Deodorization control method and system of intelligent cat litter box

Abstract

The application provides a deodorization control method and system of an intelligent cat litter box, relates to the technical field of control, and comprises the following steps: obtaining real-time data collected by a plurality of sensors, judging a cat entering state, monitoring gas concentration changes to start a deodorization system, establishing a cat toilet time regularity model to predict the next toilet time and preheat the deodorization system in advance, and dynamically adjusting the rotating speed of a deodorization fan according to the weight change rate of cat litter. The application realizes intelligent deodorization control, improves deodorization efficiency, improves the use environment of the cat litter box, and reduces energy consumption.

Description

Technical Field

[0001] This invention relates to the field of control technology, and in particular to a method and system for deodorizing intelligent cat litter boxes. Background Technology

[0002] With the improvement of people's living standards and the increasing popularity of pet ownership, cats have become important companions in many families. One of the main challenges pet owners face when keeping cats indoors is managing the odor of the litter box. Traditional litter boxes typically only have the basic function of collecting cat excrement and cannot effectively solve the problem of odor diffusion, affecting the comfort and hygiene of the home environment.

[0003] The main drawbacks of existing smart litter boxes are as follows: most smart litter boxes use a simple timed deodorization mode, which cannot flexibly adjust the deodorization intensity and timing according to changes in actual gas concentration, resulting in poor deodorization effect or waste of resources; existing technologies lack the ability to learn and predict cats' toilet habits, and cannot prepare in advance before the cat is about to use the litter box, thus failing to achieve preventive deodorization; existing deodorization systems usually operate with fixed parameters, failing to dynamically adjust the working intensity of the deodorizing fan according to the amount of excrement, which cannot meet the deodorization needs under heavy pollution conditions, and will also cause energy waste under mild pollution conditions. Summary of the Invention

[0004] This invention provides a method and system for deodorizing a smart cat litter box, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides a method for deodorizing a smart cat litter box, comprising:

[0006] The system determines whether a cat has entered the smart litter box based on data collected by an infrared sensor. When a cat is detected entering the smart litter box, the system monitors the gas concentration data collected by a gas concentration sensor. When the rate of change of the gas concentration data exceeds a preset rate threshold, the system controls the deodorization system to start working.

[0007] Based on the historical usage data of the smart litter box, the next toilet time interval is predicted according to the preset cat toilet time pattern model. The deodorization system is preheated and deodorant is sprayed in advance during the predicted toilet time interval.

[0008] After the cat leaves the smart litter box, the weight change rate of the litter is calculated based on the weight change data collected by the weight sensor. The speed of the deodorizing fan is dynamically adjusted according to the comparison result of the weight change rate of the litter and the preset weight change rate threshold.

[0009] Monitoring gas concentration data collected by a gas concentration sensor, and controlling the deodorization system to start operation when the rate of change of the gas concentration data exceeds a preset rate threshold includes:

[0010] The data acquisition time interval parameter of the gas concentration sensor installed in the smart litter box is obtained, and the gas concentration sensor is controlled to periodically acquire gas concentration data in the smart litter box according to the data acquisition time interval parameter.

[0011] A moving average filtering algorithm is used to process gas concentration data from multiple consecutive sampling time points to obtain smoothed gas concentration data.

[0012] The rate of change of the smoothed gas concentration data is calculated in real time. When the rate of change of the gas concentration data exceeds a preset rate threshold, a deodorization system start signal is generated. Based on the deodorization system start signal, the deodorization system is controlled to enter the working state.

[0013] The gas concentration data from multiple consecutive sampling time points are processed using a moving average filtering algorithm to obtain smoothed gas concentration data, including:

[0014] A raw gas concentration data sequence is obtained based on the gas concentration data, and the size of the moving average window is determined based on the raw gas concentration data sequence.

[0015] A weighting coefficient is set for each sampling time point within the moving average window, and a weighted average calculation is performed on the gas concentration data within the moving average window based on the weighting coefficient to obtain the smoothed gas concentration value at the current time.

[0016] The moving average window is moved forward one sampling time point at a time, and the weighted average calculation is repeated until all gas concentration data at all sampling time points have been processed.

[0017] The weighting coefficients of the starting and ending segments of the moving average window are corrected to obtain a smoothed gas concentration data sequence.

[0018] Based on the cat's littering time pattern model, the system predicts the next littering time interval and preheats the deodorizing system and sprays deodorant during the predicted littering time interval, including:

[0019] Collect the time points and durations of each littering event of a cat to obtain historical data containing time point sequences and duration sequences;

[0020] The historical data is preprocessed by standardization. The time series is subtracted from its mean and then divided by its standard deviation to obtain a standardized time series. The standardized time series is then subjected to time series analysis to extract the temporal regularity features of cat toileting behavior and to establish a cat toileting time regularity model.

[0021] Based on the cat's toilet time pattern model, the time interval of the next toilet event is predicted to obtain the predicted time interval; the preheating start time of the deodorization system is determined according to the predicted time interval.

[0022] The deodorization system is started at the preheating start time to preheat, and the spraying amount of deodorant is adaptively controlled according to the predicted duration within the predicted time interval.

[0023] Based on the cat's littering time pattern model, the time interval of the next littering event is predicted to obtain the predicted time interval; the preheating start time of the deodorization system is determined according to the predicted time interval, including:

[0024] The cat's toilet time pattern model is used to predict the time of the next toilet event. By determining the start and end times of the prediction time interval under a preset confidence level, the complete prediction time interval is obtained.

