Pet excretion need estimation method, electronic device, and computer program product

Abstract

The application discloses a pet excretion demand estimation method, an electronic device and a computer program product. The method comprises the following steps: acquiring primary determination data, and further acquiring secondary determination data when it is determined that the primary determination data meets a first condition; one group of the primary determination data and the secondary determination data is pet excretion rule data, and the other group is pet real-time state data; the first condition is a determination condition for determining whether the pet has a potential excretion demand; and the primary determination data and the secondary determination data are input into an excretion estimation model to obtain an excretion estimation time output by the model. Through a hierarchical triggering mechanism, step-by-step collection and model fusion are realized, long-term rhythm can be grasped by relying on pet excretion rule data, and instant demand can be captured in combination with pet real-time state data; the two types of data are mutually verified and complementary to each other, the accuracy of excretion time estimation is improved, and invalid data processing and waste of computing power are reduced.

Description

Technical Field

[0001] This application relates to the field of pet feeding technology, and more specifically, to a method for predicting pet excretion needs, electronic devices, and computer program products. Background Technology

[0002] Pets are animals kept by people for emotional or psychological reasons, not for economic purposes. Pet excretion is a persistent headache for owners. Owners typically want their pets to relieve themselves in designated areas or using provided containers within the home, or outdoors during outdoor activities. However, more often than not, this is uncontrollable, with pets frequently defecating and urinating indiscriminately throughout the house, directly impacting the home environment. Summary of the Invention

[0003] In view of the above problems, this application is made to provide a method, electronic device, and computer program product for predicting pet elimination needs, so as to predict the pet's elimination time, allowing owners to promptly arrange for their pets to eliminate in designated areas and maintain a clean and hygienic home environment. The specific solution is as follows:

[0004] Firstly, a method for predicting pet elimination needs is provided, including:

[0005] Obtain primary determination data, and if the primary determination data satisfies the configured first condition, further obtain secondary determination data;

[0006] Among them, one set of the primary judgment data and the secondary judgment data is pet excretion pattern data, and the other set is pet real-time status data. The first condition is a judgment condition used to determine whether a pet has a potential excretion need.

[0007] The primary and secondary judgment data are input into the configured excretion prediction model to obtain the excretion prediction time output by the model.

[0008] In one possible design, in another implementation of the first aspect of the embodiments of this application, the primary determination data is pet excretion pattern data, which includes historical excretion data and dietary data;

[0009] The secondary judgment data is real-time status data of the pet, which includes physiological state change data and behavioral state data.

[0010] In one possible design, in another implementation of the first aspect of the embodiments of this application, the first condition includes:

[0011] Given that the dietary data represents a normal diet, the current time falls within the predicted excretion period determined based on the historical excretion data;

[0012] When the dietary data indicates overeating, the difference between the current time and the predicted excretion period is within a first preset time threshold range;

[0013] When the dietary data indicates a low-volume diet, the difference between the current time and the predicted excretion period falls within the second preset time threshold range.

[0014] In one possible design, in another implementation of the first aspect of the embodiments of this application, the primary determination data is real-time pet status data, which includes physiological state change data and behavioral state data.

[0015] The secondary judgment data is pet excretion pattern data, which includes historical excretion data and dietary data;

[0016] Then, the first condition includes:

[0017] The physiological state change data reaches the preset physiological change level, and the behavioral state data shows typical behavioral characteristics related to excretion needs.

[0018] In one possible design, in another implementation of the first aspect of the embodiments of this application, the primary determination data is excretion pattern data, and the secondary determination data is pet real-time status data;

[0019] Before inputting the primary judgment data and secondary judgment data into the configured excretion prediction model, the method further includes: adaptive weight allocation of the primary judgment data and the secondary judgment data;

[0020] The process of inputting the primary and secondary judgment data into the configured excretion prediction model includes:

[0021] The primary and secondary judgment data are fused according to their respective weights, and the excretion prediction model is used to predict the pet's excretion time based on the fused data.

[0022] The adaptive weight allocation process includes:

[0023] If the primary judgment data is within the normal historical fluctuation range, the fusion weight of the primary judgment data is increased, and the weight of the secondary judgment data is reduced accordingly.

[0024] If the primary judgment data exceeds the historical normal fluctuation range, the fusion weight of the primary judgment data is reduced, and the weight of the secondary judgment data is increased accordingly.

