Pet intelligent induction necklace and control method thereof

Abstract

The embodiment of the present specification provides a control method of a pet intelligent sensing necklace. The pet intelligent sensing necklace at least comprises a sound detection unit, a positioning detection unit, an activity detection unit, a physiological detection unit, a processor and a light unit. The method comprises: obtaining sound information of a target pet based on the sound detection unit; obtaining positioning information of the target pet based on the positioning detection unit; obtaining activity information of the target pet based on the activity detection unit; obtaining physiological information of the target pet based on the physiological detection unit; determining behavior information of the target pet based on the sound information, the activity information and the physiological information; processing the positioning information and the behavior information based on a demand model to determine a demand type of the target pet; and controlling the light unit to display the demand type.

Description

Technical Field

[0001] This manual relates to the field of pet supplies technology, and in particular to a smart sensor collar for pets and its control method. Background Technology

[0002] With the development of the times and the improvement of people's living standards, more and more families are starting to keep pets. However, due to language barriers, humans cannot accurately understand the meaning behind a pet's barking and behavior. This is especially true for working people, who often lack experience in identifying abnormal barking and behavior from their pets because they spend relatively little time with them.

[0003] In the prior art, CN105706951B provides a smart pet collar and its implementation method. This prior art analyzes, recognizes, and translates pet language, and limits the pet's activity distance and time by setting a map range or time. It only involves recognizing pet language through a pet language database and does not involve analyzing the pet's needs. CN114999501A provides a pet voice recognition method and system based on neural networks. This prior art can identify the intensity of a pet's emotions based on its voice, but it does not involve pet behavior recognition or pet need assessment.

[0004] Therefore, there is a need to provide a smart sensor collar for pets and its control method, which can accurately and reliably determine the needs of pets and improve user experience and satisfaction. Summary of the Invention

[0005] This specification provides one or more embodiments of a control method for a smart pet sensor collar. The smart pet sensor collar includes at least a sound detection unit, a location detection unit, an activity detection unit, a physiological detection unit, a processor, and a lighting unit. The method includes: acquiring sound information of a target pet based on the sound detection unit; acquiring location information of the target pet based on the location detection unit; acquiring activity information of the target pet based on the activity detection unit; acquiring physiological information of the target pet based on the physiological detection unit; determining behavioral information of the target pet based on the sound information, activity information, and physiological information; processing the location information and behavioral information based on a demand model, where the demand model is a machine learning model; and controlling the lighting unit to display the demand type. This specification uses a trained neural network model to process multiple types of information, which can provide a more accurate determination of pet needs.

[0006] This specification provides one or more embodiments of a smart pet sensor collar, which includes at least a sound detection unit, a location detection unit, an activity detection unit, a physiological detection unit, a processor, and a light unit. The sound detection unit detects the sound information of the target pet; the location detection unit detects the location information of the target pet; the activity detection unit detects the activity information of the target pet; the physiological detection unit detects the physiological information of the target pet; the processor determines the behavioral information of the target pet based on the sound information, activity information, and physiological information; it processes the location information and behavioral information based on a demand model (a machine learning model) to determine the demand type of the target pet; and it controls the light unit to display the demand type.

[0007] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes the control method for the pet smart sensor collar described above. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is an exemplary structural diagram of a smart sensor collar for pets, as shown in some embodiments of this specification.

[0010] Figure 2 This is an exemplary flowchart of a control method for a smart sensor collar for pets, as shown in some embodiments of this specification.

[0011] Figure 3 This is an exemplary flowchart illustrating the acquisition of a requirement model through pre-training, according to some embodiments of this specification;

[0012] Figure 4 This is an exemplary flowchart illustrating the acquisition of a demand model through reinforcement training, according to some embodiments of this specification.

[0013] Figure 5 This is an exemplary schematic diagram illustrating the acquisition of a second tag according to some embodiments of this specification. Detailed Implementation

[0014] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0015] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0016] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0017] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0018] Figure 1 This is an exemplary structural diagram of a smart sensor collar for pets, shown according to some embodiments of this specification.

[0019] like Figure 1 As shown, the pet smart sensor collar 100 may include a sound detection unit 110, a positioning detection unit 120, an activity detection unit 130, a physiological detection unit 140, a processor 150, and a light unit 160.

[0020] The sound detection unit 110 refers to a unit capable of acquiring and analyzing sound-related data. For example, the sound detection unit 110 can acquire and analyze pet barks, etc.

[0021] In some embodiments, the sound detection unit 110 can be used to acquire the sound information of the target pet. For more information about the target pet and sound information, see [link to relevant documentation]. Figure 2 And its related descriptions.

[0022] In some embodiments, the sound detection unit 110 may include a microphone, a recording device, etc. The description of the sound detection unit 110 is merely illustrative and does not constitute a limitation on the implementation.

[0023] The location detection unit 120 refers to a unit with location data acquisition and analysis functions. For example, the location detection unit 120 can collect and analyze the location data of a pet.