[0025] Calculate the prediction error between the complete predicted time interval and the actual toilet time, and dynamically correct the complete predicted time interval based on the prediction error to obtain the corrected predicted time interval.

[0026] The preheating start time of the deodorization system is determined based on the corrected prediction time interval, and the deodorization system is controlled to start preheating operation according to the preheating start time, so as to realize the advance preheating preparation of the deodorization system.

[0027] Based on the cat litter weight change data collected by the weight sensor, the cat litter weight change rate is calculated, and the deodorizing fan speed is dynamically adjusted according to the comparison result of the cat litter weight change rate and the preset weight change rate threshold, including:

[0028] The weight sensor is used to collect cat litter weight change data to obtain cat litter weight values ​​at continuous time points; the weight change at adjacent time points is calculated based on the cat litter weight values, and the cat litter weight change rate is calculated based on the weight change and the corresponding time interval;

[0029] The cat litter weight change rate is compared with a preset weight change rate threshold to obtain the deviation value of the weight change rate;

[0030] The target speed of the deodorizing fan is determined based on the deviation value. When the cat litter weight change rate is greater than the preset weight change rate threshold, the target speed of the deodorizing fan is increased. When the cat litter weight change rate is less than the preset weight change rate threshold, the target speed of the deodorizing fan is decreased.

[0031] The current speed of the deodorizing fan is gradually adjusted to the target speed to achieve dynamic adjustment of the deodorizing fan speed.

[0032] A second aspect of the present invention provides an odor control system for an intelligent cat litter box, comprising:

[0033] The first unit is used to determine whether a cat has entered the smart litter box based on data collected by an infrared sensor. When a cat is detected to have entered the smart litter box, the unit monitors the gas concentration data collected by a gas concentration sensor. When the rate of change of the gas concentration data exceeds a preset rate threshold, the unit controls the deodorization system to start working.

[0034] The second unit is used to predict the next toilet time interval based on the historical usage data of the smart litter box and a preset model of cat toilet time patterns, and to preheat the deodorization system and spray deodorant in advance during the predicted toilet time interval.

[0035] The third unit is used to calculate the cat litter weight change rate based on the cat litter weight change data collected by the weight sensor after the cat leaves the smart litter box, and dynamically adjust the speed of the deodorizing fan according to the comparison result of the cat litter weight change rate and the preset weight change rate threshold.

[0036] A third aspect of the present invention provides an electronic device, comprising:

[0037] processor;

[0038] Memory used to store processor-executable instructions;

[0039] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0040] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0041] The beneficial effects of this application are as follows:

[0042] The deodorization control method for intelligent cat litter boxes provided by this invention, through the collaborative work of multiple sensors, can accurately detect the behavior patterns of cats using the litter box and gas changes, realizing the intelligent start-up and regulation of the deodorization system, and effectively solving the problem of untimely and incomplete deodorization of traditional cat litter boxes.

[0043] Based on historical usage data, a model of cat littering time patterns is established, which can predict the time interval of the cat's next use of the litter box. This enables the preheating of the deodorization system and the pre-spraying of deodorizing agents, greatly improving deodorization efficiency, saving energy consumption, and enhancing the user experience of the smart litter box.

[0044] By dynamically adjusting the deodorizing fan speed based on the weight change data of cat litter collected by a weight sensor, the deodorizing intensity is precisely matched with the actual pollution level, avoiding over- or under-working of the deodorizing system. This ensures the deodorizing effect, extends the service life of the deodorizing system, and improves the overall system reliability and economy. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the deodorization control method of the intelligent cat litter box according to an embodiment of the present invention.

[0046] Figure 2 This is a flowchart illustrating the dynamic adjustment logic of the fan speed according to an embodiment of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0049] Figure 1 This is a flowchart illustrating the odor control method of the intelligent cat litter box according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0050] The system determines whether a cat has entered the smart litter box based on data collected by an infrared sensor. When a cat is detected entering the smart litter box, the system monitors the gas concentration data collected by a gas concentration sensor. When the rate of change of the gas concentration data exceeds a preset rate threshold, the system controls the deodorization system to start working.

[0051] Based on the historical usage data of the smart litter box, the next toilet time interval is predicted according to the preset cat toilet time pattern model. The deodorization system is preheated and deodorant is sprayed in advance during the predicted toilet time interval.

[0052] After the cat leaves the smart litter box, the weight change rate of the litter is calculated based on the weight change data collected by the weight sensor. The speed of the deodorizing fan is dynamically adjusted according to the comparison result of the weight change rate of the litter and the preset weight change rate threshold.

[0053] In one optional implementation, monitoring gas concentration data collected by a gas concentration sensor, and controlling the deodorization system to start operation when the rate of change of the gas concentration data exceeds a preset rate threshold includes:

[0054] The data acquisition time interval parameter of the gas concentration sensor installed in the smart litter box is obtained, and the gas concentration sensor is controlled to periodically acquire gas concentration data in the smart litter box according to the data acquisition time interval parameter.

[0055] A moving average filtering algorithm is used to process gas concentration data from multiple consecutive sampling time points to obtain smoothed gas concentration data.

[0056] The rate of change of the smoothed gas concentration data is calculated in real time. When the rate of change of the gas concentration data exceeds a preset rate threshold, a deodorization system start signal is generated. Based on the deodorization system start signal, the deodorization system is controlled to enter the working state.

