the acquisition module is sent to the algorithm module; the operation interface of the operation interface module displays the following contents: buttons set according to the functions of the rice cooker, the buttons flash in the set order, and prompts for confirming the execution of the rice cooker function and completing the rice cooker function;

[0045] The algorithm module uses the collected electrooculogram data to perform blink detection through a classification algorithm, determines whether to select a button displayed on the operation interface and which button is selected based on the result of the blink detection, further analyzes the function required by the user according to the set rules, obtains the corresponding instruction based on the function, and sends the instruction to the communication module;

[0046] The communication module receives various instructions sent by the algorithm module and transmits them to the control module before the electric cooker performs a function. After the electric cooker completes a function, the communication module sends the information of the function completion to the operation interface module.

[0047] The control module controls the rice cooker to complete the corresponding function according to the instruction received by the communication module, and transmits the information of the function completion to the communication module.

[0048] Furthermore, the acquisition module needs to place electrodes (Ag / AgCl electrodes) around the user's eyes, the electrodes are connected to a signal amplifier, model Neuroscan NuAmps, and then the signal amplifier is connected to the algorithm module, and the collected electrooculogram signal data will be sent to the algorithm module.

[0049] Furthermore, in the operation interface module, there are three different display modes: training mode, use mode and prompt mode.

[0050] In the training mode, the operation interface only displays one button, which flashes continuously. In the first case, the user is informed in advance that the button flashes and then blinks; in the second case, the user is informed in advance that the button flashes and then does not blink.

[0051] In the use mode, the operation interface is arranged with 7 buttons: "Porridge", "Hot Rice", "Keep Warm", "Porridge", "Steam", "Start", and "Cancel". Figure 2 As shown in the figure, "Porridge", "Hot Rice", "Keep Warm", "Porridge", and "Steam" are function buttons, and "Start" and "Undo" are decision buttons. The seven buttons flash at the set time interval and in the set order. The user is told that if he wants to trigger the function of a button, he needs to blink immediately after the button flashes.

[0052] Prompt mode prompts the user to execute and complete a function. After the algorithm module determines that a function is required by the user, the operation interface displays that the function is successfully selected and executed. After the function is completed, the operation interface displays that the function is successfully completed.

[0053] Furthermore, the algorithm module receives the electrooculogram signal data, and performs blink detection on the electrooculogram data during each round of button blinking, in combination with the order of button flashing on the operation interface. The blink detection first determines the number of blinks performed by the user after the button flashes. If there is zero blinking action or blinking action is performed after multiple buttons flash, no button is selected, and the operation interface continues with the next round of button flashing. If a blink action is performed only after one button flashes, it is further determined which button flashes after the blink action, so as to determine which button the user triggered. Specifically, in the blink detection, the electrooculogram signal data after the button flashes is classified to determine whether a blink action is performed after the button is pressed.

[0054] Regarding the specific content of the classification algorithm: we divide it into two parts: training and use. Training is divided into two parts: data preprocessing and model building. Use is divided into two parts: data preprocessing and classification using models.

[0055] In the data preprocessing part of the training, the operation interface module is in the training mode. The EO signal data collected in Case 1 and Case 2 are marked with two different labels: blinking and no blinking. The system takes the button flashing time point as the starting point and intercepts a fixed length of 600ms EO signal segment (sampling frequency; 250Hz). First, the signal segment is band-pass filtered (1-12Hz) as a whole to remove the interference of high-frequency electromyography and low-frequency head shaking, and then the difference is performed to obtain the differential signal segment. The waveform feature parameters that need to be extracted include: the maximum value a of the filtered signal segment max , the maximum value d of the differential signal segment max , the minimum value d of the differential signal segment min The time point of the maximum and minimum values ​​of the differential signal segment is t p and t v , the mean value a of the filtered signal segment ave , waveform span t d Given by the formula:

[0056] t d =|t p -t v |

[0057] a max d max d min ,t p ,t v 、a aveAs 6 feature data.

[0058] The construction of the trained model, χ represents the feature space, Y represents the label set, -1 represents no blinking, 1 represents blinking, R n represents the real number domain, N represents the total amount of data, and the i-th eigenvector The i-th label y i ∈Y={-1,+1}, there exists 1≤i≤N. Each sample point consists of features and labels, and the training data set can be expressed as:

[0059] T={(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x N ,y N )}

[0060] The classification model is constructed using the training data set T. Due to the small number of input features, in order to achieve the best classification effect, the voting classifier G(x) model is adopted, in which the basic classifiers used are: K-nearest neighbor classifier g 1 (x), Linear Discriminant Analysis classifier g 2 (x), support vector machine classifier g 3 (x). These three basic classifiers can be implemented using the MATLAB machine learning toolbox. m (x) represents the mth basic classifier, a m (m=1,2,3) is the linear combination of g m (x) corresponding to the coefficient, and Use features and labels to train the basic classifier g separately 1 (x),g 2 (x),g 3 (x). The voting classifier G(x) is essentially a boosting method, and the final classifier is a linear combination of the basic classifiers.