[0024] In some embodiments, the location detection unit 120 can be used to acquire the location information of the target pet. For more information on location information, see [link to relevant documentation]. Figure 2 And its related descriptions.

[0025] In some embodiments, the positioning detection unit 120 may include at least one of a GPS satellite-based locator, a Bluetooth-based locator, a Wi-Fi-based locator, etc. The description of the positioning detection unit 120 is merely illustrative and does not constitute a limitation on the implementation.

[0026] The activity detection unit 130 refers to a unit with the function of collecting and analyzing motion data. For example, the activity detection unit 130 can collect and analyze the motion data of a pet.

[0027] In some embodiments, the activity detection unit 130 can be used to acquire activity information of the target pet. For more information on activity information, see [link to relevant documentation]. Figure 2 And its related descriptions.

[0028] In some embodiments, the activity detection unit 130 may include an accelerometer, a velocity sensor, an orientation sensor, etc. The accelerometer is used to collect acceleration information of the target pet; the velocity sensor is used to collect velocity information of the target pet; and the orientation sensor is used to collect orientation information of the target pet. The description of the activity detection unit 130 is merely illustrative and does not constitute a limitation on the implementation.

[0029] The physiological detection unit 140 refers to a unit with physiological data acquisition and analysis functions. For example, the physiological detection unit 140 can collect and analyze the physiological data of a pet.

[0030] In some embodiments, the physiological detection unit 140 is used to acquire physiological information of the target pet. For more information on physiological information, see [link to relevant documentation]. Figure 2 And its related descriptions.

[0031] In some embodiments, the physiological detection unit 140 may include a respiratory rate sensor, a heart rate sensor, and a temperature sensor. The respiratory rate sensor is used to acquire the respiratory rate information of the target pet; the heart rate sensor is used to acquire the heart rate information of the target pet; and the temperature sensor is used to acquire the body temperature information of the target pet. In some instances, the physiological detection unit 140 may also include a blood pressure sensor or similar device. The description of the physiological detection unit 140 is merely illustrative and does not constitute a limitation on the implementation.

[0032] Processor 150 refers to a device with computing capabilities. Processor 150 can process data and / or information obtained from other units or components of the device. Processor 150 can execute program instructions based on this data, information, and / or processing results to perform one or more functions described in this application. For example, processor 110 can acquire sound information, location information, activity information, and physiological information based on sound detection unit 110, location detection unit 120, activity detection unit 130, and physiological detection unit 140, respectively. For example, processor 150 can be used to determine the behavioral information of a target pet based on the sound information collected by sound detection unit 110, the activity information collected by activity detection unit 130, and the physiological information collected by physiological detection unit 140. For example, processor 150 can also be used to process the behavioral information and location information collected by location detection unit 120 based on a demand model to determine the demand type of the target pet. As another example, processor 150 can further be used to control the lighting unit to display the demand type.

[0033] In some embodiments, processor 150 may include one or more sub-processing devices (e.g., single-core processing devices or multi-core multi-chip processing devices). By way of example only, processor 150 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or any combination thereof.

[0034] In some embodiments, processor 150 may be local or remote. In some embodiments, processor 150 may be deployed on a cloud platform.

[0035] Lighting unit 160 refers to a unit with lighting display function. For example, lighting unit 160 can display multiple colors of light and / or multiple types of lighting effects.

[0036] The lighting unit 160 can display the target pet's need type in various ways. In some embodiments, the lighting unit 160 can display different need types using different light colors. For example, when the determined need type is "rest," the lighting unit 160 displays green light. The correspondence between light color and need type can be preset by the system or manually. In some embodiments, the lighting unit 160 can display different need types using different lighting effects (e.g., different flashing frequencies, flash counts, etc., can constitute different lighting effects). For example, when the target pet's need is not met for a long time, the processor 150 controls the lighting unit 160 to flash. The correspondence between lighting effect and need type can be preset by the system or manually. The description of the lighting unit 160 is merely illustrative and does not constitute a limitation on the implementation.

[0037] In some embodiments of this specification, the processor determines the target pet's need type based on the target pet's sound information, location information, activity information, and physiological information collected by each detection unit. This comprehensive approach improves the accuracy of need type determination. Simultaneously, the lighting display unit uses different colored lights to indicate different need types, allowing owners to more intuitively understand their pet's needs and thus enhancing their pet ownership experience.

[0038] Figure 2 This is an exemplary flowchart of a control method for a smart sensor collar for pets according to some embodiments of this specification. In some embodiments, process 200 may be executed by a processor. Figure 2 As shown, process 200 includes the following steps.

[0039] Step 210: Obtain the sound information of the target pet based on the sound detection unit.

[0040] A target pet refers to a pet that needs to be monitored and sensed. For example, a user's pet cat or dog. In some embodiments, the processor can designate a pet wearing a smart pet collar as the target pet. For instance, when a pet wearing a smart pet collar is detected, the processor can obtain relevant information about the pet and bind it as the target pet based on that information. This relevant information may include voiceprints, etc. As an example only, when a pet wearing a smart pet collar is detected, the processor receives a registration command from the client, obtains the pet's voice data and generates a registration voiceprint, and binds that pet as the target pet.