[0057] The smart litter box is equipped with a gas concentration sensor and a deodorization system. The gas concentration sensor collects gas concentration data within the litter box, and when the detected gas concentration exceeds a preset threshold, it activates the deodorization system. The control system of the smart litter box includes a processor and a memory. The memory stores the control program, which, when executed by the processor, controls the gas concentration sensor and the deodorization system.

[0058] Gas concentration sensors, such as ammonia sensors, hydrogen sulfide sensors, or VOC sensors, are used to detect the concentration of odorous gases in the litter box. Deodorization systems may include components such as fans, activated carbon filters, and negative ion generators to remove odors from the litter box.

[0059] The system retrieves the data acquisition interval parameter set by the gas concentration sensor in the smart litter box from its memory. This parameter can be preset at the factory or set by the user via a smartphone application. For example, the data acquisition interval parameter of the gas concentration sensor can be set to 10 seconds, meaning that gas concentration data in the litter box is collected every 10 seconds.

[0060] Based on the data acquisition time interval parameter, the gas concentration sensor is controlled to periodically collect gas concentration data in the smart litter box. Taking an ammonia sensor as an example, the sensor collects ammonia concentration data in the litter box every 10 seconds and transmits the collected data to the processor of the control system.

[0061] A moving average filtering algorithm is used to process gas concentration data from multiple consecutive sampling time points to eliminate data fluctuations during the acquisition process, resulting in smoothed gas concentration data. The moving average filtering algorithm is a commonly used data smoothing method. Its basic idea is to take the average of the most recent N sampling points as the gas concentration value at the current moment. For example, when N=5, the gas concentration value at the current moment t is the average of the gas concentration values ​​at the most recent 5 sampling points (t, t-1, t-2, t-3, t-4). This effectively eliminates short-term data fluctuations, resulting in more stable gas concentration data.

[0062] In practice, a data buffer of length N can be maintained to store the gas concentration data of the most recent N sampling points. Whenever new sampling data arrives, it is added to the buffer while the oldest data is removed. Then, the average of all data in the buffer is calculated as the smoothed gas concentration value for the current moment. For example, assuming N=5, and the buffer stores the ammonia concentration data of the most recent 5 sampling points: 5ppm, 6ppm, 4ppm, 7ppm, and 5ppm, then the smoothed ammonia concentration value for the current moment is (5+6+4+7+5) / 5=5.4ppm.

[0063] The rate of change of the smoothed gas concentration data is calculated in real time. The rate of change can be obtained by dividing the difference in gas concentration between two adjacent sampling time points by the sampling time interval. For example, if the smoothed gas concentration at the current time t is 5.4 ppm and the smoothed gas concentration at the previous time t-1 is 4.8 ppm, and the sampling time interval is 10 seconds, then the rate of change of gas concentration is (5.4-4.8) / 10=0.06 ppm / second.

[0064] The calculated rate of change in gas concentration is compared with a preset rate threshold, which can be set according to actual needs, for example, to 0.05 ppm / second. When the rate of change in gas concentration exceeds the preset rate threshold, a deodorization system start signal is generated. In the example above, the calculated rate of change is 0.06 ppm / second, which exceeds the preset rate threshold of 0.05 ppm / second, thus generating the deodorization system start signal.

[0065] To activate the deodorization system, you can control the fan to start, increasing air circulation within the litter box; simultaneously, you can activate the activated carbon filter to absorb odorous gases; and you can activate the negative ion generator to release negative ions and decompose odor molecules in the air. These measures effectively reduce the concentration of odorous gases in the litter box, improving the user experience.

[0066] The deodorization system can operate for a preset time, such as 5 minutes, and then automatically stop; alternatively, it can continue operating until the gas concentration drops below another preset threshold, such as below 2 ppm, at which point the system automatically stops. This ensures effective deodorization while avoiding energy waste.

[0067] Using the methods described above, the smart litter box can monitor changes in gas concentration in the box in real time and activate the deodorization system promptly when the concentration of odorous gases rises rapidly, keeping the air inside the litter box fresh and improving the user experience.

[0068] In one optional implementation, a moving average filtering algorithm is used to process gas concentration data from multiple consecutive sampling time points to eliminate data fluctuations during the acquisition process, resulting in smoothed gas concentration data, including:

[0069] A raw gas concentration data sequence is obtained based on the gas concentration data, and the size of the moving average window is determined based on the raw gas concentration data sequence.

[0070] A weighting coefficient is set for each sampling time point within the moving average window, and a weighted average calculation is performed on the gas concentration data within the moving average window based on the weighting coefficient to obtain the smoothed gas concentration value at the current time.

[0071] The moving average window is moved forward one sampling time point at a time, and the weighted average calculation is repeated until all gas concentration data at all sampling time points have been processed.

[0072] The weighting coefficients of the starting and ending segments of the moving average window are corrected to obtain a smoothed gas concentration data sequence.

[0073] Gas concentration data is collected at multiple consecutive sampling time points. Sampling can be achieved through a gas sensor array. The sensors collect the concentration values ​​of specific gases in the environment at a preset sampling frequency (e.g., once per second), forming time-series data. For example, during a gas monitoring process, continuous sampling was conducted for 120 seconds at a frequency of once per second, obtaining carbon dioxide concentration data (unit: ppm) at 120 sampling points: [452, 461, 457, 463, 459, 465, 470, 468, 472, 467,...] (for simplicity, only some data are listed here). These data constitute the original gas concentration data sequence.

