WO2020078093A1 - 门锁控制方法、装置、控制设备 - Google Patents

门锁控制方法、装置、控制设备 Download PDF

Info

Publication number
WO2020078093A1
WO2020078093A1 PCT/CN2019/101018 CN2019101018W WO2020078093A1 WO 2020078093 A1 WO2020078093 A1 WO 2020078093A1 CN 2019101018 W CN2019101018 W CN 2019101018W WO 2020078093 A1 WO2020078093 A1 WO 2020078093A1
Authority
WO
WIPO (PCT)
Prior art keywords
sample
time period
time
target
operation frequency
Prior art date
Application number
PCT/CN2019/101018
Other languages
English (en)
French (fr)
Inventor
董明珠
李绍斌
谭建明
李坤
宋德超
陈道远
彭磊
Original Assignee
珠海格力电器股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 珠海格力电器股份有限公司 filed Critical 珠海格力电器股份有限公司
Publication of WO2020078093A1 publication Critical patent/WO2020078093A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07CTIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C9/00Individual registration on entry or exit
    • G07C9/00174Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys
    • G07C9/00571Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys operated by interacting with a central unit

Definitions

  • the present disclosure relates to the field of smart home technology, and in particular to a door lock control method, device, and control equipment.
  • the smart door lock will enter the sleep mode when there is no operation. Once the user operation is detected, the smart door lock will be woken up.
  • the touch module of the door lock such as touching the numeric keyboard or fingerprint recognition module, can usually detect the user's Operation, when the user approaches or touches the touch module of the door lock, the door lock will be woken up.
  • the power supply gain of the smart door locks on the existing market is determined at the factory, that is, a fixed power supply gain has been set at the factory.
  • the sensing function of the touch module is related to the power supply gain, and the sensing function of the touch module directly affects The sensing distance of the touch module, therefore, under a fixed power supply gain, the user can only wake the door lock within a certain sensing distance to complete the door opening, the operation is not flexible, and when the user does not need to open the door, the door lock still operates at a fixed power supply gain , Causing unnecessary power consumption.
  • Some embodiments of the present disclosure provide a door lock control method.
  • the method includes:
  • the method further includes: according to the current time, and the pre-saved For each time period included in the time period, determine the target time period where the current time is located;
  • the inputting the current time into a pre-trained operation frequency determination model, and determining the target operation frequency corresponding to the current time based on the operation frequency determination model includes:
  • the target time period is input into a pre-trained operation frequency determination model, and based on the operation frequency determination model, a target operation frequency corresponding to the target time period is determined, wherein the target operation frequency corresponding to the target time period is The ratio of the predicted number of user operations in the target time period to the total number of operations in a preset time period, and the target time period is located in the time period.
  • the method further includes: judging whether the time of the last determination of the target time period from the current time reaches a time interval corresponding to the pre-saved time period; if it is, executing the step according to the current time And each time period included in the pre-saved time period, the step of determining the target time period in which the current time is located.
  • the method further includes: judging whether there is only one of the target time period of the current time and the last determined target time period according to the pre-saved identification information of the time period of frequent user operations
  • the target time period is a time period where the user frequently operates; if so, the step of inputting the target time period into a pre-trained operation frequency determination model is performed.
  • the training process of the operating frequency determination model includes:
  • the total number of sample operations within a preset time period is counted
  • the corresponding sample operation frequency in the first sample time is determined according to the ratio of the number of user sample operations in the first sample time to the total number of sample operations in the time period;
  • Each first sample time and the sample operation frequency corresponding to each first sample time are input into the operation frequency determination model, and the operation frequency determination model is trained.
  • the training process of the operating frequency determination model includes:
  • the total number of sample operations within a preset time period is counted
  • each pre-saved time period For each pre-saved time period, use this time period as the sample time period; obtain each second sample time within the sample time period in the training set, and obtain the sample operation corresponding to each second sample time The number of times; according to the number of sample operations corresponding to each second sample time, count the number of sample operations of the user in the sample time period; and according to the number of sample operations of the user in the sample time period account for the time period The ratio of the total number of sample operations within the period to determine the sample operation frequency corresponding to the sample time period;
  • Each sample time period and the sample operation frequency corresponding to each sample time period are input into the operation frequency determination model, and the operation frequency determination model is trained.
  • the method further includes:
  • the number of sample operations corresponding to the sample time matching the time when the user operates the door lock is increased by a set number of times.
  • the method further includes:
  • For each first sample time in the training set determine whether the first sample time belongs to a weekend or holiday; if it belongs to a weekend or holiday, delete the number of sample operations corresponding to the first sample time in the training set.
  • a door lock control device which includes:
  • a first determining module configured to input the current time into a pre-trained operation frequency determination model, and determine a target operation frequency corresponding to the current time based on the operation frequency determination model, wherein the current time corresponds to The target operating frequency is the ratio of the predicted number of user operations at the current time to the total number of operations within a preset time period, and the current time is within the time period;
  • a second determining module configured to determine a target power supply gain corresponding to the target operating frequency according to a pre-stored correspondence between the operating frequency and the power supply gain, wherein the higher the operating frequency in the correspondence, the greater the power supply gain;
  • the control module is configured to use the target power supply gain as the power supply gain of the door lock to control the door lock.
  • the device further includes:
  • a third determining module configured to determine the target time period in which the current time is based on the current time and each time period included in the pre-saved time period;
  • the first determination module is further configured to input the target time period into a pre-trained operation frequency determination model, determine the target operation frequency corresponding to the target time period based on the operation frequency determination model, wherein The target operation frequency corresponding to the target time period is a ratio of the predicted number of user operations in the target time period to the total number of operations in a preset time period, and the target time period is located in the time period.
  • the third determination module is further used to determine whether the time of the last determination of the target time period has reached the time interval corresponding to the pre-saved time period from the current time; if so, according to the current The time and each time period included in the pre-saved time period determine the target time period in which the current time is located.
  • the first determining module is further configured to determine the target time period in which the current time is located and the target time period previously determined according to the pre-saved identification information of the time period in which the user frequently operates , Whether there is only one target time period for the user's frequent operation time period; if so, input the target time period into the operation frequency determination model that has been pre-trained.
  • the device further includes:
  • the first training module is used to count the total number of sample operations within a preset time period according to the number of sample operations corresponding to each first sample time in the training set; for each first sample time, according to the user ’s The ratio of the number of sample operations at the same time to the total number of sample operations within the time period to determine the corresponding sample operation frequency within the first sample time; The sample operation frequency corresponding to this time is input into the operation frequency determination model, and the operation frequency determination model is trained.
  • the device further includes:
  • the second training module is used to count the total number of sample operations within a preset time period according to the number of sample operations corresponding to each first sample time in the training set; for each pre-saved time period, the time period is used as Sample time period; obtain each second sample time within the sample time period in the training set, and obtain the number of sample operations corresponding to each second sample time; according to the sample corresponding to each second sample time
  • the number of operations counts the number of sample operations performed by the user in the sample period; and determines the sample period based on the ratio of the number of sample operations performed by the user in the sample period to the total number of sample operations performed in the time period Corresponding sample operation frequency; input each sample time period and the sample operation frequency corresponding to each sample time period into the operation frequency determination model, and train the operation frequency determination model.
  • the device further includes:
  • An additional module is used to obtain the time when the user operates the door lock when the user operates the door lock; in the training set, increase the number of sample operations corresponding to the sample time that matches the time the user operated the door lock frequency.
  • the device further includes:
  • the deletion module is used to determine whether the first sample time belongs to the weekend or holiday for each first sample time in the training set; if it belongs to the weekend or holiday, the sample corresponding to the first sample time in the training set The number of operations is deleted.
  • control device including: a processor and a memory
  • a computer program is stored in the memory, and when the computer program is executed by the processor, the processor is caused to perform the steps of any of the above methods.
  • Some embodiments of the present disclosure provide a non-transitory computer-readable storage medium that stores a computer program executable by a control device, and when the computer program runs on the control device, causes the control device to execute the above Any one of the steps of the method.
  • Embodiment 1 is a schematic diagram of a door lock control process provided by Embodiment 1 of the present disclosure
  • FIG. 2 is a schematic diagram of a door lock control process provided by Embodiment 1 of the present disclosure
  • Embodiment 5 of the present disclosure is a diagram of a machine learning power supply gain model provided by Embodiment 5 of the present disclosure
  • FIG. 4 is a schematic diagram of a door lock control process provided by Embodiment 7 of the present disclosure.
  • FIG. 5 is a schematic structural diagram of a control device according to Embodiment 9 of the present disclosure.
  • FIG. 6 is a schematic diagram of a door lock control device provided by an embodiment of the present disclosure.