[0041] Sound information refers to data related to a pet's vocalizations. Examples include volume, frequency (or pitch), and timbre.

[0042] The processor can acquire the target pet's sound information in various ways. In some embodiments, the processor can acquire the target pet's sound information based on a sound detection unit. For example, the sound information of the target pet can be acquired in real time or at intervals, which can be set according to actual needs. The time interval can be preset manually or by the system. More information about sound detection units can be found in [link to relevant documentation]. Figure 1 And its related descriptions.

[0043] In some embodiments, the processor may acquire initial sound information based on the sound detection unit; and extract the sound information of the target pet from the initial sound information.

[0044] Initial sound information refers to the sound information initially acquired. This initial sound information may include noise and the target pet's sound information. Noise may include, but is not limited to, ambient background noise, human voices, and sounds not belonging to the target pet; this noise can interfere with the processor's acquisition of the target pet's sound information.

[0045] Similar to the acquisition of sound information, initial sound information can be acquired through methods such as sound detection units, which will not be elaborated here.

[0046] In some embodiments, the processor can extract the target pet's sound information by processing the initial sound information. In some embodiments, the data processing may include using algorithms, machine learning models, etc. For example, the processor can perform data processing using algorithms such as linear filters and spectral subtraction. Alternatively, the processor can perform data processing using machine learning models.

[0047] In some embodiments of this specification, initial sound information is acquired based on a sound detection unit, and then the sound information of the target pet is extracted. This method can obtain sound information of the target pet with low noise and high accuracy. At the same time, it can effectively reduce the interference of noise from non-target pets on the acquisition of the target pet's sound information. For example, if a target pet wearing a collar is in the same room as a non-target pet, and the former does not make any sound while the latter makes noise, the method described in the aforementioned embodiments can eliminate the noise from the non-target pet, reduce interference, and make the acquired sound information of the target pet more realistic and reliable.

[0048] Step 220: Obtain the location information of the target pet based on the location detection unit.

[0049] Location information refers to data related to the location of the target pet. Examples include the target pet's location coordinates and location type (e.g., living room, kitchen, outdoors).

[0050] The processor can acquire the target pet's location information in various ways. In some embodiments, the processor can acquire the target pet's location information based on a location detection unit. For example, the processor can acquire the target pet's location information in real time or at regular intervals using the location detection unit, which can be set according to actual needs. More information about the location detection unit can be found in [link to relevant documentation]. Figure 1 And related descriptions. In some embodiments, if the target pet's activity area is equipped with monitoring devices such as cameras, the processor can obtain the target pet's location information through image recognition or other methods by using images captured by the monitoring devices. In some embodiments, the processor can also obtain the target pet's location information through user input, which is not limited in this specification.

[0051] Step 230: Obtain the activity information of the target pet based on the activity detection unit.

[0052] Activity information refers to data related to a pet's activities. This includes, for example, the target pet's activity status (e.g., still, moving), and activity speed.

[0053] The processor can acquire the target pet's activity information in various ways. In some embodiments, the processor can acquire the target pet's activity information based on an activity detection unit. For example, the processor can acquire the target pet's activity information in real time or at regular intervals using the activity detection unit, which can be configured according to actual needs. More information about the activity detection unit can be found in [link to relevant documentation]. Figure 1 And related descriptions. In some embodiments, the processor can obtain the activity information of the target pet through images captured by the monitoring device and through methods such as image recognition. In some embodiments, the processor can also obtain the activity information of the target pet through user input, which is not limited in this specification.

[0054] Step 240: Obtain physiological information of the target pet based on the physiological detection unit.

[0055] Physiological information refers to data related to a pet's physiological characteristics. For example, the target pet's body temperature, heart rate, and respiratory rate.

[0056] The processor can acquire the target pet's physiological information in various ways. In some embodiments, the processor can acquire the target pet's physiological information based on a physiological detection unit. For example, the processor can acquire the target pet's physiological information in real time or at regular intervals through the physiological detection unit, which can be set according to actual needs. More information about the physiological detection unit can be found in [link to relevant documentation]. Figure 1 And related descriptions. In some embodiments, the processor can also obtain the target pet's physiological information through user input, which is not limited in this specification.

[0057] Step 250: Determine the target pet's behavioral information based on sound information, activity information, and physiological information.

[0058] Behavioral information refers to data related to a pet's behavior. This information can include the target pet's behavior type (e.g., lying down, eating, sleeping, running, etc.), behavior timing, and behavior frequency. Behavioral timing information refers to information related to the duration of the pet's behavior. For example, it can include the start time, end time, and duration of the behavior. Behavioral frequency refers to the number of times a pet's behavior occurs within a unit of time (e.g., 24 hours). For example, the target pet's behavior frequency could be eating 3 times per day.