[0025] In one possible design, in another implementation of the first aspect of the embodiments of this application, the primary determination data is pet real-time status data, and the secondary determination data is excretion pattern data.

[0026] Before inputting the primary judgment data and secondary judgment data into the configured excretion prediction model, the method further includes: adaptive weight allocation of the primary judgment data and the secondary judgment data;

[0027] The process of inputting the primary and secondary judgment data into the configured excretion prediction model includes:

[0028] The primary and secondary judgment data are fused according to their respective weights, and the excretion prediction model is used to predict the pet's excretion time based on the fused data.

[0029] The adaptive weight allocation process includes:

[0030] If the primary judgment data exhibits continuous and stable pre-excretion symptoms, the fusion weight of the primary judgment data is increased, and the weight of the secondary judgment data is correspondingly decreased.

[0031] If the primary judgment data exhibits transient and sporadic characteristics, the fusion weight of the primary judgment data is reduced, and the weight of the secondary judgment data is increased accordingly.

[0032] In one possible design, another implementation of the first aspect of the embodiments of this application further includes:

[0033] Determine the stability of a pet's excretion patterns within a preset historical period;

[0034] If the stability is higher than the set stability threshold, the primary judgment data is set as pet excretion rule data, and the secondary judgment data is set as pet real-time status data.

[0035] If the stability is not higher than the set stability threshold, the primary judgment data is set as pet real-time status data, and the secondary judgment data is set as pet excretion rule data.

[0036] In one possible design, another implementation of the first aspect of the embodiments of this application further includes:

[0037] When the interval between the current time and the estimated excretion time is less than or equal to a preset first threshold, an excretion prompt message is output.

[0038] In a second aspect, an electronic device is provided, comprising: a memory and a processor;

[0039] The memory is used to store programs;

[0040] The processor is configured to execute the program to implement the steps of the pet excretion demand estimation method described in any of the first aspects of this application.

[0041] Thirdly, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the pet excretion demand prediction method described in any of the first aspects of this application.

[0042] By employing the above technical solution, this application divides the reference data for excretion prediction into two complementary dimensions: historical pattern data and real-time status data. Through a hierarchical triggering mechanism, step-by-step data collection and model fusion are achieved. This approach can grasp long-term rhythms based on pet excretion pattern data and capture immediate needs by combining pet real-time status data. The two types of data mutually verify each other and complement each other's advantages, effectively overcoming the problems of large prediction deviations and weak anti-interference capabilities of single-dimensional data. This significantly improves the accuracy, stability, and scenario adaptability of excretion time prediction.

[0043] Furthermore, this application does not collect all dimensions of reference data at once. Instead, it uses a tiered triggering mechanism to first acquire primary judgment data. Only when the primary judgment data meets the first condition is the collection of secondary judgment data and model calculation initiated. For the majority of the remaining time, it remains in a low-power monitoring state, significantly reducing invalid data processing and wasted computing power, making it particularly suitable for low-power scenarios such as pet wearable devices. In addition, schemes that collect all data at once introduce a large amount of interfering data unrelated to the current excretion needs. This scheme, through a tiered triggering mechanism, only introduces supplementary data (secondary judgment data) when the preconditions are met. This is equivalent to performing a pre-screening of the data, filtering out invalid features, making the model input more focused and the prediction more stable. Attached Figure Description

[0044] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0045] Figure 1 A schematic diagram of an implementation system architecture for the pet excretion demand estimation method provided in this application embodiment;

[0046] Figure 2 This is a schematic flowchart of a pet excretion demand estimation method provided in an embodiment of this application;

[0047] Figure 3 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] This application provides a method for predicting pet excretion needs, which can be applied to, for example... Figure 1 The system architecture shown can include a pet collar 100 and a terminal device 200.

[0050] The pet collar 100 can be used alone to execute the pet elimination demand estimation method provided in the embodiments of this application. Alternatively, the pet collar 100 and the terminal device 200 can also be used collaboratively to execute the pet elimination demand estimation method provided in the embodiments of this application.

[0051] The pet collar 100 and the terminal device 200 can communicate. For example, the pet collar 100 can send the collected real-time status data of the pet to the terminal device 200, and the terminal device 200 can run the pet's excretion demand prediction method to obtain the excretion prediction time.