  • the embodiments of the present disclosure provide a door lock control method, device, control device, and readable storage medium, thereby solving the inflexibility of user operations caused by the fixed power supply gain And the problem of large power consumption.
  • the present disclosure determines the model based on the pre-trained operating frequency according to the current time, determines the target operating frequency corresponding to the current time, and determines the target power supply gain corresponding to the target operating frequency.
  • the power supply gain of the door lock is related to the user operating frequency corresponding to the current time , And the higher the user operation frequency corresponding to the current time, the higher the power supply gain. Therefore, the power supply gain of the door lock can be adjusted in a differentiated and targeted manner, which can improve the flexibility of user operations, improve the user experience, and can be Lower power gain is used when low, reducing the power consumption of the door lock.
  • FIG. 1 is a schematic diagram of a door lock control process provided by an embodiment of the present disclosure. The process includes the following steps:
  • the door lock control method provided by the embodiments of the present disclosure is applied to a control device, which may be a terminal or an electronic device such as a door lock, as long as it has high computing power and network communication capabilities. If the control device is a terminal, the control device may be a user terminal, an intelligent gateway, or a server, etc. If the control device is a door lock, the control device may be a door lock installed in a home environment.
  • the control device can obtain the current time.
  • the current time may be a time including hours, minutes, and seconds, or may include a date in addition to including hours, minutes, and seconds.
  • control device is a terminal
  • the process of acquiring the current time by the control device refers to the prior art, and will not be repeated in the embodiments of the present disclosure.
  • control terminal is a door lock
  • the control device may obtain the current time by itself, or may obtain the current time from other devices in the home environment connected thereto.
  • S102 Input the current time into a pre-trained operation frequency determination model, and determine a target operation frequency corresponding to the current time based on the operation frequency determination model, where the target operation frequency corresponding to the current time is a prediction The ratio of the number of operations of the user at the current time to the total number of operations within a preset time period, and the current time is within the time period.
  • the operation frequency determination model is obtained by training based on the sample operation time and the sample operation frequency corresponding to the sample time.
  • the control device may store the pre-trained operation frequency determination model corresponding to the door lock, so the control device may determine the model based on the obtained current time and based on the operation frequency to determine the target operation frequency corresponding to the current time.
  • the target operating frequency corresponding to the current time is the target operating frequency predicted by the operating frequency determination model according to the current time. Specifically, the target operating frequency corresponding to the current time is the total number of operations performed by the user at the current time within a preset time period And the current time is within the time period.
  • the preset time period may be, for example, one day, one week, one month, and so on. For example, since the user's unlocking behavior within a day is regular, the time period can be set to one day.
  • Machine learning is a multi-disciplinary discipline involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. Multiple disciplines. Specially study how computers simulate or realize human learning behaviors to acquire new knowledge and skills, and reorganize existing knowledge structures to continuously improve their performance. It is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are in various fields of artificial intelligence. It mainly uses induction, synthesis rather than deduction. Therefore, this solution is based on this and can be determined by the operating frequency of pre-training. Model to predict the target operating frequency.
  • the operation frequency determination model is, for example, a certain machine learning model with a prediction function.
  • S103 Determine a target power supply gain corresponding to the target operating frequency according to a pre-stored correspondence between the operating frequency and the power supply gain, where the higher the operating frequency, the greater the power supply gain.
  • the control device pre-stores the correspondence between the operating frequency and the power supply gain. In this correspondence, the higher the operating frequency, the greater the power supply gain.
  • the corresponding relationship between the operating frequency and the power supply gain may be set by the door lock developer.
  • control device determines the model based on the operating frequency, determines the target operating frequency corresponding to the current time, and then determines the target power supply gain corresponding to the target operating frequency.
  • the target power supply gain corresponding to the time when the user's operation frequency is high is also higher, and the target power supply gain corresponding to the time when the user's operation frequency is low is also lower.
  • the ease of use of the touch module of the smart lock is related to the power supply gain of the touch module The size is closely related, when the power supply gain is large, the touch module can be awakened before the finger is touched, because the perception function becomes stronger. When the power supply gain drops, the sensing function of the touch module becomes weak, so it is only sensed when the finger touches the module.
  • the increased gain seems to enhance the function, but it consumes a lot of power.
  • S104 Use the target power supply gain as the power supply gain of the door lock to control the door lock.
  • the control device After determining the target power supply gain, the control device updates the power supply gain of the door lock to the target power supply gain to control the door lock.
  • the fixed mode of changing the power supply gain of the smart lock can be realized within a preset time period, and the power supply gain is appropriately amplified or reduced by machine learning, so that the power consumption can be optimally reduced, and the machine learning can be used to find the user to use in a day.
  • the frequency and time of the smart lock can be differentiated and targeted to increase or decrease the power supply gain, which improves the user experience and reduces system power consumption.
  • the model can be determined based on the pre-trained operating frequency according to the current time, the target operating frequency corresponding to the current time can be determined, and the target power supply gain corresponding to the target operating frequency can be determined.
  • the power supply gain of the door lock corresponds to the current time
  • the user operation frequency is related, and the higher the user operation frequency corresponding to the current time, the higher the power supply gain. Therefore, the power supply gain of the door lock can be adjusted in a differentiated and targeted manner, which can improve the flexibility of user operation, improve the user experience, and can When the user needs to open the door is low, the power supply gain is lower, which reduces the power consumption of the door lock.
  • the method further includes:
  • the inputting the current time into a pre-trained operation frequency determination model, and determining the target operation frequency corresponding to the current time based on the operation frequency determination model includes:
  • the target time period is input into a pre-trained operation frequency determination model, and based on the operation frequency determination model, a target operation frequency corresponding to the target time period is determined, wherein the target operation frequency corresponding to the target time period is The ratio of the predicted number of user operations in the target time period to the total number of operations in a preset time period, and the target time period is located in the time period.
  • Each time period included in the time period is pre-stored in the control device, and each time period included in the time period can be set by the user, can be set by the developer of the door lock, and can be learned and so on .
  • the control device may determine in which time period of the time period the current time is, that is, to determine the target time period in which the current time is located.
  • the target time period in which the current time is located may be input into the pre-trained operation frequency determination model, and the target operation frequency corresponding to the target time period may be determined based on the operation frequency determination model.
  • the operation frequency determination model is obtained by training based on the sample operation time and the sample operation frequency corresponding to the sample time.
  • the target operating frequency corresponding to the target time determined at this time is the target operating frequency predicted by the operating frequency determination model according to the target time period.
  • the target operating frequency corresponding to the target time period is the preset number of user operations in the target time period.
  • the ratio of the total number of operations within the time period, and the target time period is within the time period.
  • the control device determines the target power supply gain corresponding to the target operating frequency of the target time period according to the pre-saved correspondence between the operating frequency and the power supply gain, and then uses the target power supply gain to lock Take control.
  • the door lock is controlled according to the time period.
  • the method further includes: determining whether the time of the last determination of the target time period from the current time reaches a time interval corresponding to the pre-saved time period;
  • the control device pre-stores the time interval corresponding to the time period
  • the time interval corresponding to the time period can be used as the interval at which the time period corresponding to the time changes.
  • control device may store the time of the last determination of the target time period, so as to determine whether the current time obtained from the last determination of the target time period has reached the current time The time interval corresponding to the time period.
  • the target time period last time reaches the current time interval corresponding to that time period, it can be considered that the time period corresponding to the time has changed, so it can be based on the current time obtained and the time period For each included time period, determine the target time period for the current time.
  • the control device may save the time of determining the target time period of the current time, so that the next time the terminal obtains the time to make a judgment.
  • the time of the last determined time period of the target time period does not reach the time interval corresponding to the time period, it can be considered that the time period corresponding to the time has not changed, that is, the target corresponding to the current time and the current time The time period is the same, so there is no need to re-determine the target time period.
  • the control device may specifically determine the time difference between the time of the last determined target time period and the current time, and determine whether the value of the time interval corresponding to the time difference is the same as the time interval. If they are the same, it may be considered that the time interval corresponding to the time period has been reached , Conversely, to determine the time interval corresponding to the unreached time period, you can also control the device to reset the timer and restart timing after each determination of the target time period, so as to determine whether the time period has been reached according to the timing result of the timer Corresponding time interval.
  • the corresponding target time period is determined only when the time period corresponding to the time changes, that is, the target time period is determined only once in each time period, which further reduces the power consumption of the door lock.