[0059] The processor can acquire behavioral information about the target pet in various ways. In some embodiments, the processor can acquire behavioral information about the target pet through user input.

[0060] In some embodiments, the processor can construct a target vector based on the target pet's sound information, activity information, and physiological information, and determine the target pet's behavioral information through a vector database.

[0061] The elements of the target vector can include the target pet's vocal information, activity information, and physiological information. For example, the classification vector can be (A, B, C), where A represents vocal information, B represents activity information, and C represents physiological information.

[0062] A vector database is a database used to store, index, and query vectors. Vector databases enable rapid similarity queries and other vector management functions when dealing with large numbers of vectors.

[0063] In some embodiments, the vector database may include multiple reference vectors and their corresponding reference behavioral information. The reference vectors can be constructed based on historical data collected from the target pet or other pets (e.g., pets wearing other pet smart sensor collars), including historical voice information, historical activity information, and historical physiological information. For example, multiple reference vectors can be obtained by constructing vectors from multiple historical data points. The reference behavioral information corresponding to the reference vectors can be obtained through manual annotation of historical voice information, historical activity information, and historical physiological information. In some embodiments, a vector database can be constructed based on multiple reference vectors and their corresponding reference behavioral information.

[0064] In some embodiments, the processor can determine reference vectors that meet preset conditions as associated vectors based on the target vector by searching a vector database, and use the reference behavior information corresponding to the associated vectors as the behavior information of the target pet. The preset conditions may include vector distance being less than a distance threshold, minimum vector distance, etc. The distance threshold may be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs; this specification does not impose any restrictions on it.

[0065] In some embodiments, the processor can also process the target pet's sound information, activity information, and physiological information through a behavior determination model to determine the target pet's behavior information.

[0066] A behavior-determining model can be a machine learning model. There are various types of behavior-determining models. For example, a behavior-determining model can include a neural network (NN), a deep neural network (DNN), a recurrent neural network (RNN), or any combination thereof.

[0067] In some embodiments, the input to the behavior determination model can be the target pet's voice information, activity information, and physiological information, and the output of the behavior determination model can be the target pet's behavior information.

[0068] In some embodiments, the behavior determination model can be obtained through training. For example, a fourth training sample is input into the initial behavior determination model, and a loss function is established based on the fourth label and the output of the initial behavior determination model. The parameters of the initial behavior determination model are updated. The model training is completed when the loss function of the initial behavior determination model meets a preset condition, which may be the convergence of the loss function, the number of iterations reaching a threshold, etc.

[0069] In some embodiments, the fourth training sample may be sample sound information, sample activity information, and sample physiological information of the target pet or sample pet. The fourth label may be actual behavioral information of the target pet or sample pet.

[0070] In some embodiments, the fourth training sample can be obtained based on historical data. The fourth label can be manually labeled. The method for obtaining the fourth training sample is similar to that for obtaining the first training sample. For more information on the sample pets and the method for obtaining the first training sample, please refer to [link to relevant documentation]. Figure 3 And its related descriptions.

[0071] In some embodiments, the processor can input the output of the behavior determination model and the target pet's location information into the demand model to determine the target pet's demand type. For details on the demand model and the target pet's demand type, please refer to the relevant descriptions below.

[0072] In some embodiments, the output of the behavior determination model can be used as the input of the demand model, and the behavior determination model and the demand model can be obtained through joint training. Joint training can be performed during the pre-training process of the demand model or during the reinforcement training process of the demand model. For more information on the pre-training and reinforcement training of the demand model, please refer to [link to relevant documentation]. Figure 3 , 4 And its related descriptions.

[0073] Taking joint training during reinforcement training as an example, the exemplary joint training process includes: inputting sample sound information, sample activity information, and sample physiological information into the behavior determination model to obtain the sample behavior information output by the behavior determination model; inputting the sample behavior information and sample location information output by the behavior determination model into the initial demand model to obtain the demand type output by the initial demand model. Based on the second label and the output of the initial demand model, a loss function is constructed, and the parameters of the behavior determination model and the initial demand model are updated based on the loss function until a preset condition is met, and training is completed. The preset condition may include one or more of the following: the loss function is less than a threshold, convergence, or the training period reaches a threshold.

[0074] In some embodiments of this specification, behavior determination models can quickly and accurately determine the behavioral information of a target pet, facilitating subsequent prediction of the target pet's needs.

[0075] Step 260: Process the location information and behavioral information based on the demand model to determine the target pet's demand type.

[0076] Need type refers to the specific type of a pet's needs. For example, feeding needs, petting needs, entertainment needs, rest needs, and no needs.

[0077] In some embodiments, the processor can analyze and process the target pet's location information and behavior information to determine the target pet's need type. In some embodiments, the processor can process the location information and behavior information based on a need model to determine the target pet's need type. More information about the location information and behavior information can be found in steps 220 and 250 and their related descriptions.