[0074] The moving average window size is determined based on the acquired raw gas concentration data sequence. The choice of window size must consider the data fluctuation characteristics and sampling frequency. Generally, a larger window provides better smoothing but leads to a lag in response to short-term changes; a smaller window better preserves the data trend but has limited smoothing effect. In practical applications, the window size can be determined by analyzing the fluctuation characteristics of the raw data or set according to actual needs. In this embodiment, the standard deviation of the raw data is calculated to be 5.8 ppm. Considering the relatively stable monitoring environment and the high requirement for data smoothness, a window size of 9 sampling points is selected.

[0075] Weighting coefficients are assigned to each sampling time point within the moving average window. The method of setting these weighting coefficients affects the smoothing effect, and different weighting strategies can be selected according to actual needs. Commonly used weighting settings include uniform weighting, center weighting, and exponential weighting. In this embodiment, a center weighting strategy is adopted, so that the center point of the window has the maximum weight, decreasing towards both sides. For a window size of 9, the weighting coefficients are set to [0.05, 0.08, 0.11, 0.15, 0.22, 0.15, 0.11, 0.08, 0.05], ensuring that the sum of all weights is 1. This weighting method can preserve the overall trend of the data while effectively suppressing short-term fluctuations.

[0076] The gas concentration data within the moving average window is weighted and averaged based on the set weight coefficients to obtain the smoothed gas concentration value at the current time. For the i-th sampling point (the center point of the window) in the original data sequence, the data of the nine sampling points before and after the i-th sampling point are taken, multiplied by the corresponding weight coefficients respectively, and then summed to obtain the smoothed concentration value of the i-th sampling point. Taking the 10th sampling point (value 467 ppm) in the original data as an example, the weighted average of its window [457, 463, 459, 465, 470, 468, 472, 467, 464] is calculated as follows: 457×0.05 + 463×0.08 + 459×0.11 + 465×0.15 + 470×0.22 + 468×0.15 + 472×0.11 + 467×0.08 + 464×0.05 = 466.32 ppm. That is, the smoothed concentration value of the 10th sampling point is 466.32 ppm.

[0077] The moving average window is moved forward one sampling time point at a time, and the weighted average calculation is repeated. The weighted average is calculated once each time the window moves forward one sampling point, until the gas concentration data of all sampling points have been processed. In this way, each sampling point in the original data sequence (except for the beginning and end portions that cannot be completely covered by the window) can obtain the corresponding smoothed concentration value.

[0078] The weighting coefficients of the starting and ending segments of the moving average window are adjusted to eliminate data fluctuations caused by the window boundaries. Since the window size is 9, the first 4 and last 4 sampling points of the original data sequence cannot obtain a smoothed result for the entire window. To address this, a boundary processing technique is employed: for the starting segment (the first 4 sampling points), the weighting coefficients are adjusted. Specifically, for the first sampling point, the first 4 sampling points are weighted using weights of [0.5, 0.25, 0.15, 0.1]; for the second sampling point, the first 5 sampling points are weighted using weights of [0.35, 0.3, 0.2, 0.1, 0.05]; and so on, until the window can be fully applied. A similar method is used for the ending segment (the last 4 sampling points). In this way, the window boundary processing problem is solved, and a completely smoothed gas concentration data sequence is obtained.

[0079] After the above processing, the original gas concentration data [452, 461, 457, 463, 459, 465, 470, 468, 472, 467, ...] was smoothed to [453.1, 456.8, 458.2, 460.5, 462.7, 464.9, 466.8, 467.5, 466.9, 466.3, ...]. The results show that the smoothed data curve is more stable, effectively eliminating random fluctuations in the original data while preserving the overall trend of gas concentration changes, thus improving the reliability and practical value of the data.

[0080] This method is applicable to various gas concentration monitoring scenarios, such as industrial safety monitoring, environmental quality monitoring, and indoor air quality monitoring. It can significantly improve the quality of gas concentration data and provide more reliable data support for subsequent analysis and decision-making.

[0081] In one optional implementation, predicting the next toilet-use time interval based on the cat's toilet-use time pattern model, and preheating the deodorization system and spraying deodorant in advance during the predicted toilet-use time interval includes:

[0082] Collect the time points and durations of each littering event of a cat to obtain historical data containing time point sequences and duration sequences;

[0083] The historical data is preprocessed by standardization. The time series is subtracted from its mean and then divided by its standard deviation to obtain a standardized time series. The standardized time series is then subjected to time series analysis to extract the temporal regularity features of cat toileting behavior and to establish a cat toileting time regularity model.

[0084] Based on the cat's toilet time pattern model, the time interval of the next toilet event is predicted to obtain the predicted time interval; the preheating start time of the deodorization system is determined according to the predicted time interval.

[0085] The deodorization system is started at the preheating start time to preheat, and the spraying amount of deodorant is adaptively controlled according to the predicted duration within the predicted time interval.

[0086] This embodiment includes a litter box, a sensor module, a data processing module, a prediction module, and a deodorization execution module. The sensor module is installed around the litter box to detect when the cat enters and leaves the litter box, obtaining the duration of the cat's toilet time. The data processing module receives and preprocesses the sensor data. The prediction module builds a model of the cat's toilet time pattern based on the processed data to predict the next toilet time event. The deodorization execution module controls the activation of the deodorization system and the spraying of the deodorizing agent based on the prediction results.