  • the method before inputting the target time period into the pre-trained operation frequency determination model, the method further includes:
  • the pre-stored identification information of the time period in which the user frequently operates determine whether there is only one target time period in the target time period in which the current time is located and the target time period determined last time is the time period in which the user frequently operates;
  • multiple adjacent time periods may correspond to the time periods when the user frequently unlocks. Or multiple adjacent time periods correspond to the time period when the user unlocks infrequently, so when the state of the user's frequent unlocking or infrequent unlocking period changes, the power supply gain of the door lock can be re-determined. , No need to re-determine, further reducing the power consumption of the door lock.
  • the control device may be pre-stored with the identification information of the time period in which the user frequently operates, such as adding marks for the time period in which the user frequently operates, or for the time period in which the user frequently operates and infrequent operations
  • the corresponding time period is added with a corresponding mark, or the time period during which the user frequently operates is grouped into a set, and the identification information of the set is determined as the identification information for the time period during which the user frequently operates.
  • the control device can determine whether the target time period determined last time is the time period frequently operated by the user, and whether the target time period of the current time determined this time is the time period frequently operated by the user, and then the control device judges the last determination Of the target time period and the target time period of the current time determined this time, is there only one target time period for the user's frequent operation time period, if it is, it is considered that the last determined target time period to this current time Whether the status of the user's frequent operations in the target time period has changed. It may be that the last user's frequent operations have changed to the current user's infrequent operations, or the last user's infrequent operations have changed to the current user. Frequent operation; if not, it is considered that the status of whether the user frequently operates from the target time period determined last time to the target time period of the current time has not changed.
  • each door lock can effectively grasp the user habits, greatly improving the user's experience, forming an individual differentiation of each door lock, allowing each smart lock to adapt to the habits of the corresponding user.
  • the power supply gain of the door lock can be updated in time, and when there is no change, Keeping the power supply gain of the door lock unchanged can further reduce the power consumption of the door lock.
  • the training process of the operating frequency determination model includes:
  • the total number of sample operations within a preset time period is counted
  • the corresponding sample operation frequency in the first sample time is determined according to the ratio of the number of user sample operations in the first sample time to the total number of sample operations in the time period;
  • Each first sample time and the sample operation frequency corresponding to each first sample time are input into the operation frequency determination model, and the operation frequency determination model is trained.
  • the operation frequency determination model in the embodiment of the present disclosure may be a model based on machine learning training.
  • the training set contains a large number of first sample times.
  • the first sample time contained in the training set is a sample used for model training.
  • the training set also includes the number of sample operations corresponding to the first sample time.
  • the first sample time included in the training set and the number of sample operations corresponding to the first sample time are determined according to the time and number of operations the user operates the door lock, that is, the number of sample operations corresponding to each first sample time is The number of sample operations at the first sample time.
  • FIG. 3 it is a graph of the machine learning power supply gain model.
  • the model diagram is mainly for the number of sample operations corresponding to each first sample time (horizontal axis in FIG. 3) and the first sample time in the training set ( The vertical axis in Figure 3) is summarized.
  • the first sample time is the sample time within 24 hours of a day, and the corresponding sample operation times are the operation times from 0 to 200.
  • the total number of sample operations within a preset time period needs to be determined first.
  • the total number of sample operations in the time period it is determined according to the number of sample operations corresponding to each first sample time in the training set, which may specifically be determined by summing the number of sample operations corresponding to each first sample time in the training set. The total number of samples in the time period.
  • the sample operation frequency corresponding to each first sample time it is determined according to the ratio of the number of sample operations corresponding to each first sample time to the total number of sample operations within the time period.
  • the ratio of the number of sample operations corresponding to the sample time to the total number of sample operations within the time period is determined as the frequency of sample operations corresponding to each first sample time, which may be the sample operations corresponding to each first sample time.
  • the ratio of the number of times to the total number of sample operations in the time period and the set weight value jointly determine the sample operation frequency corresponding to each first sample time. For example, the product of the ratio and the set weight value is determined as the sample operation frequency.
  • the operating frequency determining model After determining each first sample time and the sample operating frequency corresponding to each first sample time are input into the operating frequency determining model, the operating frequency determining model is trained.
  • the process of training the model according to the data used for training can be implemented by using the existing technology, and will not be repeated in the embodiments of the present disclosure.
  • the training process of the operating frequency determination model includes:
  • the total number of sample operations within a preset time period is counted
  • each pre-saved time period For each pre-saved time period, use this time period as the sample time period; obtain each second sample time within the sample time period in the training set, and obtain the sample operation corresponding to each second sample time The number of times; according to the number of sample operations corresponding to each second sample time, count the number of sample operations of the user in the sample time period; and according to the number of sample operations of the user in the sample time period account for the time period The ratio of the total number of sample operations within the period to determine the sample operation frequency corresponding to the sample time period;
  • Each sample time period and the sample operation frequency corresponding to each sample time period are input into the operation frequency determination model, and the operation frequency determination model is trained.
  • the operating frequency determination model in the embodiment of the present disclosure may be a model based on machine learning training.
  • training is performed according to the operating frequency corresponding to the time period.
  • Each time period is pre-stored in the terminal.
  • the total number of sample operations within a preset time period needs to be determined first.
  • the total number of sample operations in the time period it is determined according to the number of sample operations corresponding to each first sample time in the training set, which may specifically be determined by summing the number of sample operations corresponding to each first sample time in the training set. The total number of samples in the time period.
  • each time period After determining the number of sample operations corresponding to each time period, use each time period as the sample time period, obtain each second sample time within the sample time period in the training set, and obtain the corresponding time of each second sample time
  • the number of sample operations corresponding to the sample time period is counted according to the number of sample operations corresponding to each second sample time in the sample time period. Specifically, the sum of the number of sample operations corresponding to each sample time in the sample time period may be determined as the number of sample operations corresponding to the sample time period.
  • the frequency of sample operations corresponding to each sample time can be determined.
  • the sample operation frequency corresponding to each sample time period it is determined according to the ratio of the number of sample operations corresponding to each sample time period to the total number of sample operations within the time period, which may be directly corresponding to each sample time period
  • the ratio of the number of sample operations to the total number of sample operations in the time period is determined as the frequency of sample operations corresponding to each sample time period, which may be the number of sample operations corresponding to each sample time period in the total number of sample operations in the time period
  • the ratio of, and the set weight value together determine the sample operating frequency corresponding to each sample time period. For example, the product of the ratio and the set weight value is determined as the sample operating frequency.
  • the operation frequency determination model After determining each sample time period and the sample operation frequency corresponding to each sample time period are input into the operation frequency determination model, the operation frequency determination model is trained.
  • the process of training the model according to the data used for training may be implemented using existing technology, and will not be described in detail in the embodiments of the present disclosure.
  • the method further includes:
  • the number of sample operations corresponding to the sample time matching the time when the user operates the door lock is increased by a set number of times.
  • the present disclosure provides a way to collect data in the training set, so that the operation frequency determination model is trained based on the data in the training set.
  • the time when the user operates the door lock is acquired.
  • the control device recognizes whether the user has operated the door lock, and may send corresponding information to the control device when the door lock is operated by the user.
  • the time for obtaining the user's operation of the door lock is similar to the above-mentioned process for obtaining the current time, and will not be described in detail in the embodiments of the present disclosure.
  • the number of sample operations corresponding to the sample time that matches the time for the user to operate the door lock is increased by the set number of times.
  • the set number of times may be saved in the control device, for example, the set number of times may be 1, 3, or 5, etc., which is not limited in the embodiments of the present disclosure.
  • the operation frequency determination model can be further trained according to the updated training set, so as to update the power supply gain of the door lock.
  • the user's operation data at each time of the day is organized as input data into the chart, and each operation will correspond to the chart in the database
  • the value of is superimposed by 1, so that (the corresponding map area / total map area at each moment) is used as a strategy indication for power supply gain adjustment, based on the historical data of user operations in the training set, the experience summarized by machine learning, and input to
  • the data collected by the new user using the door lock in the icon shown in FIG. 3 processes the data in the image, that is, continues to train the operation frequency determination model, so as to update the power supply gain time period, and then realize the door lock Control of the power supply gain.
  • the method further includes:
  • the training set For each first sample time in the training set, it is determined whether the first sample time belongs to a weekend or a holiday; if either is true, the number of sample operations corresponding to the first sample time is deleted in the training set.
  • the influence caused by the messy and irregular data on weekends or holidays can be removed.
  • the first sample time in the training set For each first sample time in the training set, whether the first sample time belongs to a weekend or a holiday can be judged according to the calendar, if either is true, that is, if the first sample time belongs to a weekend or a holiday, because the weekend is considered Or, the data operated by the user in holidays is messy information, so the number of sample operations corresponding to the first sample time can be deleted. If the first sample time neither belongs to a weekend nor a holiday, the number of sample operations corresponding to the first sample time may be retained in the training set.
  • the embodiments of the present disclosure also provide a control device 500, as shown in FIG. 5, including: a processor 501 and a memory 503, and optionally, a communication interface 502 and a communication bus 504 etc.
  • the processor 501, the communication interface 502, and the memory 503 complete communication with each other through the communication bus 504;
  • a computer program is stored in the memory 503, and when the computer program is executed by the processor 501, the processor 501 executes the steps of the door lock control method of any embodiment, for example, performing the following steps:
  • the control device provided by the embodiment of the present disclosure may specifically be an electronic device such as a desktop computer, a server, and a network-side device.
  • the communication bus mentioned in the above control device may be a peripheral component interconnection (Peripheral Component Interconnect, PCI) bus or an extended industry standard structure (Extended Industry Standard Architecture, EISA) bus, etc.
  • PCI peripheral component interconnection
  • EISA Extended Industry Standard Architecture
  • the communication bus can be divided into an address bus, a data bus, and a control bus.
  • the figure is only represented by a thick line, but it does not mean that there is only one bus or one type of bus.
  • the communication interface 502 is used for communication between the control device and other devices.
  • the memory may include random access memory (Random Access Memory, RAM), or non-volatile memory (Non-Volatile Memory, NVM), for example, at least one disk memory.
  • RAM Random Access Memory
  • NVM Non-Volatile Memory
  • the memory may also be at least one storage device located away from the aforementioned processor.
  • the above processor may be a general-purpose processor, including a central processor, a network processor (Network Processor, NP), etc .; it may also be a digital instruction processor (Digital Signal Processing, DSP), an application specific integrated circuit, a field programmable gate display or Other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • NP Network Processor
  • DSP Digital Signal Processing
  • the embodiments of the present disclosure also provide a non-transitory computer storage readable storage medium, the computer readable storage medium stores a computer program executable by the control device, when the computer When the program runs on the control device, when the control device is executed, the steps of the door lock control method of any embodiment are implemented, for example, the following steps are implemented:
  • the above computer-readable storage medium may be any available medium or data storage device that can be accessed by the processor in the control device, including but not limited to magnetic storage such as floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc., optical storage such as CD , DVD, BD, HVD, etc., as well as semiconductor memory such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state hard disk (SSD), etc.
  • magnetic storage such as floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.
  • optical storage such as CD , DVD, BD, HVD, etc.
  • semiconductor memory such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state hard disk (SSD), etc.
  • FIG. 6 is a schematic diagram of a door lock control device 600 provided by an embodiment of the present disclosure, which is applied to a control device.
  • the device includes:
  • the obtaining module 601 is used to obtain the current time
  • the first determining module 602 is configured to input the current time into a pre-trained operation frequency determination model, determine the target operation frequency corresponding to the current time based on the operation frequency determination model, wherein the current time corresponds to The target operating frequency is the ratio of the predicted number of user operations at the current time to the total number of operations within a preset time period, where the current time is within the time period;
  • the second determining module 603 is configured to determine a target power supply gain corresponding to the target operating frequency according to a pre-stored correspondence between the operating frequency and the power supply gain, where the higher the operating frequency in the correspondence, the greater the power supply gain;
  • the control module 604 is configured to use the target power supply gain as the power supply gain of the door lock to control the door lock.
  • the device further includes: a third determination module 605, configured to determine a target time period in which the current time is located based on the current time and each time period included in a pre-saved time period; the first The determination module 602 is further configured to input the target time period into a pre-trained operation frequency determination model, determine the target operation frequency corresponding to the target time period based on the operation frequency determination model, wherein the target time The target operation frequency corresponding to the period is the ratio of the predicted number of user operations in the target time period to the total number of operations in the preset time period, and the target time period is located in the time period.
  • a third determination module 605 configured to determine a target time period in which the current time is located based on the current time and each time period included in a pre-saved time period
  • the first The determination module 602 is further configured to input the target time period into a pre-trained operation frequency determination model, determine the target operation frequency corresponding to the target time period based on the operation frequency determination model, wherein the target
  • the third determination module 605 is also used to determine whether the time of the last determination of the target time period has reached the time interval corresponding to the pre-saved time period from the current time; if so, according to the current time, and the pre-save For each time period included in the time period of, determine the target time period for the current time.
  • the first determination module 602 is further configured to determine whether there is only one target in the target time period of the current time and the target time period determined last time according to the pre-stored identification information of the time period in which the user frequently operates
  • the time period is a period of frequent operation by the user; if it is, the target time period is input into a pre-trained operation frequency determination model.
  • the apparatus further includes: a first training module 606, configured to count the total number of sample operations within a preset time period according to the number of sample operations corresponding to each first sample time in the training set; for each first sample Time, according to the ratio of the number of user sample operations in the first sample time to the total number of sample operations in the time period, determine the corresponding sample operation frequency in the first sample time; The time and the sample operation frequency corresponding to each first sample time are input into the operation frequency determination model, and the operation frequency determination model is trained.
  • a first training module 606 configured to count the total number of sample operations within a preset time period according to the number of sample operations corresponding to each first sample time in the training set; for each first sample Time, according to the ratio of the number of user sample operations in the first sample time to the total number of sample operations in the time period, determine the corresponding sample operation frequency in the first sample time; The time and the sample operation frequency corresponding to each first sample time are input into the operation frequency determination model, and the operation frequency determination model is trained.
  • the device further includes: a second training module 607, configured to count the total number of sample operations within a preset time period according to the number of sample operations corresponding to each first sample time in the training set; for each time saved in advance Segment, use this period as the sample period; obtain each second sample time within the sample period in the training set, and obtain the number of sample operations corresponding to each second sample time; according to each The number of sample operations corresponding to the second sample time, counting the number of sample operations performed by the user in the sample time period; and according to the number of sample operations performed by the user in the sample time period in the total number of sample operations in the time period
  • the ratio determines the sample operation frequency corresponding to the sample time period; input each sample time period and the sample operation frequency corresponding to each sample time period into the operation frequency determination model, and train the operation frequency determination model.
  • the apparatus further includes: an addition module 608, configured to obtain a time when the user operates the door lock when the user operates the door lock is identified; in the training set, a sample time matching the time when the user operates the door lock is corresponded to The number of sample operations increases by the set number of times.
  • an addition module 608 configured to obtain a time when the user operates the door lock when the user operates the door lock is identified; in the training set, a sample time matching the time when the user operates the door lock is corresponded to The number of sample operations increases by the set number of times.
  • the device further includes: a deletion module 609, for each first sample time in the training set, to determine whether the first sample time belongs to weekends or holidays; if any is yes, then the first in the training set The number of sample operations corresponding to a sample time is deleted.
  • the model can be determined based on the pre-trained operating frequency according to the current time, the target operating frequency corresponding to the current time can be determined, and the target power supply gain corresponding to the target operating frequency can be determined.
  • the power supply gain of the door lock corresponds to the current time
  • the user operation frequency is related, and the higher the user operation frequency corresponding to the current time, the higher the power supply gain. Therefore, the power supply gain of the door lock can be adjusted in a differentiated and targeted manner, which can improve the flexibility of user operation, improve the user experience, and can When the user needs to open the door is low, the power supply gain is lower, which reduces the power consumption of the door lock.
  • the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.
  • computer usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions may also be stored in a computer readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, the instructions The device implements the functions specified in one block or multiple blocks of the flowchart one flow or multiple flows and / or block diagrams.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operating steps are performed on the computer or other programmable device to produce computer-implemented processing, which is executed on the computer or other programmable device
  • the instructions provide steps for implementing the functions specified in one block or multiple blocks of the flowchart one flow or multiple flows and / or block diagrams.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Lock And Its Accessories (AREA)