[0078] The demand model can be a machine learning model. There can be various types of demand models. For example, a demand model can include neural network models, deep neural network models, recurrent neural network models, or any combination thereof.

[0079] In some embodiments, the input to the demand model can be the target pet's location and behavior information, and the output of the demand model can be the target pet's demand type.

[0080] In some embodiments, the output of the demand model may also include the demand probability corresponding to the demand type. The demand probability reflects the likelihood that a certain demand type represents an actual pet demand. The demand probability can be characterized by a quantitative indicator. For example, the demand probability can be expressed as a percentage; the higher the percentage, the higher the likelihood that the demand type output by the demand model represents an actual pet demand. The demand probability can also be represented in other ways, such as using letter grades, etc., and this specification does not limit this. As an example only, the output of the demand model is [(j1, k1), (j2, k2), (j3, k3)], indicating that the demand probability of demand type j1 is k1, the demand probability of demand type j2 is k2, and the demand probability of demand type j3 is k3.

[0081] In some embodiments, the demand model can be obtained through training. In some embodiments, the processor can obtain an initial demand model through pre-training before the pet smart sensor collar leaves the factory. In some embodiments, the pre-training process includes: acquiring a first training sample set; training an initial machine learning model based on the first training sample set to determine the initial demand model. Further explanation of pre-training can be found in [link to relevant documentation]. Figure 3 And its related descriptions.

[0082] In some embodiments, the processor can perform reinforcement training on the initial demand model during the use of the pet smart sensor collar. In some embodiments, the reinforcement training process includes: acquiring a second training sample set; training the initial demand model based on the second training sample set; and determining the demand model. Further explanation of reinforcement training can be found at [link to relevant documentation]. Figure 4 And its related descriptions.

[0083] Step 270: Control the lighting unit to display the required type.

[0084] In some embodiments, the processor can generate corresponding control instructions based on the demand type, and control the lighting unit to display the demand type through these control instructions. Control instructions refer to instructions that control the lighting unit to display the target pet's demand type. In some embodiments, the processor can generate different control instructions based on different demand types. For example, the processor can generate control instructions to control the lighting unit to display different colors, durations, and frequencies based on different demand types. The specific correspondence between the demand type and the display color, duration, and frequency of the lighting unit can be preset according to user needs. As an example only, the processor can generate control instructions to control the lighting unit to display red for 0.5 seconds and once for a demand type of feeding need.

[0085] In some embodiments, when a user provides a corresponding service to a target pet based on a predicted demand type, the processor can further acquire user actions and determine the target pet's satisfaction with the user actions.

[0086] User actions refer to the specific actions a user performs on their pet. Examples include feeding, petting, providing companionship and entertainment, putting the pet to sleep, and no action. In some embodiments, user actions can correspond to different types of needs. For example, a feeding need corresponds to a feeding action, and an entertainment need corresponds to a companionship and entertainment action.

[0087] In some embodiments, the processor can acquire user actions in various ways. For example, the processor can acquire user actions based on user input. Another example is that the processor can use user actions corresponding to a request type as the actual user action. Yet another example is that the processor can acquire user actions through images captured by a monitoring device, using image recognition or other methods.

[0088] Satisfaction refers to a parameter that characterizes the degree to which a pet is satisfied with a user's actions. Satisfaction can be represented by quantitative indicators. For example, satisfaction can be represented by a number from 1 to 10, with higher numbers indicating greater satisfaction. Satisfaction can also be represented in other ways, such as using a rating scale, without limitation.

[0089] In some embodiments, the processor can analyze and process the target pet's location information, behavior information, and user actions to determine the target pet's satisfaction with the user actions. In some embodiments, the processor can process the target pet's location information, behavior information, and user actions based on a satisfaction model to determine the target pet's satisfaction with the user actions. More information about location information and behavior information can be found in [link to relevant documentation]. Figure 2 And its related descriptions.

[0090] The satisfaction model can be a machine learning model. In some embodiments, the satisfaction model may include a neural network model, a deep neural network model, a recurrent neural network model, or any combination thereof.

[0091] In some embodiments, the input to the satisfaction model can be the target pet's location information, the target pet's behavior information, and user actions. The output of the satisfaction model can be the target pet's satisfaction with the user actions. The user actions input to the satisfaction model can be user actions performed on the target pet based on the demand types predicted by the demand model.

[0092] In some embodiments, the processor may determine the training label (i.e., the second label) for reinforcement training of the demand model based on the output of the satisfaction model.

[0093] In some embodiments, the processor can input the second sample location information, second sample behavior information, and at least one candidate user operation into a satisfaction model to obtain the satisfaction level corresponding to at least one candidate user operation. The demand type corresponding to the candidate user operation with the highest satisfaction level is then used as the demand type corresponding to the second sample location information and second sample behavior information of the target pet, i.e., as the second label. For further explanation on obtaining the second label, please refer to [link to documentation]. Figure 5 And its related descriptions.