[0052] Alternatively, the pet collar 100 can acquire pet elimination pattern data recorded by the pet owner and sent by the terminal device 200. The pet collar 100 then runs a pet elimination need prediction method to obtain the estimated elimination time. Furthermore, the estimated elimination time can be output to the terminal device 200.

[0053] Terminal device 200 can be a smart home device or a user terminal, such as a mobile phone or tablet computer.

[0054] The pet collar 100 can be equipped with multimodal sensors to collect real-time status data of the pet. These multimodal sensors include, but are not limited to, physiological state sensors (such as heart rate sensors, pressure sensors, etc.) and motion sensors (such as a six-axis sensor). The real-time status data collected by the multimodal sensors can include physiological state change data and behavioral state data.

[0055] This application provides a method for predicting pet excretion needs. Taking the application of this method to a computer device as an example, the computer device can specifically be... Figure 1 The system comprises a pet collar 100 or a pet collar 100 and a terminal device 200. (Refer to...) Figure 2 The method for predicting a pet's elimination needs specifically includes the following steps:

[0056] Step S100: Obtain preliminary judgment data.

[0057] Specifically, this application can divide the reference data for excretion prediction into two types of data: one is pet excretion pattern data, and the other is pet real-time status data.

[0058] Optionally, pet excretion pattern data may include historical excretion data and dietary data.

[0059] Historical excretion data can include historical excretion periods and the excretion frequency for each period. The historical excretion period is the time interval within which the pet habitually exhibits excretion behavior during a historical cycle, and the excretion frequency is the cumulative number of times the pet defecates and urinates within that time interval. Excretion behavior can refer to both defecation and urination.

[0060] Dietary data can include the pet's eating times within a historical period, as well as the amount of food consumed at each time point. The historical period can be user-defined or a system default setting, such as the past day. The dietary data can refer to both eating and drinking data.

[0061] Optionally, real-time pet status data may include physiological status change data and behavioral status data.

[0062] Physiological state change data refers to data on changes in physiological parameters of a pet's excretory system, such as changes in heart rate and abdominal pressure. This data can be acquired through various sensors attached to the pet's collar. A pet's need to excrete is often accompanied by slight stress-related fluctuations in heart rate, abdominal hardening, and changes in bladder volume. Therefore, physiological state change data can serve as a reference for analyzing whether a pet has an urge to excrete.

[0063] Behavioral status data can include a pet's physical and vocal behaviors, such as circling, sniffing, crouching, arching its back, and lifting its leg. This data can be identified using motion sensors and microphones attached to the pet's collar. The behavioral status data can be a sequence of behaviors, such as a sequence of various pet behaviors over a period of time. For example, if a dog circles, sniffs, and then lifts its leg, it is highly likely that the pet needs to relieve itself. Therefore, a pet's behavioral status data can serve as a reference factor in determining whether a pet needs to relieve itself.

[0064] To avoid the waste of computing power and data interference that would result from collecting all reference data at once, this application designs a tiered triggering mechanism. Accordingly, either of the two types of reference data (pet excretion pattern data and pet real-time status data) can be set as primary judgment data, and the other can be set as secondary judgment data.

[0065] For example, the primary judgment data is pet excretion pattern data, and the corresponding secondary judgment data is pet real-time status data. Alternatively, the primary judgment data is pet real-time status data, and the corresponding secondary judgment data is pet excretion pattern data.

[0066] This application can acquire preliminary judgment data in real time or periodically.

[0067] Step S110: Determine whether the initial judgment data meets the first condition of the configuration.

[0068] The first condition is used to determine whether a pet has a potential need to eliminate waste. This first condition is compatible with the type of data used in the initial determination.

[0069] If the initial judgment data meets the first condition, it indicates that the pet has a potential need to defecate, thus triggering further execution step S120.

[0070] Optionally, if the initial judgment data does not meet the configured first condition, it indicates that the pet currently does not have a need to relieve itself, and therefore the process can return to step S100 to obtain new initial judgment data. Of course, after determining in step S110 that the first condition is not met, the process can also return to step S100 after a certain period of time.

[0071] Step S120: Obtain secondary judgment data.