Abstract

一种门锁控制方法、装置、控制设备,该方法包括:获取当前时间(S101);将当前时间输入到预先训练完成的操作频率确定模型中,基于操作频率确定模型,确定当前时间对应的目标操作频率,其中当前时间对应的目标操作频率为预测的用户在当前时间的操作次数占预设的时间周期内的操作总次数的比值,当前时间位于所述时间周期内(S102);根据预先保存的操作频率与供电增益的对应关系,确定目标操作频率对应的目标供电增益,在对应关系中操作频率越高供电增益越大(S103);将目标供电增益作为门锁的供电增益,对门锁进行控制(S104)。能够提高用户操作的灵活性,减少了门锁的电量消耗。

Description

门锁控制方法、装置、控制设备
相关申请的交叉引用
本申请是以CN申请号为201811196162.1,申请日为2018年10月15日的申请为基础,并主张其优先权,该CN申请的公开内容在此作为整体引入本申请中。
技术领域
本公开涉及智能家居技术领域,尤其涉及一种门锁控制方法、装置、控制设备。
背景技术
通常智能门锁在无操作时会进入睡眠模式,一旦检测到用户操作时,智能门锁就会被唤醒,门锁的触摸模块,如触摸数字键盘或指纹识别模块等,通常能够检测到用户的操作,当用户靠近或触摸门锁的触摸模块时,门锁就会被唤醒。
但是现有市场上智能门锁的供电增益在出厂的时候就确定了,即在出厂的时候已设置了固定的供电增益,触摸模块的感知功能与供电增益有关,而触摸模块的感知功能直接影响触摸模块的感应距离,因此在固定的供电增益下,用户只能在特定的感应距离内唤醒门锁完成开门,操作不灵活,并且当用户没有开门需求时,门锁仍以固定的供电增益运行,造成了不必要的电量消耗。
发明内容
本公开一些实施例提供了一种门锁控制方法,该方法包括:
获取当前时间;
将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内;
根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中在所述对应关系中操作频率越高供电增益越大;
将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
在一些实施例中,在所述获取当前时间之后,所述将所述当前时间输入到预先训练完 成的操作频率确定模型中之前,所述方法还包括:根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段;
所述将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率包括:
将所述目标时间段输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述目标时间段对应的目标操作频率,其中所述目标时间段对应的目标操作频率为预测的用户在所述目标时间段内的操作次数占预设的时间周期内的操作总次数的比值,所述目标时间段位于所述时间周期内。
在一些实施例中,所述方法还包括:判断上一次确定目标时间段的时间距离所述当前时间是否到达了预先保存的时间段对应的时间间隔;如果是,执行所述根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段的步骤。
在一些实施例中,所述方法还包括:根据预先保存的用户频繁操作的时间段的标识信息,判断所述当前时间所处的目标时间段及上一次确定的目标时间段中,是否只有一个目标时间段为用户频繁操作的时间段;如果是,执行所述将所述目标时间段输入到预先训练完成的操作频率确定模型中的步骤。
在一些实施例中,所述操作频率确定模型的训练过程包括:
根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;
针对每个第一样本时间,根据用户在该第一样本时间的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该第一样本时间内对应的样本操作频率;
将每个第一样本时间及每个第一样本时间对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
在一些实施例中,所述操作频率确定模型的训练过程包括:
根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;
针对预先保存的每个时间段,将该时间段作为样本时间段;在训练集中获取处于该样本时间段内的每个第二样本时间,并获取所述每个第二样本时间对应的样本操作次数;根据所述每个第二样本时间对应的样本操作次数,统计用户在该样本时间段内的样本操作次数;并根据所述用户在该样本时间段内的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该样本时间段对应的样本操作频率;
将每个样本时间段及每个样本时间段对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
在一些实施例中,所述方法还包括:
当识别到用户操作门锁时,获取用户操作门锁的时间;
在所述训练集中,将与所述用户操作门锁的时间匹配的样本时间对应的样本操作次数增加设定次数。
在一些实施例中,所述方法还包括:
针对训练集中的每个第一样本时间,判断该第一样本时间是否属于周末或节假日;如果属于周末或节假日,则在训练集中将该第一样本时间对应的样本操作次数删除。
本公开一些实施例提供了一种门锁控制装置,该装置包括:
获取模块,用于获取当前时间;
第一确定模块,用于将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内;
第二确定模块,用于根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中在所述对应关系中操作频率越高供电增益越大;
控制模块,用于将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
在一些实施例中,所述装置还包括:
第三确定模块,用于根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段;
所述第一确定模块,还用于将所述目标时间段输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述目标时间段对应的目标操作频率,其中所述目标时间段对应的目标操作频率为预测的用户在所述目标时间段内的操作次数占预设的时间周期内的操作总次数的比值,所述目标时间段位于所述时间周期内。
在一些实施例中,所述第三确定模块,还用于判断上一次确定目标时间段的时间距离所述当前时间是否到达了预先保存的时间段对应的时间间隔;如果是,根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段。
在一些实施例中,所述第一确定模块,还用于根据预先保存的用户频繁操作的时间段的标识信息,判断所述当前时间所处的目标时间段及上一次确定的目标时间段中,是否只有一个目标时间段为用户频繁操作的时间段;如果是,将所述目标时间段输入到预先训练 完成的操作频率确定模型中。
在一些实施例中,所述装置还包括:
第一训练模块,用于根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;针对每个第一样本时间,根据用户在该第一样本时间的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该第一样本时间内对应的样本操作频率;将每个第一样本时间及每个第一样本时间对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
在一些实施例中,所述装置还包括:
第二训练模块,用于根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;针对预先保存的每个时间段,将该时间段作为样本时间段;在训练集中获取处于该样本时间段内的每个第二样本时间,并获取所述每个第二样本时间对应的样本操作次数;根据所述每个第二样本时间对应的样本操作次数,统计用户在该样本时间段内的样本操作次数;并根据所述用户在该样本时间段内的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该样本时间段对应的样本操作频率;将每个样本时间段及每个样本时间段对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
在一些实施例中,所述装置还包括:
增加模块,用于当识别到用户操作门锁时,获取用户操作门锁的时间;在所述训练集中,将与所述用户操作门锁的时间匹配的样本时间对应的样本操作次数增加设定次数。
在一些实施例中,所述装置还包括:
删除模块,用于针对训练集中的每个第一样本时间,判断该第一样本时间是否属于周末或节假日;如果属于周末或节假日,则在训练集中将该第一样本时间对应的样本操作次数删除。
本公开一些实施例提供了一种控制设备,包括:处理器和存储器;
所述存储器中存储有计算机程序,当所述计算机程序被所述处理器执行时,使得所述处理器执行上述任一项所述方法的步骤。
本公开一些实施例提供了一种非瞬时性计算机可读存储介质,其存储有可由控制设备执行的计算机程序,当所述计算机程序在所述控制设备上运行时,使得所述控制设备执行上述任一项所述方法的步骤。
附图说明
为了更清楚地说明本公开实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本公开的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为本公开实施例1提供的一种门锁控制过程的示意图;
图2为本公开实施例1提供的一种门锁控制过程示意图;
图3为本公开实施例5提供的一种机器学习供电增益模型图;
图4为本公开实施例7提供的一种门锁控制过程示意图;
图5为本公开实施例9提供的一种控制设备的结构示意图;
图6为本公开实施例提供的一种门锁控制装置示意图。
具体实施方式
为了提高用户操作的灵活性,减少门锁的电量消耗,本公开实施例提供了一种门锁控制方法、装置、控制设备及可读存储介质,从而解决供电增益固定所导致的用户操作不灵活及电量消耗大的问题。
本公开根据当前时间,基于预先训练完成的操作频率确定模型,确定当前时间对应的目标操作频率,并确定目标操作频率对应的目标供电增益,门锁的供电增益与当前时间对应的用户操作频率有关,并且当前时间对应的用户操作频率越高供电增益越大,因此可以差异化针对性地调整门锁的供电增益,能够提高用户操作的灵活性,提高了用户体验,并且能够在用户开门需求较低时采用较低供电增益,减少了门锁的电量消耗。
为了使本公开的目的、技术方案和优点更加清楚,下面将结合附图本公开作在一些实施例中详细描述,显然,所描述的实施例仅仅是本公开的一部分实施例,而不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。
实施例1:
图1为本公开实施例提供的一种门锁控制过程的示意图,该过程包括以下步骤:
S101:获取当前时间。
本公开实施例提供的门锁控制方法应用于控制设备,该控制设备可以为终端,也可以为门锁等电子设备,只要能具有较高的计算能力和网络通信能力即可。如果该控制设备为 终端,则该控制设备可以为用户终端、智能网关或服务器等,如果该控制设备为门锁,则该控制设备可以为家居环境中安装的门锁。
该控制设备能够获取到当前时间,该当前时间可以是为包括时分秒的时间,还可以是除包括时分秒外,包括日期的时间。
如果该控制设备为终端,则该控制设备获取当前时间的过程参考现有技术,在本公开实施例中不做赘述。
如果该控制终端为门锁,则该控制设备自身可以获取当前时间,也可以是在与其连接的家居环境中的其他设备中获取到当前时间。
S102:将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内。
该操作频率确定模型为根据样本时间与样本时间对应的样本操作频率训练得到的。
控制设备中可以保存有门锁对应的预先训练完成的操作频率确定模型,因此控制设备可以根据获取到的当前时间,并基于该操作频率确定模型,确定该当前时间对应的目标操作频率,此时该当前时间对应的目标操作频率为操作频率确定模型根据当前时间预测的目标操作频率,具体该当前时间对应的目标操作频率为用户在当前时间的操作次数占预设的时间周期内的操作总次数的比值,且该当前时间位于时间周期内。
该预设的时间周期例如可以为一天、一周、一个月等。例如,由于用户在一天内的开锁行为具有规律性,因此可以将该时间周期设定为一天。