[0061]

[0062] Calculate g m The classification error rate e on the training data of (x) m :

[0063]

[0064] The coefficient

[0065] According to the above formula, the voting classifier G(x) is obtained.

[0066] In the data preprocessing part used, the operation interface module is in the use mode. After the button flashes, the same data processing method as the training is adopted.

[0067] The classification part of the model used is a max d max d min ,t p ,t v 、a ave These 6 feature data are put into the trained voting classifier, and the classification output result is used as the result of blink detection. The flowchart of blink detection is shown in Figure 3 shown.

[0068] When two consecutive selected buttons are function buttons and decision buttons respectively, a judgment is made. The function button is selected first, and then the cancel button is selected, and it is considered that the previously selected function button is canceled; the function button is selected first, and then the start button is selected, and it is considered that the function corresponding to the selected function button is determined to be executed. The algorithm module sends the instruction corresponding to the function to the communication module, and the operation interface module in the electrooculogram acquisition module displays that the function is successfully selected.

[0069] Further, the communication module is a Bluetooth interface, which can carry out two-way communication. Before the instruction is issued, it specifically receives the instruction information sent by the algorithm module, and controls the rice cooker to work according to the instruction information. After the function corresponding to the instruction is completed, the information of the function completion is sent to the operation interface module.

[0070] Further, the control module receives the instruction sent by the communication module, then controls the function corresponding to the electric rice cooker execution instruction, sets the timer (the time is the completion time corresponding to the set function) and performs temperature detection at the same time. The timing can be completed using a 555 timer chip, and the temperature detection device is completed using a digital temperature measurement chip DS18B20. When the timer timing ends and the temperature is greater than the completion temperature corresponding to the set function, it is judged that the function is completed, and then the information of the function completion is fed back to the communication module.

[0071] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A rice cooker control system based on electrooculography, characterized in that: include: The electrooculogram acquisition module is divided into two parts, one is the acquisition module and the other is the operation interface module; The electrooculogram signal data collected by the acquisition module is sent to the algorithm module; The content displayed on the operation interface of the operation interface module includes: buttons arranged according to the electric cooker function, the buttons flashing respectively in the set order, and prompts for confirming the execution of the electric cooker function and completing the prompts of the electric cooker function; The algorithm module uses the collected electrooculogram data to perform blink detection through a classification algorithm, determines whether to select a button displayed on the operation interface and which button is selected based on the result of the blink detection, further analyzes the function required by the user according to the set rules, obtains the corresponding instruction based on the function, and sends the instruction to the communication module; The communication module receives various instructions sent by the algorithm module and transmits them to the control module before the electric cooker performs a function. After the electric cooker completes a function, the communication module sends the information of the function completion to the operation interface module. The control module controls the rice cooker to complete the corresponding function according to the instruction received by the communication module, and transmits the information of the function completion to the communication module; The algorithm module receives the electrooculogram signal data, and performs blink detection on the electrooculogram signal data during each round of button blinking in combination with the order of button blinking on the operation interface; the blink detection first determines the number of blinking actions performed by the user after the button blinks; if there is no blinking action or blinking actions are performed after multiple buttons blink, no button is selected, and the operation interface continues with the next round of button blinking; if only one button blinks and a blink action is performed, it is further determined which button the blinking action is after, so as to determine which button the user has triggered; specifically, in the blink detection, the electrooculogram signal data after the button blinks is classified to determine whether a blink action is performed after the button is clicked; The specific content of the classification algorithm is divided into two parts: training and use. The training is divided into two parts: data preprocessing and model building. The use is divided into two parts: data preprocessing and classification using the model. In the data preprocessing part of the training, the operation interface module is in training mode. The electrooculogram signal data collected in Case 1 and Case 2 are labeled as blinking and no blinking. The system takes the button flashing time point as the starting point and intercepts a fixed length of electrooculogram signal segment; first, the signal segment is band-pass filtered as a whole to remove the interference of high-frequency electromyography and low-frequency head shaking, and then the difference is performed to obtain the differential signal segment. The waveform feature parameters that need to be extracted include: the maximum value a of the filtered signal segment max , the maximum value d of the differential signal segment max , the minimum value d of the differential signal segment min The time point of the maximum and minimum values ​​of the differential signal segment is t p and t v , the mean value a of the filtered signal segment ave , waveform span t d Given by the formula: t d =|t p -t v | a max d max d min ,t p ,t v 、a ave As 6 feature data; The construction of the trained model, χ represents the feature space, Y represents the label set, -1 represents no blinking, 1 represents blinking, R n represents the real number domain, N represents the total amount of data, and the i-th eigenvector The i-th label y i ∈Y={-1,+1}, there exists 1≤i≤N; each sample point consists of features and labels, and the training data set is expressed as: T={(x1,y1),(x2,y2),…,(x N ,y N )} The classification model is constructed using the training data set T. Due to the small number of input features, the voting classifier G(x) model is used to improve the classification effect. The basic classifiers used are: K-nearest neighbor classifier g1(x), linear discriminant analysis classifier g2(x), support vector machine classifier g3(x); g m (x) represents the mth basic classifier, a m is the linear combination of g m (x) corresponding to the coefficient, and Use features and labels to train basic classifiers g1(x), g2(x), and g3(x) respectively; the voting classifier G(x) is essentially a boosting method, and the final classifier is a linear combination of the basic classifiers: Calculate g m The classification error rate e on the training data of (x) m : Among them, the coefficient According to the above formula, we get the voting classifier G(x); For the data preprocessing part used, the operation interface module is in the use mode, and after the button flashes, the same data processing method as the training is adopted; The classification part of the model used is a max ,d max ,d min ,t p ,t v 、a ave These 6 feature data are put into the trained voting classifier, and the classification output is used as the result of blink detection; When two consecutive selected buttons are respectively a function button and a decision button, a judgment is made: the function button is selected first, and then the cancel button is selected, and it is considered that the previously selected function button is canceled; the function button is selected first, and then the start button is selected, and it is considered that the function corresponding to the selected function button is determined to be executed. The algorithm module sends the instruction corresponding to the function execution to the communication module, and displays the successful selection of this function in the operation interface module in the electrooculogram acquisition module.