[0087] Collect the time points and durations of each litter box activity of the cat. For example, record the litter box data of a cat for 7 consecutive days: Day 1: litter box activities at 06:30, 12:15, and 18:45, with durations of 2 minutes, 1.5 minutes, and 2.5 minutes respectively; Day 2: litter box activities at 06:25, 12:30, and 19:00, with durations of 2.2 minutes, 1.8 minutes, and 2.3 minutes respectively; and so on up to Day 7. Organize this data into time point series and duration series for use as historical data in subsequent analysis.

[0088] To ensure model accuracy, historical data is standardized before preprocessing. The mean and standard deviation of the time series are calculated; for example, if the mean is 12:25 and the standard deviation is 5.6 hours, then the mean is subtracted from each time point, and the result is divided by the standard deviation to obtain the standardized time series. For example, 06:30 is 6.5 hours in hourly representation, and after standardization, it becomes (6.5 - 12.42) / 5.6 = -1.06. Similarly, the duration series is also standardized. For example, if the duration mean is 2.1 minutes and the standard deviation is 0.48 minutes, then the standardized value for 2 minutes is (2 - 2.1) / 0.48 = -0.21.

[0089] After standardization, time-series data were analyzed to extract the temporal patterns of cat littering behavior. Analysis of the daily distribution of littering times revealed that cats typically litter in the morning (06:25-06:35), at noon (12:15-12:30), and in the evening (18:45-19:00). The time intervals between adjacent littering events were also analyzed, finding an average interval of approximately 6 hours and a standard deviation of 0.3 hours. Furthermore, analysis of the duration of littering at different times revealed that evening littering sessions were typically 0.3-0.5 minutes longer than those in the morning and at noon.

[0090] Based on these analytical results, a cat littering time pattern model was established. This model employs a deep neural network structure, comprising an input layer, three hidden layers, and an output layer. The input layer receives standardized time-series data and has 30 neurons, corresponding to 30 days of historical data. The first hidden layer contains 64 neurons, using the ReLU activation function to extract basic time-series features. The second hidden layer contains 32 neurons, also using the ReLU activation function, to further extract higher-level time features. The third hidden layer contains 16 neurons, using the tanh activation function to perform a non-linear transformation on the features. The output layer contains two neurons, predicting the start and end times of the next littering session. This model comprehensively considers intraday periodicity, time interval regularity, and duration characteristics, enabling it to predict the time range of the cat's next littering session. For example, if the current time is 08:00 AM on day 8, the cat completed its first littering session of the day at 06:28, lasting 2.1 minutes. Based on historical data, the model predicts that the time range for the second toilet visit on the same day will be 12:20±15 minutes (i.e., 12:05-12:35), and the expected duration will be 1.7±0.3 minutes.

[0091] The preheating start time of the deodorization system is determined based on the predicted time interval. Assuming the deodorization system requires 5 minutes to complete preheating, preheating will begin 5 minutes before the start of the predicted time interval (i.e., 12:00). During the preheating process, parameters such as the remaining amount of deodorant and the temperature of the heating element are checked to ensure they are in optimal working condition.

[0092] After the deodorizing system preheats, it enters standby mode within the predicted time interval (12:05-12:35). When the sensor detects a cat entering the litter box, confirming the start of the toilet activity, it begins timing and prepares to spray deodorant. The amount of deodorant sprayed is adaptively controlled based on the predicted duration (1.7 ± 0.3 minutes). For example, if the predicted duration is short (approximately 1.4 minutes), the amount of deodorant sprayed will be reduced to 80% of the standard amount; if the predicted duration is long (approximately 2.0 minutes), the amount of deodorant sprayed will be increased to 120% of the standard amount.

[0093] When the sensor detects that the cat has left the litter box, confirming the end of the toilet-use event, the actual duration is recorded, and then the deodorization program is activated. The amount of deodorant sprayed is adjusted based on the difference between the actual and predicted duration. For example, if the actual duration is 2.2 minutes, 0.5 minutes longer than the predicted value, the amount of deodorant sprayed is increased by 25%. Simultaneously, the data from this toilet-use event (time point, duration, etc.) is added to the historical data to update the cat's toilet-use time pattern model, improving the accuracy of future predictions.

[0094] The above methods can accurately predict cats' litter box behavior, preheat the deodorizing system, and adaptively control the deodorizing agent spraying according to specific circumstances, effectively reducing litter box odor and improving the quality of the home environment.

[0095] In one optional implementation, the time interval of the next toilet-use event is predicted based on the cat's toilet-use time pattern model to obtain a predicted time interval; the preheating start-up time point of the deodorization system is determined according to the predicted time interval, including:

[0096] The cat's toilet time pattern model is used to predict the time of the next toilet event. By determining the start and end times of the prediction time interval under a preset confidence level, the complete prediction time interval is obtained.

[0097] Calculate the prediction error between the complete predicted time interval and the actual toilet time, and dynamically correct the complete predicted time interval based on the prediction error to obtain the corrected predicted time interval.

[0098] The preheating start time of the deodorization system is determined based on the corrected prediction time interval, and the deodorization system is controlled to start preheating operation according to the preheating start time, so as to realize the advance preheating preparation of the deodorization system.