时代的发展已经步入了人工智能阶段,其中机器学习也扮演着重要的角色,机器学习是一门多领域交叉的学科,涉及概率论、统计学、逼近论、凸分析、算法复杂度理论等多门学科。专门研究计算机怎样模拟或实现人类的学习行为,以获取新的知识和技能,重新组织已有的知识结构使之不断改善自身的性能。它是人工智能的核心,是使计算机具有智能的根本途径,其应用遍及人工智能的各个领域,它主要使用归纳、综合而不是演绎,因此本方案基于此,可以通过预先训练完成的操作频率确定模型,预测目标操作频率。该操作频率确定模型例如为有预测功能的某种机器学习模型。
S103:根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中操作频率越高供电增益越大。
控制设备中预先保存有操作频率与供电增益的对应关系,该对应关系中操作频率越高,供电增益越大。该操作频率与供电增益的对应关系可以是门锁开发人员设定的。
因此控制设备基于操作频率确定模型,确定当前时间对应的目标操作频率后,根据确定该目标操作频率对应的目标供电增益。
这样对于用户操作频率较高的时间对应的目标供电增益也越高,对于用户操作频率较低的时间对应的目标供电增益也越低。如图2所示,基于机器学习确定的操作频率确定模型,在拥有指纹识别模块和触摸数字键盘的智能锁的控制过程中,智能锁的触摸模块的易用情况都与该触摸模块的供电增益大小息息相关,当供电增益较大时,触摸模块可以在手指未触碰之前就被唤醒,因为感知功能变强了。当供电增益下降时,触摸模块的感知功能变弱,所以当手指触摸到模块时才会被感知。然而增益升高看起来使功能增强了,但是却极大的消耗的电量,可以差异化针对性地调整门锁的供电增益,能够提高用户操作的灵活性,提高了用户体验,并且能够在用户开门需求较低时采用较低供电增益,减少了门锁的电量消耗,能够在让用户使用效果好和电量消耗小的矛盾体中取得平衡点极为重要。
S104:将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
控制设备确定目标供电增益后,将门锁的供电增益更新为该目标供电增益,对门锁进行控制。
此时可以实现在预设的时间周期内改变智能锁供电增益的固定模式,采用机器学习的方式恰当放大或减小供电增益,这样可以最优化的降低电量损耗,利用机器学习找到用户一天中使用智能锁的频繁程度和时间,就可以差异化和针对性的提高或降低供电增益,提高了用户体验的同时还会降低系统功耗。
本公开实施例中能够根据当前时间,基于预先训练完成的操作频率确定模型,确定当前时间对应的目标操作频率,并确定目标操作频率对应的目标供电增益,门锁的供电增益与当前时间对应的用户操作频率有关,并且当前时间对应的用户操作频率越高供电增益越大,因此可以差异化针对性地调整门锁的供电增益,能够提高用户操作的灵活性,提高了用户体验,并且能够在用户开门需求较低时采用较低供电增益,减少了门锁的电量消耗。
实施例2:
在上述实施例的基础上,本公开实施例中,所述获取当前时间后,所述将所述当前时间输入到预先训练完成的操作频率确定模型中之前,所述方法还包括:
根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段;
所述将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率包括:
将所述目标时间段输入到预先训练完成的操作频率确定模型中,基于所述操作频率确 定模型,确定所述目标时间段对应的目标操作频率,其中所述目标时间段对应的目标操作频率为预测的用户在所述目标时间段内的操作次数占预设的时间周期内的操作总次数的比值,所述目标时间段位于所述时间周期内。
由于用户在预设时间周期内打开门锁的行为具有规律性,并且集中在特定的时间段内,因此无需针对每次获取到的时间进行相应控制,可以是针对时间段为单位对门锁进行控制。
控制设备中预先保存有时间周期中包含的每个时间段,该时间周期中包含的每个时间段可以是用户设置的,可以是门锁的开发人员设置的,可以是及其学习到的等。
因此控制设备在获取到当前时间后,可以确定该当前时间处于时间周期中的哪个时间段中,即可以确定该当前时间所处的目标时间段。
此时可以是将该当前时间所处的目标时间段输入到预先训练完成的操作频率确定模型中,基于该操作频率确定模型确定该目标时间段对应的目标操作频率。该操作频率确定模型为根据样本时间与样本时间对应的样本操作频率训练得到的。
此时确定的目标时间对应的目标操作频率为操作频率确定模型根据目标时间段预测的目标操作频率,具体该目标时间段对应的目标操作频率为用户在目标时间段内的操作次数占预设的时间周期内的操作总次数的比值,且该目标时间段位于时间周期内。
因此控制设备确定目标时间段对应的目标操作频率后,根据预先保存的操作频率与供电增益的对应关系,确定该目标时间段的目标操作频率对应的目标供电增益,进而通过该目标供电增益对门锁进行控制。
由于本公开实施例中以时间段为单位,基于操作频率确定模型,确定时间段对应的目标操作频率,实现了根据时间段对门锁进行控制。
实施例3:
在上述各实施例的基础上,本公开实施例中,所述根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段之前,所述方法还包括:判断上一次确定目标时间段的时间距离所述当前时间是否到达了预先保存的时间段对应的时间间隔;
如果是,执行根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段的步骤。
由于多个当前时间可能对应同一时间段,因此无需针对获取到的每个当前时间确定其对应的目标时间段,即只对时间对应的时间段发生变化时确定对应的目标时间段即可,进一步降低了门锁的电量消耗。
由于控制设备中预先保存有时间段对应的时间间隔,因此可以将该时间段对应的时间间隔作为时间对应的时间段发生变化的间隔。
控制设备为了确定时间对应的时间段是否发生了变化,控制设备中可以保存有上一次确定目标时间段时的时间,从而判断上一次确定目标时间段的时间距离获取到的当前时间是否达到了该时间段对应的时间间隔。
如果上一次确定目标时间段的时间距离获取到的当前时间达到了该时间段对应的时间间隔,则可以认为时间对应的时间段发生了变化,因此可以根据获取到的当前时间,及时间周期中包含的每个时间段,确定该当前时间所处的目标时间段。
控制设备在确定当前时间所处的目标时间段后,可以将确定该当前时间所处的目标时间段的时间进行保存,以便于下一次终端获取到时间后进行判断。
如果上一次确定目标时间段的时间距离获取到的当前时间未达到该时间段对应的时间间隔,则可以认为时间对应的时间段未发生变化,即上一次获取到的时间与当前时间对应的目标时间段相同,因此无需重新确定目标时间段。
控制设备具体可以是确定上一确定目标时间段的时间与当前时间的时间差,判断该时间差与该时间段对应的时间间隔的值是否相同,如果相同,可以认为是达到了时间段对应的时间间隔,反之,确定未达到时间段对应的时间间隔,还可以控制设备在每次确定目标时间段后,对计时器清零并重新开始计时,从而根据计时器的计时结果判断是否到达了该时间段对应的时间间隔。
由于本公开实施例中,只对时间对应的时间段发生变化时确定对应的目标时间段,即在每个时间段内仅确定一次目标时间段,进一步降低了门锁的电量消耗。
实施例4:
在上述各实施例的基础上,本公开实施例中,所述将所述目标时间段输入到预先训练完成的操作频率确定模型中之前,所述方法还包括:
根据预先保存的用户频繁操作的时间段的标识信息,判断所述当前时间所处的目标时间段及上一次确定的目标时间段中,是否只有一个目标时间段为用户频繁操作的时间段;
如果是,执行将所述目标时间段输入到预先训练完成的操作频率确定模型中的步骤。
由于用户开锁行为具有规律性,如对于上班族,只有在上下班的时候才会少量使用门锁,其余时间通常无开锁行为,因此可能多个相邻时间段均对应用户频繁开锁的时间段,或多个相邻时间段均对应用户非频繁开锁的时间段,因此可以在用户频繁开锁或非频繁开锁的时间段的状态改变时,再去重新确定门锁的供电增益,在状态未改变时,无需重新确定,进一步降低了门锁的电量消耗。
为了进一步降低门锁的电量消耗,控制设备中可以与预先保存有用户频繁操作的时间段的标识信息,如针对用户频繁操作的时间段添加标记,或针对用户频繁操作的时间段及非频繁操作的时间段均添加相应的标记,或将用户频繁操作的时间段归纳为一个集合,将该集合的标识信息确定为用户频繁操作的时间段的标识信息等。
因此控制设备可以确定上一次确定的目标时间段是否为用户频繁操作的时间段,以及本次确定的当前时间所处的目标时间段是否为用户频繁操作的时间段,之后控制设备判断上一次确定的目标时间段及本次确定的当前时间所处的目标时间段中,是否只有一个目标时间段为用户频繁操作的时间段,如果是,则认为上一次确定的目标时间段到本次当前时间所处的目标时间段的用户是否频繁操作的状态发生了改变,可能是由上一次的用户频繁操作变为了本次用户非频繁操作,可能是由上一次的用户非频繁操作变为了本次用户频繁操作;如果否,则认为上一次确定的目标时间段到本次当前时间所处的目标时间段的用户是否频繁操作的状态未发生变化。
当用户是否频繁操作的状态发生了变化,为了能够及时对门锁的供电增益进行更新,进行后续将目标时间段输入到操作频率确定模型中的操作,从而确定该目标时间段对应的目标操作频率,并确定该目标操作频率对应的目标供电增益。
如用户是上班族,那么只有上下班的时候才会少量的使用门锁,那么机器学习捕捉到用户的习惯后,就会在上下班的时间内提高供电增益,让用户有良好的体验,其他时间降低供电增益,以减少门锁功耗。这样每一款门锁都能有效的掌握用户习惯,极大地提高了用户的体验效果,形成了每一把门锁的个体差异化,让每一把智能锁适配相应用户的习惯。
由于本公开实施例中根据上一次确定的目标时间段及本次确定的目标时间段,在用户是否频繁操作的状态发生改变时,能够及时对门锁的供电增益进行更新,在未发生改变时,保持门锁的供电增益不变,能够进一步减少门锁的用电消耗。
实施例5:
在上述各实施例的基础上,本公开实施例中,所述操作频率确定模型的训练过程包括:
根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;
针对每个第一样本时间,根据用户在该第一样本时间的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该第一样本时间内对应的样本操作频率;
将每个第一样本时间及每个第一样本时间对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
本公开实施例中操作频率确定模型可以是基于机器学习训练完成的模型。
具体地,训练集中包含大量的第一样本时间,训练集中包含的第一样本时间为用于进行模型训练的样本,该训练集中还包括第一样本时间对应的样本操作次数。训练集中包含的第一样本时间及第一样本时间对应的样本操作次数为根据用户操作门锁的时间及操作次数确定的,即每个第一样本时间对应的样本操作次数为用户在第一样本时间的样本操作次数。
如图3所示,为机器学习供电增益模型图,该模型图中主要是对训练集中每个第一样本时间(图3中的横轴)和第一样本时间对应的样本操作次数(图3中的纵轴)进行了归纳,第一样本时间为一天24小时内的样本时间,对应的样本操作次数为在0到200次的操作次数。
为了确定每个第一样本时间对应的样本操作频率,需要先确定预设的时间周期内的样本操作总次数。
在确定时间周期内的样本操作总次数时,根据训练集中每个第一样本时间对应的样本操作次数确定,具体可以是将训练集中每个第一样本时间对应的样本操作次数的和确定为时间周期内的样本总次数。
在确定每个第一样本时间对应的样本操作频率时,根据每个第一样本时间对应的样本操作次数占时间周期内的样本操作总次数的比值确定,具体可以是直接将每个第一样本时间对应的样本操作次数占时间周期内的样本操作总次数的比值,确定为每个第一样本时间对应的样本操作频率,可以是将每个第一样本时间对应的样本操作次数占时间周期内的样本操作总次数的比值,与设定权重值,共同确定每个第一样本时间对应的样本操作频率,如将比值与设定权重值的乘积确定为样本操作频率。
在确定每个第一样本时间及每个第一样本时间对应的样本操作频率输入到操作频率确定模型中,对操作频率确定模型进行训练。
根据用于训练的数据对模型进行训练的过程可以采用现有技术实现,在本公开实施例中不做赘述。
本公开实施例中通过对操作频率确定模型进行训练,保证了在进行门锁控制时,能够准确地确定操作频率及门锁的供电增益。
实施例6:
在上述各实施例的基础上,本公开实施例中,所述操作频率确定模型的训练过程包括:
根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;
针对预先保存的每个时间段,将该时间段作为样本时间段;在训练集中获取处于该样 本时间段内的每个第二样本时间,并获取所述每个第二样本时间对应的样本操作次数;根据所述每个第二样本时间对应的样本操作次数,统计用户在该样本时间段内的样本操作次数;并根据所述用户在该样本时间段内的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该样本时间段对应的样本操作频率;