[0094] In some embodiments, the satisfaction model can be obtained by: acquiring a third training sample set; training an initial satisfaction model based on the third training sample set; and determining the satisfaction model.

[0095] The third training sample set refers to the set of samples used to train the satisfaction model. In some embodiments, the third training sample set may include third training samples and their third labels.

[0096] In some embodiments, the third training sample may include the target pet's third sample location information, third sample behavior information, and at least one sample user operation.

[0097] The third sample location information refers to the target pet's historical location information. The third sample location information may be the same as or different from the second sample location information.

[0098] Third-sample behavioral information refers to the target pet's historical behavioral information. The third-sample behavioral information may or may not be the same as the second-sample behavioral information. For more information on location information and behavioral information, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0099] In some embodiments, the third label may be the satisfaction level of at least one sample user's operation.

[0100] In some embodiments, the processor can acquire a third training sample set in multiple ways. For example, it can acquire the target pet's historical location information and historical behavior information over a certain historical period as the third training sample, while also using at least one sample user operation as the third training sample. In some embodiments, the at least one sample user operation used as the third training sample can include all types of user operations. Accordingly, the target pet's reactions to each of the at least one sample user operation can be manually collected and labeled to obtain a third label. As an example only, if the target pet does not eat after a user performs a feeding operation on it during that historical period, then the target pet's satisfaction with this "feeding operation" sample user operation is labeled as low.

[0101] In some embodiments, the processor can train an initial satisfaction model based on a third training sample set to determine the satisfaction model. In some embodiments, the processor can input the third training samples into the initial satisfaction model, establish a loss function based on the third label and the output of the initial satisfaction model, update the parameters of the initial satisfaction model, and complete the model training when the loss function of the initial satisfaction model meets preset conditions, thus determining the satisfaction model. The preset conditions may include loss function convergence, the number of iterations reaching a threshold, etc.

[0102] The target pet's satisfaction with user actions is influenced by numerous information features, which are insufficient to determine satisfaction solely through individual rules or simple rules. The satisfaction-based model described in some embodiments of this specification, which determines the target pet's satisfaction with user actions, can leverage a large number and wide range of information features, overcoming the limitations of traditional rule-based satisfaction determination. Rule-based methods, limited by their complexity, can only rely on a limited number of information features and are constrained by manually defined rules, making it difficult to accurately obtain the target pet's actual satisfaction with user actions. In contrast, prediction based on machine learning techniques can utilize more and richer information features, and can use the target pet's actual feedback information as training data, resulting in higher accuracy in determining satisfaction levels.

[0103] In some embodiments, the processor can also determine the health status of the target pet based on the target pet's behavioral information and basic information.

[0104] Basic information refers to information related to the pet's fundamental characteristics. This includes, for example, the target pet's breed, age, weight, body length, past medical history, deworming history, and health check records.

[0105] Health status refers to data information that characterizes the degree of health. For example, health status can include healthy, unhealthy, etc. Health status can be characterized by quantitative indicators. For example, health status can be represented by a number from 1 to 10, with a higher number indicating a higher degree of health for the target pet. In some embodiments, health status can also be represented in other ways, such as using letters, etc., without limitation.

[0106] In some embodiments, the processor can also determine the health status of the target pet based on its behavioral information and basic information, using preset rules. Preset rules can refer to pre-defined rules used to determine the health status of the target pet. These preset rules can be determined based on experience.

[0107] In some embodiments, the preset rule can be: when the total daily sleep time of the target pet does not meet a first preset condition for a consecutive preset number of days, the target pet's health status is determined to be unhealthy. The first preset condition can be that the total daily sleep time is less than a duration threshold. The consecutive preset number of days and the duration threshold can be set based on experience or historical data.

[0108] In some embodiments, the preset rule may also be: when the target pet's breathing rate is within a frequency threshold, the target pet's health status is considered healthy; if the breathing rate exceeds the frequency threshold, the target pet's health status is considered unhealthy. The frequency threshold can be set based on experience or historical data.

[0109] In some embodiments, the preset rule may further be: when the target pet's daily food intake and / or daily feeding frequency do not meet the second preset condition for a consecutive preset number of days, the target pet's health status is determined to be unhealthy. The second preset condition may be that the daily food intake exceeds a food intake threshold and / or the daily feeding frequency exceeds a feeding frequency threshold. The consecutive preset number of days, the food intake threshold, and the feeding frequency threshold can be set based on experience or historical data.

[0110] In some embodiments, the preset rule may further be: when the target pet's daily water intake and / or daily water intake frequency does not meet a third preset condition for a consecutive preset number of days, the target pet's health status is determined to be unhealthy. The third preset condition may be that the daily water intake exceeds a water intake threshold and / or the daily water intake frequency exceeds a water intake frequency threshold. The consecutive preset number of days, water intake threshold, and water intake frequency threshold can be set based on experience or historical data.