[0072] If, in step S110 above, the initial judgment data meets the first condition, it indicates that the pet has a potential need to defecate. At this point, secondary judgment data is further obtained to obtain more comprehensive reference data. This data is used to provide more comprehensive and complete reference data for the next step of model prediction.

[0073] Step S130: Input the primary judgment data and secondary judgment data into the configured excretion prediction model to obtain the excretion prediction time output by the model.

[0074] The excretion prediction model is configured to predict the estimated time of a pet's excretion based on the input data.

[0075] In some possible implementations, the excretion prediction model can employ a deep neural network model, including but not limited to convolutional neural networks, recurrent neural networks, and fully connected neural networks. This model is trained using a training dataset containing multiple sets of sample data. Each set of sample data includes excretion pattern data and real-time status data of the sample pet, along with the corresponding actual excretion time label.

[0076] In other possible implementations, the excretion prediction model can also be implemented using a large-scale model architecture. Leveraging the feature understanding and time-series fitting capabilities of large models, it performs comprehensive reasoning on both regular and state-related data, further improving the accuracy and generalization ability of excretion prediction time. The model is trained offline before use and can be iteratively optimized online based on user-annotated real excretion data during actual use to adapt to the individual excretion habits of different pets.

[0077] The pet excretion demand prediction method provided in this application divides the reference data for excretion prediction into two complementary dimensions: historical pattern data and real-time status data. Through a hierarchical triggering mechanism, it achieves step-by-step data collection and model fusion. It can grasp long-term rhythms based on pet excretion pattern data and capture immediate needs by combining pet real-time status data. The two types of data are mutually verified and complementary, effectively overcoming the problems of large prediction deviation and weak anti-interference ability of single-dimensional data, and significantly improving the accuracy, stability and scenario adaptability of excretion time prediction.

[0078] Furthermore, this application does not collect all dimensions of reference data at once. Instead, it uses a tiered triggering mechanism to first acquire primary judgment data. Only when the primary judgment data meets the first condition is the collection of secondary judgment data and model calculation initiated. For the majority of the remaining time, it remains in a low-power monitoring state, significantly reducing invalid data processing and wasted computing power, making it particularly suitable for low-power scenarios such as pet wearable devices. In addition, schemes that collect all data at once introduce a large amount of interfering data unrelated to the current excretion needs. This scheme, through a tiered triggering mechanism, only introduces supplementary data (secondary judgment data) when the preconditions are met. This is equivalent to performing a pre-screening of the data, filtering out invalid features, making the model input more focused and the prediction more stable.

[0079] In some embodiments of this application, primary judgment data is defined as pet excretion pattern data, which includes historical excretion data and diet data; secondary judgment data is pet real-time status data, which includes physiological state change data and behavioral state data.

[0080] Based on the above data settings, this embodiment provides an optional judgment method for determining whether the initial judgment data meets the configured first condition. This first condition combines dietary conditions and time offset for comprehensive judgment, as follows:

[0081] 1. If the pet's diet data indicates that it is eating normally, determine whether the current time falls within the predicted excretion period determined based on historical excretion data; if it does, then the first condition is met.

[0082] 2. If the dietary data indicates that the pet is overeating, determine whether the difference between the current time and the predicted excretion time is within the first preset time threshold range; if it is within the first preset time threshold range, then the first condition is met.

[0083] 3. If the dietary data indicates that the pet is eating a low amount of food, determine whether the delay difference between the current time and the predicted excretion time is within the second preset time threshold range; if it is within the second preset time threshold range, then the first condition is met.

[0084] The first preset time threshold interval and the second pre-review time threshold interval can be the same or different.

[0085] In this embodiment, whether the pet's food intake is considered normal, excessive, or insufficient can be determined based on user-recorded dietary data and a relative comparison with the pet's historical typical dietary levels. One possible implementation is as follows:

[0086] The system pre-calculates the pet's normal dietary reference range based on the food and water intake data manually entered by the user within a preset historical period (such as the last 7 days or the last 15 days). This reference range can be reflected as the normal fluctuation range of the historical average daily food intake and water intake.

[0087] After the current dietary data is entered, compare the actual food intake and water intake with the above-mentioned general dietary reference range:

[0088] If the amount of food consumed this time is within the reference range for a normal diet, it is considered a normal diet.

[0089] If the amount of food consumed exceeds the upper limit of the normal dietary reference range, and the excess reaches a preset proportion or value, it is judged as excessive eating.