将每个样本时间段及每个样本时间段对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
本公开实施例中操作频率确定模型可以是基于机器学习训练完成的模型,在本公开实施例是根据时间段对应的操作频率进行训练。
终端中预先保存有每个时间段,为了确定每个时间段对应的样本操作频率,需要先确定预设的时间周期内的样本操作总次数。
在确定时间周期内的样本操作总次数时,根据训练集中每个第一样本时间对应的样本操作次数确定,具体可以是将训练集中每个第一样本时间对应的样本操作次数的和确定为时间周期内的样本总次数。
在确定每个时间段对应的样本操作频率时,需要先确定每个时间段对应的样本操作次数,即训练集中用户在每个时间段内对门锁的样本操作次数。
在确定每个时间段对应的样本操作次数,将该每个时间段作为样本时间段,在训练集中获取处于样本时间段内的每个第二样本时间,并获取每个第二样本时间对应的样本操作次数,针对每个样本时间段,根据该样本时间段内每个第二样本时间对应的样本操作次数,统计该样本时间段对应的样本操作次数。具体地,可以将该样本时间段内每个样本时间对应的样本操作次数的和,确定为该样本时间段对应的样本操作次数。
针对每个样本时间段对应的样本操作次数后,可以针对每个样本时间对应的样本操作频率进行确定。
在确定每个样本时间段对应的样本操作频率时,根据每个样本时间段对应的样本操作次数占时间周期内的样本操作总次数的比值确定,具体可以是直接将每个样本时间段对应的样本操作次数占时间周期内的样本操作总次数的比值,确定为每个样本时间段对应的样本操作频率,可以是将每个样本时间段对应的样本操作次数占时间周期内的样本操作总次数的比值,与设定权重值,共同确定每个样本时间段对应的样本操作频率,如将比值与设定权重值的乘积确定为样本操作频率。
在确定每个样本时间段及每个样本时间段对应的样本操作频率输入到操作频率确定模型中,对操作频率确定模型进行训练。
根据用于训练的数据对模型进行训练的过程可以采用现有技术实现,在本公开实施例 中不做赘述。
本公开实施例中通过对操作频率确定模型进行训练,保证了在进行门锁控制时,能够准确地确定操作频率及门锁的供电增益。
实施例7:
在上述各实施例的基础上,本公开实施例中,所述方法还包括:
当识别到用户操作门锁时,获取用户操作门锁的时间;
在所述训练集中,将与所述用户操作门锁的时间匹配的样本时间对应的样本操作次数增加设定次数。
本公开提供了训练集中数据的收集方式,从而根据训练集中的数据对操作频率确定模型进行训练。
当识别到用户操作门锁时,获取用户操作门锁的时间。控制设备识别用户是否操作了门锁,可以是当门锁被用户操作时向控制设备发送相应信息。该获取用户操作门锁的时间与上述获取当前时间的过程相似在本公开实施例中不做赘述。
获取到用户操作门锁的时间后,在训练集中,将与用户操作门锁的时间匹配的样本时间对应的样本操作次数增加设定次数。该设定次数可以保存控制设备中,例如该设定次数可以为1、3或5等,在本公开实施例中不做限定。
训练集中数据更新后,可以根据更新数据后的训练集中对操作频率确定模型继续进行训练,从而实现对门锁的供电增益的更新。
下面以一个具体的实施例对本公开实施例进行说明,如图4所示,将用户在一天中每个时刻的操作数据作为输入数据整理到图表中,每一次的操作都会在数据库中的图表对应的值叠加1,这样将(每一时刻所对应的图谱面积/总图谱面积)作为供电增益调整的策略指示,根据训练集中用户操作的历史数据,以及机器学习归纳出的经验,以及输入到如图3所示的图标中的采集到的新的用户使用门锁的数据,对图像中的数据进行处理,即继续对操作频率确定模型训练,从而实现供电增益时间段的更新,进而实现对门锁的供电增益的控制。
实施例8:
在上述各实施例的基础上,本公开实施例中,所述方法还包括:
针对训练集中的每个第一样本时间,判断该第一样本时间是否属于周末或节假日;如果任一为是,则在训练集中将该第一样本时间对应的样本操作次数删除。
为了提高训练得到的操作频率确定模型的准确性,本公开实施例中可以去除周末或节假日的杂乱不规律的数据造成的影响。
针对训练集中的每个第一样本时间,可以根据日历判断该第一样本时间是否属于周末或节假日,如果任一为是,即如果该第一样本时间属于周末或节假日,由于认为周末或节假日中用户操作的数据为杂乱信息,因此可以将该第一样本时间对应的样本操作次数删除。如果该第一样本时间既不属于周末也不属于节假日,则可以在训练集中保留该第一样本时间对应的样本操作次数。
实施例9:
在上述各实施例的基础上,本公开实施例还提供了一种控制设备500,如图5所示,包括:处理器501和存储器503,可选地,还可以包括通信接口502和通信总线504等。其中,处理器501,通信接口502,存储器503通过通信总线504完成相互间的通信;
所述存储器503中存储有计算机程序,当所述计算机程序被所述处理器501执行时,使得所述处理器501执行任一个实施例的门锁控制方法的步骤,例如执行如下步骤:
获取当前时间;
将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内;
根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中操作频率越高供电增益越大;
将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
本公开实施例提供的控制设备具体可以为桌面计算机、服务器、网络侧设备等电子设备。
上述控制设备提到的通信总线可以是外设部件互连标准(Peripheral Component Interconnect,PCI)总线或扩展工业标准结构(Extended Industry Standard Architecture,EISA)总线等。该通信总线可以分为地址总线、数据总线、控制总线等。为便于表示,图中仅用一条粗线表示,但并不表示仅有一根总线或一种类型的总线。
通信接口502用于上述控制设备与其他设备之间的通信。
存储器可以包括随机存取存储器(Random Access Memory,RAM),也可以包括非易失性存储器(Non-Volatile Memory,NVM),例如至少一个磁盘存储器。可选地,存储器还可以是至少一个位于远离前述处理器的存储装置。
上述处理器可以是通用处理器,包括中央处理器、网络处理器(Network Processor,NP)等;还可以是数字指令处理器(Digital Signal Processing,DSP)、专用集成电路、 现场可编程门陈列或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。
实施例10:
在上述各实施例的基础上,本公开实施例还提供了一种非瞬时性计算机存储可读存储介质,所述计算机可读存储介质内存储有可由控制设备执行的计算机程序,当所述计算机程序在所述控制设备上运行时,使得所述控制设备执行时实现任一个实施例的门锁控制方法的步骤,例如实现如下步骤:
获取当前时间;
将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内;
根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中操作频率越高供电增益越大;
将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
上述计算机可读存储介质可以是控制设备中的处理器能够存取的任何可用介质或数据存储设备,包括但不限于磁性存储器如软盘、硬盘、磁带、磁光盘(MO)等、光学存储器如CD、DVD、BD、HVD等、以及半导体存储器如ROM、EPROM、EEPROM、非易失性存储器(NAND FLASH)、固态硬盘(SSD)等。
图6为本公开实施例提供的一种门锁控制装置600示意图,应用于控制设备,该装置包括:
获取模块601,用于获取当前时间;
第一确定模块602,用于将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内;
第二确定模块603,用于根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中在对应关系中操作频率越高供电增益越大;
控制模块604,用于将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
所述装置还包括:第三确定模块605,用于根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段;所述第一确定模块602, 还用于将所述目标时间段输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述目标时间段对应的目标操作频率,其中所述目标时间段对应的目标操作频率为预测的用户在所述目标时间段内的操作次数占预设的时间周期内的操作总次数的比值,所述目标时间段位于所述时间周期内。
所述第三确定模块605,还用于判断上一次确定目标时间段的时间距离所述当前时间是否到达了预先保存的时间段对应的时间间隔;如果是,根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段。
所述第一确定模块602,还用于根据预先保存的用户频繁操作的时间段的标识信息,判断所述当前时间所处的目标时间段及上一次确定的目标时间段中,是否只有一个目标时间段为用户频繁操作的时间段;如果是,将所述目标时间段输入到预先训练完成的操作频率确定模型中。
所述装置还包括:第一训练模块606,用于根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;针对每个第一样本时间,根据用户在该第一样本时间的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该第一样本时间内对应的样本操作频率;将每个第一样本时间及每个第一样本时间对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
所述装置还包括:第二训练模块607,用于根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;针对预先保存的每个时间段,将该时间段作为样本时间段;在训练集中获取处于该样本时间段内的每个第二样本时间,并获取所述每个第二样本时间对应的样本操作次数;根据所述每个第二样本时间对应的样本操作次数,统计用户在该样本时间段内的样本操作次数;并根据所述用户在该样本时间段内的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该样本时间段对应的样本操作频率;将每个样本时间段及每个样本时间段对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
所述装置还包括:增加模块608,用于当识别到用户操作门锁时,获取用户操作门锁的时间;在所述训练集中,将与所述用户操作门锁的时间匹配的样本时间对应的样本操作次数增加设定次数。
所述装置还包括:删除模块609,用于针对训练集中的每个第一样本时间,判断该第一样本时间是否属于周末或节假日;如果任一为是,则在训练集中将该第一样本时间对应的样本操作次数删除。
本公开实施例中能够根据当前时间,基于预先训练完成的操作频率确定模型,确定当 前时间对应的目标操作频率,并确定目标操作频率对应的目标供电增益,门锁的供电增益与当前时间对应的用户操作频率有关,并且当前时间对应的用户操作频率越高供电增益越大,因此可以差异化针对性地调整门锁的供电增益,能够提高用户操作的灵活性,提高了用户体验,并且能够在用户开门需求较低时采用较低供电增益,减少了门锁的电量消耗。
对于系统/装置实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者一个操作与另一个实体或者另一个操作区分开来,而不一定要求或者暗示这些实体或者操作之间存在任何这种实际的关系或者顺序。
本领域内的技术人员应明白,本申请的实施例可提供为方法、系统、或计算机程序产品。因此,本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本申请的优选实施例,但本领域内的技术人员一旦得知了基本创造性概念,则可对这些实施例做出另外的变更和修改。所以,所附权利要求意欲解释为包括优选 实施例以及落入本申请范围的所有变更和修改。
显然,本领域的技术人员可以对本公开进行各种改动和变型而不脱离本公开的精神和范围。这样,倘若本公开的这些修改和变型属于本公开权利要求及其等同技术的范围之内,则本公开也意图包含这些改动和变型在内。