[0111] In some embodiments, the preset rule may further be: if the target pet's daily urine output and / or daily urine frequency do not meet the fourth preset condition for a consecutive preset number of days, the target pet's health status is determined to be unhealthy. The fourth preset condition may be that the daily urine output exceeds a urine output threshold and / or the daily urine frequency exceeds a urine frequency threshold. The consecutive preset number of days, urine output threshold, and urine frequency threshold can be set based on experience or historical data.

[0112] In some embodiments, the preset rules may also take various other forms, which are not limited in this specification.

[0113] The health status of a target pet can be determined based on its behavioral information and basic information, as described in some embodiments of this manual. This method can comprehensively consider the influence of multiple factors on the health status of the target pet, making the determination process convenient, efficient, and accurate. It avoids the consequences of delays in diagnosis caused by human observation errors, and allows users to take timely measures to keep their pets healthy.

[0114] The methods described in some embodiments of this specification can comprehensively analyze a pet's vocalizations, physiological condition, behavior, location, and the correlation between them, accurately and reliably obtain the pet's needs, reduce errors and time costs caused by manual judgment of the pet's needs, ensure the healthy growth of the pet, and improve user satisfaction and experience.

[0115] Figure 3 This is an exemplary flowchart illustrating a requirement model according to some embodiments of this specification. In some embodiments, process 300 may be executed by a processor. Figure 3 As shown, process 300 includes the following steps.

[0116] Step 310: Obtain the first training sample set.

[0117] The first training sample set refers to the collection of training data used for pre-training. Pre-training refers to model training conducted before the pet smart sensor collar leaves the factory or before the user purchases it. Pre-training does not require high accuracy and has low training cost requirements.

[0118] In some embodiments, the first training sample set may include a first training sample and its first label.

[0119] The first training sample refers to the training sample used in pre-training.

[0120] In some embodiments, the first training sample may include the first sample location information and the first sample behavior information of the sample pet.

[0121] A sample pet refers to the object collected during the training data acquisition process for pre-training. In some embodiments, a sample pet may be a pet used for experimental data collection. For example, a cat or dog may be placed in a test room (e.g., a room simulating a pet's home environment) to collect relevant information. In some embodiments, a sample pet may also be a pet that has historically worn other smart sensor collars. For example, a sample pet may be a pet wearing another smart sensor collar that is already in use at the factory.

[0122] The first sample location information refers to the location information of the sample pet. The first sample behavior information refers to the behavior information of the sample pet.

[0123] The types of sample pets used during pre-training can include one or more. For example, the sample pets can be cats and dogs. In some embodiments, a demand model for a broad pet category can be trained based on the experimental data collected from the sample pets. In some embodiments, demand models for a specific pet type can be trained separately based on the experimental data collected from the sample pets.

[0124] In some embodiments, the first label can be the actual need type of the sample pet. For more information on need types, see [link to relevant documentation]. Figure 2 And its related descriptions.

[0125] The first training sample and the first label can be obtained in a variety of ways.

[0126] In some embodiments, a large number of first training samples and their corresponding first labels can be obtained through experimental collection. For example, a set of first sample location information and first sample behavior information of sample pets in the experimental room can be obtained as a set of first training samples; after the sample location information and sample behavior information of a set of sample pets are collected experimentally, the actual need type of the sample pets can be labeled by manual observation and analysis based on the subsequent behavior of the sample pets, and this label can be used as the first label corresponding to the set of first training samples.

[0127] In some embodiments, a first training sample and its first label can be obtained based on historical data collected from other smart sensor collars already in use. In some embodiments, historical location information and historical behavior information collected from other smart sensor collars already in use can be obtained as the first training sample. In some embodiments, the first label can be obtained based on historical prediction results from other smart sensor collars already in use. For example, when other smart sensor collars already in use predict a demand type based on certain historical location information and historical behavior information, the accuracy of the prediction can be determined based on the satisfaction of the sample pets after performing the user operation corresponding to the demand type. When the prediction is accurate, the predicted demand type can be used as the first label corresponding to the historical location information and historical behavior information for pre-training of smart sensor collars not yet manufactured.

[0128] Step 320: Train an initial machine learning model based on the first training sample set to determine the initial requirement model.

[0129] An initial machine learning model refers to a raw, untrained machine learning model. In some embodiments, the initial machine learning model may include any one or a combination of various feasible initial models such as neural network models, deep neural network models, and recurrent neural network models.

[0130] The initial requirement model refers to the machine learning model obtained after the initial machine learning model has been pre-trained.

[0131] In some embodiments, the processor can determine an initial demand model through training.

[0132] In some embodiments, the processor can train an initial demand model based on a first training sample set to obtain a demand model. An exemplary training process includes: inputting the first training samples into the initial machine learning model, constructing a loss function based on the output of the initial machine learning model and a first label, updating the parameters of the initial machine learning model through the loss function, until the trained initial machine learning model meets preset conditions to obtain the initial demand model, wherein the preset conditions may be that the loss function is less than a threshold, convergence, or the training period reaches a threshold, etc.