[0090] If the amount of food consumed is lower than the lower limit of the normal dietary reference range, and the difference is within a preset proportion or value, it is judged as a low-volume diet.

[0091] Furthermore, a specific example will be used to illustrate the judgment process of the first condition.

[0092] If determined based on a pet's historical excretion data, the pet's usual predicted excretion time is between 8:00 and 9:00 AM daily.

[0093] If the pet's food and water intake are within the normal range for the day, i.e., normal diet, then the current time is between 8:00 and 9:00, which is considered to meet the first condition.

[0094] If a pet's daily food and water intake is significantly higher than normal (overeating), its need to eliminate will usually occur earlier. Therefore, a range of 0-30 minutes is allowed for this earlier elimination, corresponding to the actual judgment period of 7:30-8:00. If the current time falls within this time period, the first condition is considered met.

[0095] If the pet has consumed a small amount of food that day, excretion will usually be delayed. Therefore, the allowable delay range is set to 0-30 minutes, corresponding to the actual judgment period of 9:00-9:30. If the current time falls within this extended time period, the first condition is considered met.

[0096] This embodiment dynamically adjusts the predicted excretion time by incorporating dietary status. Compared to methods that rely solely on fixed historical time periods, this approach more closely aligns with the pet's actual physiological rhythms and improves the accuracy of initial assessments. Specifically, the timing of a pet's excretion is significantly affected by the amount of food consumed; overeating tends to cause premature excretion, while undereating tends to cause delayed excretion. By distinguishing between normal, overeating, and undereating dietary statuses and setting corresponding time-based assessment rules, the influence of diet on excretion rhythms can be more realistically reflected, preventing the initial assessment from failing due to dietary fluctuations.

[0097] Furthermore, considering the differences in dietary habits among different pets, the first preset time threshold range and the second preset time threshold range can be flexibly configured, enabling the overall prediction scheme to better adapt to the individual characteristics of different pets and improve the generalization ability of the scheme.

[0098] In some other embodiments of this application, the primary judgment data is defined as real-time pet status data, which includes physiological state change data and behavioral state data; the secondary judgment data is pet excretion pattern data, which includes historical excretion data and dietary data.

[0099] Based on the above data settings, this embodiment provides an optional judgment method for determining whether the initial judgment data meets the configured first condition. This first condition combines physiological changes and behavioral characteristics for comprehensive judgment, as follows:

[0100] The physiological state change data reached the preset physiological change level, and the behavioral state data showed typical behavioral characteristics related to excretion needs.

[0101] The physiological state change data can be obtained from information such as abdominal pressure, body activity intensity, and posture changes collected by wearable devices on the pet. The preset physiological change level refers to the change in the pet's excretion-related physiological parameters exceeding a preset baseline value per unit time. These physiological parameters include, but are not limited to, abdominal pressure, bioimpedance, and body activity intensity. The preset physiological change level can be pre-configured according to the pet's body size and breed.

[0102] Typical behavioral characteristics related to excretion needs include, but are not limited to: circling in place, frequently sniffing the ground, restlessness, frequently looking back at the tail, and attempting to go to a fixed excretion area, which reflect the intention to excrete.

[0103] Furthermore, a specific example will be used to illustrate the judgment process of the first condition.

[0104] When the wearable device detects that the pet's abdominal pressure continues to rise and exceeds the preset physiological threshold, and at the same time, it identifies typical excretion behaviors such as circling and sniffing the ground continuously within the preset time window, and the duration reaches the preset duration, then it is determined that the physiological state change data has reached the preset physiological change level, and the behavioral state data shows typical excretion-related behavioral characteristics, thus satisfying the first condition.

[0105] If only the data on changes in physiological state reach the preset level of physiological change, but no typical excretory behavior is identified, then the first condition is not met; if only brief behavioral actions occur, but there is no obvious corresponding change in physiological state, then the first condition is also not met.

[0106] This embodiment, by combining physiological state changes with behavioral characteristics, significantly improves the reliability of initial judgment and reduces false triggers compared to methods that rely solely on a single behavior or physiological data point. Specifically, pets may exhibit defecation-like behaviors due to play or stress, but these are usually not accompanied by corresponding physiological state changes; they may also experience brief physiological fluctuations due to movement or changes in body position, without a genuine intention to defecate. By setting a combined judgment condition of "physiological change + typical behavior," it is possible to effectively filter out incidental behavioral interference and physiological noise, and more accurately identify genuine defecation needs.