[0099] By collecting historical data on cat littering, a model of cat littering time patterns can be established. In practical applications, smart litter boxes can be used. These litter boxes are equipped with pressure sensors, infrared sensors, and other devices, which can accurately record the time when the cat enters and leaves the litter box, as well as the duration of each littering session. For example, for a cat named "Xiaohua," data collection over 30 consecutive days recorded its daily littering times. It was found that the cat typically littered during three time periods: 6:00-7:30 AM, 12:00-1:30 PM, and 9:00-10:30 PM, with each session lasting approximately 3-5 minutes.

[0100] Based on historical data, a time series analysis method was used to construct a model of cat littering behavior patterns. This model considers the periodicity, temporal distribution, and environmental influences of cat littering. Specifically, the model analyzes the daily distribution of cat littering behavior, identifying high-frequency littering periods; it also examines the variation patterns of cat littering behavior on different dates, such as differences between weekdays and weekends; and it further analyzes the impact of factors such as cat diet and owner's lifestyle on littering behavior.

[0101] Using a model of cat littering patterns, the timing of the next littering event is predicted. During implementation, the probability distribution of littering events for each time period is calculated based on historical data, with a prediction confidence level of 95%. For example, based on "Xiaohua's" historical data, the predicted littering event for tomorrow morning is around 6:45 AM. At a 95% confidence level, the predicted time interval is 6:15-7:15 AM. This time interval is determined considering both the central trend of historical littering times and the magnitude of time fluctuations.

[0102] To continuously improve prediction accuracy, the prediction error between the predicted time interval and the actual toilet time is calculated, and the predicted time interval is dynamically corrected based on this error. For example, if the actual toilet time is found to be 10 minutes earlier than the predicted time for three consecutive days, the predicted time interval is adjusted forward by 10 minutes; if the toilet time on weekends is found to be 30 minutes later than on weekdays, the predicted time interval is shifted backward accordingly on weekends. Through this adaptive adjustment mechanism, the accuracy of the predicted time interval is significantly improved, with experimental data showing that the prediction accuracy rate increased from the initial 75% to 92%.

[0103] Based on the revised prediction time interval, the preheating start-up time of the deodorization system is determined. In practical applications, considering the time required for the deodorization equipment to reach its optimal operating state from a cold start, it is typically set to 5 minutes. Therefore, the preheating start-up time is set to 5 minutes before the start time of the prediction time interval. For example, if the revised prediction time interval is 6:20-7:20, then the preheating start-up time of the deodorization system is 6:15.

[0104] Based on the predetermined preheating start time, the deodorizing system begins preheating operation. During the preheating phase, the system operates at a lower power to ensure optimal deodorization while the cat is using the litter box, while also conserving energy. The deodorizing power is automatically adjusted based on the cat's actual entry into the litter box; once the cat is inside, the system switches from preheating to full power operation.

[0105] In actual testing, the smart litter box using the above method successfully predicted 92% of cats' littering behaviors during a 30-day testing period. The timely start-up rate of the deodorization system preheating reached 95%, improving deodorization efficiency by 45% compared to traditional methods while reducing energy consumption by 30%. User feedback showed a significant improvement in odor problems in the home environment, with user satisfaction reaching 4.8 / 5.

[0106] Furthermore, it possesses self-learning capabilities, automatically adjusting the predictive model based on long-term changes in a cat's litter box habits. For example, if a cat's litter box habits change due to seasonal variations or aging, the predictive model is updated by analyzing data from the last 30 days, ensuring that predictive accuracy does not decrease over time. Experimental data shows that even when the cat's living environment changes, the updated model still maintains a predictive accuracy rate of over 90%.

[0107] Through the above-mentioned technologies, this invention not only solves the problems of untimely deodorization and energy waste in traditional cat litter boxes, but also improves the user experience through intelligent prediction and dynamic adjustment, providing cats with a more comfortable toilet environment, representing the development direction of intelligent pet products.

[0108] In one optional implementation, based on the cat litter weight change data collected by the weight sensor, the cat litter weight change rate is calculated, and the speed of the deodorizing fan is dynamically adjusted according to the comparison result of the cat litter weight change rate and a preset weight change rate threshold, including:

[0109] The weight sensor is used to collect cat litter weight change data to obtain cat litter weight values ​​at continuous time points; the weight change at adjacent time points is calculated based on the cat litter weight values, and the cat litter weight change rate is calculated based on the weight change and the corresponding time interval;

[0110] The cat litter weight change rate is compared with a preset weight change rate threshold to obtain the deviation value of the weight change rate;

[0111] The target speed of the deodorizing fan is determined based on the deviation value. When the cat litter weight change rate is greater than the preset weight change rate threshold, the target speed of the deodorizing fan is increased. When the cat litter weight change rate is less than the preset weight change rate threshold, the target speed of the deodorizing fan is decreased.

[0112] The current speed of the deodorizing fan is gradually adjusted to the target speed to achieve dynamic adjustment of the deodorizing fan speed.

[0113] like Figure 2 As shown, the method includes:

[0114] The intelligent cat litter box system includes a litter box, a weight sensor, an odor-removing fan, and a control unit. The weight sensor is located at the bottom of the litter box to collect real-time litter weight data. The odor-removing fan is installed on the side wall of the litter box and connected to the exhaust pipe. The control unit is electrically connected to the weight sensor and the odor-removing fan to receive weight data and control the fan speed.