Claims (18)

  1. 一种门锁控制方法,该方法包括:
    获取当前时间;
    将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内;
    根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中在所述对应关系中操作频率越高供电增益越大;
    将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
  2. 如权利要求1所述的方法,在所述获取当前时间之后,以及在所述将所述当前时间输入到预先训练完成的操作频率确定模型中之前,所述方法还包括:
    根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段;
    所述将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率包括:
    将所述目标时间段输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述目标时间段对应的目标操作频率,其中所述目标时间段对应的目标操作频率为预测的用户在所述目标时间段内的操作次数占预设的时间周期内的操作总次数的比值,所述目标时间段位于所述时间周期内。
  3. 如权利要求2所述的方法,所述方法还包括:
    判断上一次确定目标时间段的时间距离所述当前时间是否到达了预先保存的时间段对应的时间间隔;
    如果是,执行所述根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段的步骤。
  4. 如权利要求2所述的方法,所述方法还包括:
    根据预先保存的用户频繁操作的时间段的标识信息,判断所述当前时间所处的目标时间段及上一次确定的目标时间段中,是否只有一个目标时间段为用户频繁操作的时间段;
    如果是,执行所述将所述目标时间段输入到预先训练完成的操作频率确定模型中的步 骤。
  5. 如权利要求1所述的方法,其中,所述操作频率确定模型的训练过程包括:
    根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;
    针对每个第一样本时间,根据用户在该第一样本时间的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该第一样本时间内对应的样本操作频率;
    将每个第一样本时间及每个第一样本时间对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
  6. 如权利要求2所述的方法,其中,所述操作频率确定模型的训练过程包括:
    根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;
    针对预先保存的每个时间段,将该时间段作为样本时间段;在训练集中获取处于该样本时间段内的每个第二样本时间,并获取所述每个第二样本时间对应的样本操作次数;根据所述每个第二样本时间对应的样本操作次数,统计用户在该样本时间段内的样本操作次数;并根据所述用户在该样本时间段内的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该样本时间段对应的样本操作频率;
    将每个样本时间段及每个样本时间段对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
  7. 如权利要求5或6所述的方法,所述方法还包括:
    当识别到用户操作门锁时,获取用户操作门锁的时间;
    在所述训练集中,将与所述用户操作门锁的时间匹配的样本时间对应的样本操作次数增加设定次数。
  8. 如权利要求5或6所述的方法,所述方法还包括:
    针对训练集中的每个第一样本时间,判断该第一样本时间是否属于周末或节假日;如果属于周末或节假日,则在训练集中将该第一样本时间对应的样本操作次数删除。
  9. 一种门锁控制装置,该装置包括:
    获取模块,用于获取当前时间;
    第一确定模块,用于将所述当前时间输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述当前时间对应的目标操作频率,其中所述当前时间对应的目标操作频率为预测的用户在所述当前时间的操作次数占预设的时间周期内的操作总次数的比值,所述当前时间位于所述时间周期内;
    第二确定模块,用于根据预先保存的操作频率与供电增益的对应关系,确定所述目标操作频率对应的目标供电增益,其中在所述对应关系中操作频率越高供电增益越大;
    控制模块,用于将所述目标供电增益作为门锁的供电增益,对所述门锁进行控制。
  10. 如权利要求9所述的装置,所述装置还包括:
    第三确定模块,用于根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段;
    所述第一确定模块,还用于将所述目标时间段输入到预先训练完成的操作频率确定模型中,基于所述操作频率确定模型,确定所述目标时间段对应的目标操作频率,其中所述目标时间段对应的目标操作频率为预测的用户在所述目标时间段内的操作次数占预设的时间周期内的操作总次数的比值,所述目标时间段位于所述时间周期内。
  11. 如权利要求10所述的装置,其中,所述第三确定模块,还用于判断上一次确定目标时间段的时间距离所述当前时间是否到达了预先保存的时间段对应的时间间隔;如果是,根据所述当前时间,及预先保存的时间周期中包含的每个时间段,确定所述当前时间所处的目标时间段。
  12. 如权利要求10所述的装置,其中,所述第一确定模块,还用于根据预先保存的用户频繁操作的时间段的标识信息,判断所述当前时间所处的目标时间段及上一次确定的目标时间段中,是否只有一个目标时间段为用户频繁操作的时间段;如果是,将所述目标时间段输入到预先训练完成的操作频率确定模型中。
  13. 如权利要求9所述的装置,所述装置还包括:
    第一训练模块,用于根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;针对每个第一样本时间,根据用户在该第一样本时间的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该第一样本时间内对应的样本操作频率;将每个第一样本时间及每个第一样本时间对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
  14. 如权利要求10所述的装置,所述装置还包括:
    第二训练模块,用于根据训练集中每个第一样本时间对应的样本操作次数,统计预设的时间周期内的样本操作总次数;针对预先保存的每个时间段,将该时间段作为样本时间段;在训练集中获取处于该样本时间段内的每个第二样本时间,并获取所述每个第二样本时间对应的样本操作次数;根据所述每个第二样本时间对应的样本操作次数,统计用户在该样本时间段内的样本操作次数;并根据所述用户在该样本时间段内的样本操作次数占所述时间周期内的样本操作总次数的比值,确定该样本时间段对应的样本操作频率;将每个 样本时间段及每个样本时间段对应的样本操作频率输入到操作频率确定模型中,对所述操作频率确定模型进行训练。
  15. 如权利要求13或14所述的装置,所述装置还包括:
    增加模块,用于当识别到用户操作门锁时,获取用户操作门锁的时间;在所述训练集中,将与所述用户操作门锁的时间匹配的样本时间对应的样本操作次数增加设定次数。
  16. 如权利要求13或14所述的装置,所述装置还包括:
    删除模块,用于针对训练集中的每个第一样本时间,判断该第一样本时间是否属于周末或节假日;如果属于周末或节假日,则在训练集中将该第一样本时间对应的样本操作次数删除。
  17. 一种控制设备,包括:处理器和存储器;
    所述存储器中存储有计算机程序,当所述计算机程序被所述处理器执行时,使得所述处理器执行权利要求1~8任一项所述方法的步骤。
  18. 一种非瞬时性计算机可读存储介质,其存储有可由控制设备执行的计算机程序,当所述计算机程序在所述控制设备上运行时,使得所述控制设备执行权利要求1~8任一项所述方法的步骤。
PCT/CN2019/101018 2018-10-15 2019-08-16 门锁控制方法、装置、控制设备 WO2020078093A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811196162.1 2018-10-15
CN201811196162.1A CN109461231B (zh) 2018-10-15 2018-10-15 一种门锁控制方法、装置、控制设备及可读存储介质

Publications (1)

Publication Number Publication Date
WO2020078093A1 true WO2020078093A1 (zh) 2020-04-23

Family

ID=65607589

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/101018 WO2020078093A1 (zh) 2018-10-15 2019-08-16 门锁控制方法、装置、控制设备

Country Status (2)

Country Link
CN (1) CN109461231B (zh)
WO (1) WO2020078093A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109461231B (zh) * 2018-10-15 2021-07-23 珠海格力电器股份有限公司 一种门锁控制方法、装置、控制设备及可读存储介质
CN113898254B (zh) * 2021-10-11 2022-07-22 石家庄华泰电力工具有限公司 远程控制方法及适用于智能安全管理柜的管理系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010117354A1 (en) * 2009-04-06 2010-10-14 Utc Fire & Security Corporation Power management circuitry for electronic door locks
CN106056712A (zh) * 2016-06-07 2016-10-26 杭州虹云网络科技有限公司 一种低功耗的智能门锁及其降低功耗的方法
CN106557146A (zh) * 2015-09-28 2017-04-05 北京金山安全软件有限公司 一种省电方法、装置及终端
CN107728772A (zh) * 2017-09-30 2018-02-23 广东欧珀移动通信有限公司 应用程序的处理方法、装置、存储介质及电子设备
CN108234758A (zh) * 2017-12-28 2018-06-29 广东欧珀移动通信有限公司 应用的显示方法、装置、存储介质及电子设备
CN109461231A (zh) * 2018-10-15 2019-03-12 珠海格力电器股份有限公司 一种门锁控制方法、装置、控制设备及可读存储介质

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103294158B (zh) * 2012-03-05 2016-12-14 联想(北京)有限公司 一种信息处理方法及电子设备
US9347242B2 (en) * 2013-10-28 2016-05-24 Smartlabs, Inc. Systems and methods to automatically detect a door state
CN105094272A (zh) * 2014-05-14 2015-11-25 中兴通讯股份有限公司 一种终端的硬件刷新率的调节方法及装置
CN104952131A (zh) * 2015-06-19 2015-09-30 福建星网锐捷通讯股份有限公司 一种楼宇对讲系统访客来访时间登记方法及装置
CN105096418A (zh) * 2015-07-23 2015-11-25 青岛文创科技有限公司 一种新型安全门控系统
CN106549833B (zh) * 2015-09-21 2020-01-21 阿里巴巴集团控股有限公司 一种智能家居设备的控制方法和装置
CN106023381A (zh) * 2016-06-17 2016-10-12 北京千丁互联科技有限公司 蓝牙终端、蓝牙门禁管理系统及方法
CN106527664B (zh) * 2016-11-07 2020-10-30 东软集团股份有限公司 自助终端节能控制方法及装置
CN106803301A (zh) * 2017-03-28 2017-06-06 广东工业大学 一种基于深度学习的人脸识别门禁方法及系统
CN107707759B (zh) * 2017-10-16 2021-01-15 北京小米移动软件有限公司 终端控制方法、装置以及系统、存储介质
CN107943583B (zh) * 2017-11-14 2022-02-22 Oppo广东移动通信有限公司 应用程序的处理方法、装置、存储介质及电子设备
CN108022352A (zh) * 2017-12-27 2018-05-11 北京洪泰同创信息技术有限公司 门禁控制方法、门禁终端及门禁系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010117354A1 (en) * 2009-04-06 2010-10-14 Utc Fire & Security Corporation Power management circuitry for electronic door locks
CN106557146A (zh) * 2015-09-28 2017-04-05 北京金山安全软件有限公司 一种省电方法、装置及终端
CN106056712A (zh) * 2016-06-07 2016-10-26 杭州虹云网络科技有限公司 一种低功耗的智能门锁及其降低功耗的方法
CN107728772A (zh) * 2017-09-30 2018-02-23 广东欧珀移动通信有限公司 应用程序的处理方法、装置、存储介质及电子设备
CN108234758A (zh) * 2017-12-28 2018-06-29 广东欧珀移动通信有限公司 应用的显示方法、装置、存储介质及电子设备
CN109461231A (zh) * 2018-10-15 2019-03-12 珠海格力电器股份有限公司 一种门锁控制方法、装置、控制设备及可读存储介质

Also Published As

Publication number Publication date
CN109461231A (zh) 2019-03-12
CN109461231B (zh) 2021-07-23

Similar Documents

Publication Publication Date Title
CN110298501B (zh) 基于长短时记忆神经网络的电负荷预测方法
CN107463329B (zh) 黑屏手势的检测方法、装置、存储介质及移动终端
CN107678845B (zh) 应用程序管控方法、装置、存储介质及电子设备
US10673707B2 (en) Systems and methods for managing lifecycle and reducing power consumption by learning an IoT device
CN107515890A (zh) 一种识别常驻点的方法及终端
WO2019062417A1 (zh) 应用清理方法、装置、存储介质及电子设备
CN105468823B (zh) 一种自助设备节能控制方法和装置
KR20170141248A (ko) 컴퓨팅 디바이스의 사용자에게 권고들을 선행적으로 제공하기 위한 시스템들 및 방법들
CN105068710A (zh) 一种桌面应用图标的动态排列方法及系统、移动终端
CN108829457A (zh) 应用程序预测模型更新方法、装置、存储介质及终端
WO2020078093A1 (zh) 门锁控制方法、装置、控制设备
CN107567083A (zh) 进行省电优化处理的方法和装置
CN112150311A (zh) 能耗异常状态检测方法、装置、计算机设备和存储介质
CN107886028A (zh) 一种冰箱的食材录入方法及食材录入装置
CN107391021B (zh) 黑屏手势的识别方法、装置、存储介质及移动终端
CN107479816A (zh) 黑屏手势的识别方法、装置、存储介质及移动终端
US11481014B2 (en) Power control method based on user habit and terminal
CN106775193A (zh) 一种应用程序界面显示方法及装置
WO2019085754A1 (zh) 应用清理方法、装置、存储介质及电子设备
CN107402713A (zh) 加快黑屏手势处理的方法、装置、存储介质及移动终端
CN109685255A (zh) 一种预测用户流失的方法和装置
CN113934615A (zh) 一种数据监控方法、装置及设备
CN107145215A (zh) 一种后台应用程序清理方法及移动终端
CN107748697B (zh) 应用关闭方法、装置、存储介质及电子设备
CN109951515A (zh) 卫生间信息推送方法、终端设备及计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19872990

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19872990

Country of ref document: EP

Kind code of ref document: A1