[0107] Furthermore, considering the differences in behavioral habits and physiological characteristics of different pets, the preset physiological change levels and typical behavior identification standards can be flexibly configured, enabling the overall prediction plan to better adapt to different individuals and improve the applicability and robustness of the plan.

[0108] In some embodiments of this application, after determining the primary and secondary judgment data and before inputting them into the excretion prediction model, an adaptive weight allocation scheme for the two types of data is provided. By dynamically assigning fusion weights to data of different dimensions, the influence of each dimension in the model prediction can be adjusted according to the credibility of the current data, thereby making the output of the excretion prediction model more consistent with the pet's actual physiological state and improving prediction accuracy and robustness. The assigned weights characterize the importance of the corresponding data in the feature fusion and prediction process; a higher weight indicates a greater influence of that dimension of data on the final excretion prediction time.

[0109] The process of adaptively weighting the primary and secondary judgment data and then inputting them into the configured excretion prediction model may include:

[0110] The primary and secondary judgment data are fused according to their respective weights, and the excretion prediction model is used to predict the estimated time of excretion of pets based on the fused data.

[0111] In practical applications, the specific implementation strategy of adaptive weight allocation varies depending on the types of primary and secondary decision data. The following sections explain the two data combination methods respectively.

[0112] In this embodiment, the primary judgment data is excretion pattern data, and the secondary judgment data is pet real-time status data.

[0113] Before inputting the primary and secondary judgment data into the configured excretion prediction model, adaptive weight allocation is first performed on the two types of data.

[0114] The specific adaptive weight allocation process may include:

[0115] If the primary judgment data is within the normal historical fluctuation range, it indicates that the pet's excretion pattern is stable and highly reliable. Therefore, the fusion weight of the primary judgment data is increased, and the weight of the secondary judgment data is reduced accordingly.

[0116] If the primary judgment data exceeds the historical normal fluctuation range, it indicates that the reliability of the pattern is reduced due to factors such as diet and rest. In this case, the fusion weight of the primary judgment data is reduced, and the weight of the secondary judgment data is increased accordingly.

[0117] After that, the primary judgment data and secondary judgment data are fused according to their respective weights, and the weighted fused data is input into the excretion prediction model. The model then predicts the pet's excretion time based on the fused features.

[0118] In another embodiment, the primary judgment data is pet real-time status data, and the secondary judgment data is excretion pattern data.

[0119] Before inputting the primary and secondary judgment data into the excretion prediction model, adaptive weight allocation is also performed on the two types of data.

[0120] The specific adaptive weight allocation process includes:

[0121] If the primary judgment data shows continuous and stable pre-excretion symptoms, indicating that the current physiological and behavioral characteristics have high reliability, then the fusion weight of the primary judgment data is increased, and the weight of the secondary judgment data is reduced accordingly.

[0122] If the primary judgment data only exhibits transient and sporadic characteristics, is easily interfered with, and has low reliability, then the fusion weight of the primary judgment data is reduced, and the weight of the secondary judgment data is increased accordingly.

[0123] Subsequently, the primary and secondary judgment data are fused according to their respective weights, and the fused data is input into the excretion prediction model to achieve a more accurate prediction of excretion time.

[0124] The method described in the above embodiments, which adaptively assigns weights to the two types of data before inputting the primary and secondary judgment data into the excretion prediction model, has the following advantages:

[0125] 1. Improve prediction accuracy and better reflect the pet's actual condition.

[0126] When patterns are stable, the model relies more on pattern data; when patterns are abnormal, it relies more on real-time status. When behavioral intentions are clear, it focuses more on real-time status; when behavior is ambiguous, it relies more on pattern data. This ensures that the model predictions are always biased towards data dimensions with higher credibility, effectively improving the accuracy of excretion prediction time.

[0127] 2. Enhance the anti-interference capability of the solution and reduce the probability of misjudgment.

[0128] It can automatically filter unreliable information such as dietary fluctuations, random behaviors, and instantaneous noise, avoiding the significant impact of single-dimensional abnormal fluctuations on the overall prediction results and improving the robustness of the system.