[0115] A weight sensor collects data on the weight changes of the litter box. The sensor measures the weight of the litter box every second and sends the data to the control unit. The control unit records the litter weight values ​​at consecutive time points, forming a time-series data. For example, in a measurement, the weight sensor records the litter weight as 3500 grams at time t1, 3650 grams at time t2 (1 second later), and 3800 grams at time t3 (2 seconds later).

[0116] Based on the collected cat litter weight values, the control unit calculates the weight change at adjacent time points and, combined with the corresponding time intervals, calculates the cat litter weight change rate. The weight change is equal to the weight value at the next moment minus the weight value at the previous moment, and the weight change rate is equal to the weight change divided by the time interval. In the example above, the weight change from time t1 to t2 is 150 grams (3650 grams - 3500 grams), with a time interval of 1 second, therefore the weight change rate is 150 grams / second; the weight change from time t2 to t3 is 150 grams (3800 grams - 3650 grams), with a time interval of 1 second, therefore the weight change rate is 150 grams / second.

[0117] The control unit compares the calculated cat litter weight change rate with a preset weight change rate threshold to obtain the deviation value. The preset weight change rate threshold is determined based on the average amount and frequency of the cat's excretion and is set to 100 grams per second. The deviation value is equal to the cat litter weight change rate minus the weight change rate threshold. In the example above, the deviation value for the time period t1 to t2 is 50 grams per second (150 grams per second - 100 grams per second), and the deviation value for the time period t2 to t3 is 50 grams per second (150 grams per second - 100 grams per second).

[0118] The control unit determines the target speed of the deodorizing fan based on the deviation value. When the cat litter weight change rate is greater than the preset weight change rate threshold, it indicates that the cat is excreting, and the target speed of the deodorizing fan needs to be increased. When the cat litter weight change rate is less than the preset weight change rate threshold, it indicates that the cat's excretion activity has weakened or stopped, and the target speed of the deodorizing fan can be decreased. The target speed is adjusted using a proportional control method, meaning that the increase or decrease in the target speed is directly proportional to the deviation value.

[0119] In this embodiment, the base speed of the deodorizing fan is set to 1000 rpm, and the speed adjustment coefficient is set to 10 (rpm) / (g / s). When the deviation value is positive, the target speed is equal to the base speed plus the deviation value multiplied by the speed adjustment coefficient; when the deviation value is negative, the target speed is equal to the base speed plus the deviation value multiplied by the speed adjustment coefficient, and is not lower than the minimum speed of 500 rpm. In the above example, the deviation value for the time period t1 to t2 is 50 g / s, and the target speed is 1500 rpm (1000 rpm + 50 g / s × 10 (rpm) / (g / s)); the deviation value for the time period t2 to t3 is 50 g / s, and the target speed is maintained at 1500 rpm.

[0120] To avoid startling pets due to sudden changes in the deodorizing fan's speed, the control unit employs a gradual speed adjustment strategy, progressively adjusting the fan's current speed to the target speed. This gradual speed adjustment is achieved by setting a speed change rate of 100 revolutions per minute (rpm). For example, if the current deodorizing fan speed is 1000 rpm and the target speed is 1500 rpm, the control unit will gradually adjust the speed from 1000 rpm to 1500 rpm over 5 seconds, increasing by 100 rpm per second.

[0121] To further optimize the deodorization effect, a speed maintenance time and a recovery time are also set. When the cat litter weight change rate is below the threshold, the deodorizing fan will not immediately reduce its speed, but will maintain a high speed for a period of time (e.g., 30 seconds) to ensure that the odor of the excrement is fully treated. The recovery time refers to the time required for the deodorizing fan to return from high speed to the base speed, which is set to 20 seconds, that is, the speed decreases by 50 revolutions per minute until it returns to the base speed.

[0122] Through the above technical solution, the intelligent cat litter box system can automatically adjust the deodorizing fan speed according to the change rate of cat litter weight, improve deodorizing efficiency during cat excretion, and gradually return to normal after excretion. This not only ensures the deodorizing effect but also reduces energy consumption and avoids fan noise disturbing pets, thus improving the user experience for both users and pets.

[0123] A second aspect of the present invention provides an odor control system for an intelligent cat litter box, comprising:

[0124] The first unit is used to determine whether a cat has entered the smart litter box based on data collected by an infrared sensor. When a cat is detected to have entered the smart litter box, the unit monitors the gas concentration data collected by a gas concentration sensor. When the rate of change of the gas concentration data exceeds a preset rate threshold, the unit controls the deodorization system to start working.

[0125] The second unit is used to predict the next toilet time interval based on the historical usage data of the smart litter box and a preset model of cat toilet time patterns, and to preheat the deodorization system and spray deodorant in advance during the predicted toilet time interval.

[0126] The third unit is used to calculate the cat litter weight change rate based on the cat litter weight change data collected by the weight sensor after the cat leaves the smart litter box, and dynamically adjust the speed of the deodorizing fan according to the comparison result of the cat litter weight change rate and the preset weight change rate threshold.