2. The electric cooker control system based on electrooculography according to claim 1 is characterized in that: The acquisition module needs to place electrodes around the user's eyes, the electrodes are connected to a signal amplifier, and then the signal amplifier is connected to the algorithm module, and the collected electrooculogram signal data will be sent to the algorithm module.

3. The electric rice cooker control system based on electrooculography according to claim 1, characterized in that: In the operation interface module, there are three different display modes: training mode, use mode and prompt mode; In the training mode, the operation interface only displays one button, which flashes continuously. In the first case, the user is informed in advance that the button flashes and then blinks. In the second case, the user is informed in advance that the button flashes and then does not blink. In the use mode, the operation interface is arranged with 7 buttons: "Porridge", "Hot Rice", "Keep Warm", "Porridge", "Steam", "Start" and "Cancel". Among them, "Porridge", "Hot Rice", "Keep Warm", "Porridge" and "Steam" are function buttons, and "Start" and "Cancel" are decision buttons. The 7 buttons flash at the set time interval and in the set order. Users are told that if they want to trigger the function of a button, they need to blink immediately after the button flashes. Prompt mode prompts the user to execute and complete functions; after the algorithm module determines that a certain function is required by the user, the operation interface displays that the function is successfully selected and executed; after the function is completed, the operation interface displays that the function is successfully completed.

4. The electric rice cooker control system based on electrooculography according to claim 1, characterized in that: The communication module is a Bluetooth interface capable of two-way communication. Before the instruction is issued, it specifically receives the instruction information sent by the algorithm module and controls the rice cooker to work according to the instruction information. After the function corresponding to the instruction is completed, the function completion information is sent to the operation interface module.

5. The electric rice cooker control system based on electrooculography according to claim 1, characterized in that: The control module receives the instruction sent by the communication module, then controls the rice cooker to execute the function corresponding to the instruction, sets a timer and performs temperature detection at the same time, wherein the time of the timer is the completion time corresponding to the set function; when the timer timing ends and the temperature is greater than the completion temperature corresponding to the set function, it is judged that the function is completed, and then the information of the completion of the function is fed back to the communication module.

Citation Information

Patent Citations

  • Unmanned aerial vehicle control system based on portable electro-oculogram acquisition device

    CN112904893A

  • Intelligent wheelchair based on eye electric signals and head movement signals

    CN202096374U

  • EOG-based method and apparatus for asynchronous character input

    US20210278894A1

  • Program, information processing device, and eyewear

    WO2017064800A1