[0129] 3. Enhance the generalizability and individual adaptability of the solution.

[0130] There is no need to set fixed experience parameters for different pets. Through adaptive weight allocation, it can automatically adapt to the different pets' excretion habits, behavioral characteristics and physiological features, and maintain a stable and reliable prediction effect in a variety of scenarios.

[0131] In some embodiments of this application, in order to better adapt the excretion prediction scheme to the individual habits of different pets and further improve the prediction accuracy and scenario adaptability, an implementation method is also provided that adaptively configures the primary judgment data and secondary judgment data types based on the stability of the pet's historical excretion patterns.

[0132] The specific implementation process is as follows:

[0133] First, determine the stability of the pet's excretion patterns within a preset historical period.

[0134] The preset historical period can be a time period configured by the system or defined by the user, such as the most recent 7 days or the most recent 15 days. Based on the excretion time data recorded within this historical period, the system performs statistical analysis on the concentration and regularity of excretion periods to obtain the corresponding excretion pattern stability. The higher the stability, the more fixed the pet's excretion time and the more stable the rhythm; the lower the stability, the greater the fluctuation in the pet's excretion time and the poorer the regularity.

[0135] After obtaining the stability of the excretion pattern, it is compared with the set stability threshold, and the types of primary and secondary judgment data are adaptively configured according to the comparison result:

[0136] If the stability is higher than the set stability threshold, it indicates that the pet's excretion rhythm is stable and regular. A relatively reliable preliminary judgment can be made based on historical patterns. In this case, the primary judgment data is set as excretion pattern data, and the secondary judgment data is set as real-time status data, thus entering the pattern-first judgment process.

[0137] If the stability is not higher than the set stability threshold, it indicates that the pet's excretion habits fluctuate greatly and have weak regularity. Relying solely on historical patterns is prone to significant deviations. In this case, the primary judgment data is set as real-time status data, and the secondary judgment data is set as excretion pattern data, thus entering a status-priority judgment process.

[0138] Through the above methods, the system can automatically select a more reasonable judgment path based on the pet's own excretion habits: when the pattern is stable, the pattern is the core and the status is the secondary factor; when the pattern is unstable, the real-time status is the core and the pattern is the secondary factor, thus maintaining high judgment accuracy and robustness for different individual pets.

[0139] By employing the adaptive configuration scheme for primary and secondary judgment data types provided in this embodiment, the system automatically determines stability and switches modes based on historical excretion data, achieving individual pet self-adaptation without manual configuration. For pets with stable patterns, the system fully leverages their habitual advantages; for pets with irregular patterns, it prioritizes real-time status to avoid inherent biases caused by fixed patterns.

[0140] In some embodiments of this application, after obtaining the estimated excretion time output by the excretion prediction model, the following steps may be further performed:

[0141] When the interval between the current time and the estimated excretion time is less than or equal to a preset first threshold, an excretion prompt message is output.

[0142] Specifically, the time interval between the current time and the estimated time of excretion is calculated. When the time interval is less than or equal to a preset first threshold, it is determined that the excretion period is about to arrive, and the corresponding excretion prompt information is output, such as issuing a voice or other form of prompt information through the pet collar, or sending a prompt information to the terminal device associated with the pet collar to remind the user to guide the pet to excrete in time.

[0143] The preset first threshold is the advance reminder duration, which can be set to 5 minutes, 10 minutes, etc., and can be configured by the system default or customized by the user according to their usage habits.

[0144] This application also provides an electronic device in its embodiments. (See reference...) Figure 3 The diagram illustrates a structural schematic suitable for implementing the electronic device in the embodiments of this application. The electronic device in the embodiments of this application may include, but is not limited to, fixed terminals such as pet collars, pet tokens, wearable devices, etc. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0145] like Figure 3 As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 1, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 2 or a program loaded from a storage device 8 into a random access memory (RAM) 3, to implement the pet excretion demand estimation method of the foregoing embodiments of this application. When the electronic device is powered on, the RAM 3 also stores various programs and data required for the operation of the electronic device. The processing unit 1, ROM 2, and RAM 3 are interconnected via a bus 4. An input / output (I / O) interface 5 is also connected to the bus 4.