[0127] A third aspect of the present invention provides an electronic device, comprising:

[0128] processor;

[0129] Memory used to store processor-executable instructions;

[0130] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0131] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0132] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for deodorizing and controlling the odor of an intelligent cat litter box, characterized in that, include: The system determines whether a cat has entered the smart litter box based on data collected by an infrared sensor. When a cat is detected entering the smart litter box, the system monitors the gas concentration data collected by a gas concentration sensor. When the rate of change of the gas concentration data exceeds a preset rate threshold, the system controls the deodorization system to start working. Based on the historical usage data of the smart litter box, the next toilet time interval is predicted according to the preset cat toilet time pattern model. The deodorization system is preheated and deodorant is sprayed in advance during the predicted toilet time interval. After the cat leaves the smart litter box, the cat litter weight change rate is calculated based on the cat litter weight change data collected by the weight sensor, and the deodorizing fan speed is dynamically adjusted according to the comparison result of the cat litter weight change rate with the preset weight change rate threshold. Monitoring gas concentration data collected by a gas concentration sensor, and controlling the deodorization system to start operation when the rate of change of the gas concentration data exceeds a preset rate threshold includes: The data acquisition time interval parameter of the gas concentration sensor installed in the smart litter box is obtained, and the gas concentration sensor is controlled to periodically acquire gas concentration data in the smart litter box according to the data acquisition time interval parameter. A moving average filtering algorithm is used to process gas concentration data from multiple consecutive sampling time points to eliminate data fluctuations during the acquisition process, resulting in smoothed gas concentration data, including: A raw gas concentration data sequence is obtained based on the gas concentration data. The size of the moving average window is determined based on the raw gas concentration data sequence. Weighting coefficients are set for each sampling time point within the moving average window. A weighted average calculation is performed on the gas concentration data within the moving average window based on the weighting coefficients to obtain the smoothed gas concentration value at the current time. The moving average window is then moved forward one sampling time point at a time, and the weighted average calculation is repeated until the gas concentration data of all sampling time points have been processed. The weighting coefficients of the starting and ending segments of the moving average window are corrected to obtain the smoothed gas concentration data. The rate of change of the smoothed gas concentration data is calculated in real time. When the rate of change of the gas concentration data exceeds a preset rate threshold, a deodorization system start signal is generated. Based on the deodorization system start signal, the deodorization system is controlled to enter the working state.

2. The method according to claim 1, characterized in that, Based on the cat's littering time pattern model, the system predicts the next littering time interval and preheats the deodorizing system and sprays deodorant during the predicted littering time interval, including: Collect the time points and durations of each littering event of a cat to obtain historical data containing time point sequences and duration sequences; The historical data is preprocessed by standardization. The time series is subtracted from its mean and then divided by its standard deviation to obtain a standardized time series. The standardized time series is then subjected to time series analysis to extract the temporal regularity features of cat toileting behavior and to establish a cat toileting time regularity model. Based on the cat's toilet time pattern model, the time interval of the next toilet event is predicted to obtain the predicted time interval; the preheating start time of the deodorization system is determined according to the predicted time interval. The deodorization system is started at the preheating start time to preheat, and the spraying amount of deodorant is adaptively controlled according to the predicted duration within the predicted time interval.

3. The method according to claim 2, characterized in that, Based on the cat's littering time pattern model, the time interval of the next littering event is predicted to obtain the predicted time interval; the preheating start time of the deodorization system is determined according to the predicted time interval, including: The cat's toilet time pattern model is used to predict the time of the next toilet event. By determining the start and end times of the prediction time interval under a preset confidence level, the complete prediction time interval is obtained. Calculate the prediction error between the complete predicted time interval and the actual toilet time, and dynamically correct the complete predicted time interval based on the prediction error to obtain the corrected predicted time interval. The preheating start time of the deodorization system is determined based on the corrected prediction time interval, and the deodorization system is controlled to start preheating operation according to the preheating start time, so as to realize the advance preheating preparation of the deodorization system.

4. The method according to claim 1, characterized in that, Based on the cat litter weight change data collected by the weight sensor, the cat litter weight change rate is calculated, and the deodorizing fan speed is dynamically adjusted according to the comparison result of the cat litter weight change rate and the preset weight change rate threshold, including: The weight sensor is used to collect cat litter weight change data to obtain cat litter weight values ​​at continuous time points; the weight change at adjacent time points is calculated based on the cat litter weight values, and the cat litter weight change rate is calculated based on the weight change and the corresponding time interval; The cat litter weight change rate is compared with a preset weight change rate threshold to obtain the deviation value of the weight change rate; The target speed of the deodorizing fan is determined based on the deviation value. When the cat litter weight change rate is greater than the preset weight change rate threshold, the target speed of the deodorizing fan is increased. When the cat litter weight change rate is less than the preset weight change rate threshold, the target speed of the deodorizing fan is decreased. The current speed of the deodorizing fan is gradually adjusted to the target speed to achieve dynamic adjustment of the deodorizing fan speed.

5. An odor control system for an intelligent cat litter box, used to implement the method as described in any one of claims 1-4, characterized in that, include: The first unit is used to determine whether a cat has entered the smart litter box based on data collected by an infrared sensor. When a cat is detected to have entered the smart litter box, the unit monitors the gas concentration data collected by a gas concentration sensor. When the rate of change of the gas concentration data exceeds a preset rate threshold, the unit controls the deodorization system to start working. The second unit is used to predict the next toilet time interval based on the historical usage data of the smart litter box and a preset model of cat toilet time patterns, and to preheat the deodorization system and spray deodorant in advance during the predicted toilet time interval. The third unit is used to calculate the cat litter weight change rate based on the cat litter weight change data collected by the weight sensor after the cat leaves the smart litter box, and dynamically adjust the speed of the deodorizing fan according to the comparison result of the cat litter weight change rate and the preset weight change rate threshold.

6. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 4.

Images