[0146] Typically, the following devices can be connected to I / O interface 5: input devices 6 including, for example, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 7 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 8 including, for example, memory cards, hard drives, etc.; and communication devices 9. Communication device 9 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0147] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement the various steps of any of the pet excretion demand prediction methods provided in this application.

[0148] This application also provides a computer-readable storage medium carrying one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the various steps of any of the pet excretion demand estimation methods provided in this application.

[0149] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0150] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0151] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0152] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0153] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

Claims

1. A method for predicting pet excretion needs, characterized in that, include: Obtain primary determination data, and if the primary determination data satisfies the configured first condition, further obtain secondary determination data; Among them, one set of the primary judgment data and the secondary judgment data is pet excretion pattern data, and the other set is pet real-time status data. The first condition is a judgment condition used to determine whether a pet has a potential excretion need. The primary and secondary judgment data are input into the configured excretion prediction model to obtain the excretion prediction time output by the model.

2. The method according to claim 1, characterized in that, The initial judgment data is pet excretion pattern data, which includes historical excretion data and dietary data; The secondary judgment data is real-time status data of the pet, which includes physiological state change data and behavioral state data.

3. The method according to claim 2, characterized in that, The first condition includes: Given that the dietary data represents a normal diet, the current time falls within the predicted excretion period determined based on the historical excretion data; When the dietary data indicates overeating, the difference between the current time and the predicted excretion period is within a first preset time threshold range; When the dietary data indicates a low-volume diet, the difference between the current time and the predicted excretion period falls within the second preset time threshold range.

4. The method according to claim 1, characterized in that, The initial judgment data is real-time status data of the pet, which includes physiological state change data and behavioral state data; The secondary judgment data is pet excretion pattern data, which includes historical excretion data and dietary data; Then, the first condition includes: The physiological state change data reaches the preset physiological change level, and the behavioral state data shows typical behavioral characteristics related to excretion needs.

5. The method according to claim 1, characterized in that, The primary judgment data is excretion pattern data, and the secondary judgment data is real-time status data of the pet; Before inputting the primary judgment data and secondary judgment data into the configured excretion prediction model, the method further includes: adaptive weight allocation of the primary judgment data and the secondary judgment data; The process of inputting the primary and secondary judgment data into the configured excretion prediction model includes: The primary and secondary judgment data are fused according to their respective weights, and the excretion prediction model is used to predict the pet's excretion time based on the fused data. The adaptive weight allocation process includes: If the primary judgment data is within the normal historical fluctuation range, the fusion weight of the primary judgment data is increased, and the weight of the secondary judgment data is reduced accordingly. If the primary judgment data exceeds the historical normal fluctuation range, the fusion weight of the primary judgment data is reduced, and the weight of the secondary judgment data is increased accordingly.

6. The method according to claim 1, characterized in that, The primary judgment data is pet real-time status data, and the secondary judgment data is excretion pattern data. Before inputting the primary judgment data and secondary judgment data into the configured excretion prediction model, the method further includes: adaptive weight allocation of the primary judgment data and the secondary judgment data; The process of inputting the primary and secondary judgment data into the configured excretion prediction model includes: The primary and secondary judgment data are fused according to their respective weights, and the excretion prediction model is used to predict the pet's excretion time based on the fused data. The adaptive weight allocation process includes: If the primary judgment data exhibits continuous and stable pre-excretion symptoms, the fusion weight of the primary judgment data is increased, and the weight of the secondary judgment data is correspondingly decreased. If the primary judgment data exhibits transient and sporadic characteristics, the fusion weight of the primary judgment data is reduced, and the weight of the secondary judgment data is increased accordingly.

7. The method according to claim 1, characterized in that, Also includes: Determine the stability of a pet's excretion patterns within a preset historical period; If the stability is higher than the set stability threshold, the primary judgment data is set as pet excretion rule data, and the secondary judgment data is set as pet real-time status data. If the stability is not higher than the set stability threshold, the primary judgment data is set as pet real-time status data, and the secondary judgment data is set as pet excretion rule data.

8. The method according to any one of claims 1-7, characterized in that, Also includes: When the interval between the current time and the estimated excretion time is less than or equal to a preset first threshold, an excretion prompt message is output.

9. An electronic device, characterized in that, include: Memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the pet excretion demand prediction method as described in any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the pet excretion demand prediction method as described in any one of claims 1 to 8.

Images