[0133] Step 330: Based on the initial requirement model, determine the requirement model.

[0134] In some embodiments, the processor can directly train an initial machine learning model using a large amount of data related to the target pet to obtain a demand model. For more information on target pets, see [link to relevant documentation]. Figure 2 And its related descriptions.

[0135] In some instances, the processor can perform reinforcement training on the initial demand model to determine the demand model. Reinforcement training refers to retraining the pre-trained model after the pet smart sensor collar has left the factory, further optimizing the parameters of the pre-trained model so that it is more adapted to the target pet. For more information on how to obtain a demand model through reinforcement training, see [link to relevant documentation]. Figure 4 And its related descriptions.

[0136] The methods described in some embodiments of this specification allow for the pre-training of an initial demand model based on a large number of sample pets' first-sample location information and first-sample behavior information before the pet smart sensor collar leaves the factory. This pre-training enables the initial demand model to have strong generalization ability. Furthermore, pre-training eliminates the need for extensive training data when applying the initial demand model to specific individual pets, reducing training costs.

[0137] Figure 4 This is an exemplary flowchart illustrating the acquisition of a demand model through reinforcement training, according to some embodiments of this specification. In some embodiments, Figure 4 One or more operations in the process 400 shown can be performed Figure 1 This is implemented in the pet smart sensor collar 100 shown. In some embodiments, process 400 can be executed by a processor. Figure 4 As shown, process 400 includes the following steps.

[0138] Intensive training can be conducted during the manufacturing process of the pet smart needs collar. Targeted intensive training can be performed on the target pet for a period of time before it starts wearing the collar. See the related description below for more information on intensive training.

[0139] Step 410: Obtain the second training sample set.

[0140] The second training sample set refers to the collection of training data used to reinforce the initial training requirement model. In some embodiments, the second training sample set may include second training samples and their second labels.

[0141] The second training sample refers to the training sample corresponding to reinforcement training.

[0142] In some embodiments, the second training sample may include second sample location information and second sample behavior information of the target pet.

[0143] The second sample location information of the target pet refers to the location information collected from the target pet as training samples. In some embodiments, the second sample location information of the target pet may include the location information of the target pet collected by the location detection unit. For more information on location information, please refer to [link to relevant documentation].Figure 2 And its related descriptions.

[0144] The second sample behavioral information of the target pet refers to the behavioral information collected from the target pet as training samples. For more information on behavioral information, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0145] In some embodiments, the second label is the target pet's actual need type. For more information on need types, see [link to relevant documentation]. Figure 2 And its related descriptions.

[0146] The second training sample set can be obtained in various ways. In some embodiments, the processor can obtain the second training samples based on the target pet's historical detection data. For example, the processor can use the historical location information and historical behavior information obtained from the pet smart sensor collar's historical detection of the target pet as the second training samples. In some embodiments, the second label can be obtained through manual annotation. As an example only, after collecting the target pet's historical location information and historical behavior information, the target pet's actual need type can be manually labeled based on observation and analysis of the target pet's subsequent behavior, and this label can be used as the second label.

[0147] In some embodiments, the processor can process the second sample location information, second sample behavior information, and at least one candidate user operation of the target pet in the second training samples based on a satisfaction model to determine the target pet's sample satisfaction with at least one candidate user operation; determine the target user operation based on the at least one sample satisfaction; and determine the demand type corresponding to the target user operation as the target pet's true demand type. This embodiment can determine the second label corresponding to each second training sample. For more information on how to obtain the second label based on the satisfaction model, see [link to documentation]. Figure 5 And related content.

[0148] Step 420: Train the initial demand model based on the second training sample set and determine the demand model.

[0149] In some embodiments, the processor can train the initial demand model based on a second training sample set to obtain a demand model. An exemplary training process includes: inputting the second training samples into the initial demand model; constructing a loss function based on the output of the initial demand model and a second label; updating the parameters of the initial demand model using the loss function; and continuing until the trained initial demand model meets preset conditions to obtain the demand model. These preset conditions may include the loss function being less than a threshold, convergence, or the training period reaching a threshold. For more information on the initial demand model, see [link to relevant documentation]. Figure 3 And its related descriptions.

[0150] In some embodiments of this specification, data can be collected on the target pet for a period of time before the target pet wears the smart pet collar to reinforce the pre-trained initial demand model. Further reinforcement training can be carried out on specific individual pets to ensure that the demand model has high adaptability to different individual pets and effectively improve the accuracy of the smart pet collar in predicting the demand type of the target pet.

[0151] Figure 5 This is an exemplary schematic diagram illustrating the acquisition of a second tag according to some embodiments of this specification.

[0152] In some embodiments, the processor can process the second sample location information 511, the second sample behavior information 512, and at least one candidate user operation 513 of the target pet based on the satisfaction model 520, to determine the target pet's sample satisfaction 530 with the at least one candidate user operation 513; determine the target user operation 540 based on the at least one sample satisfaction 530; and determine the demand type 550 corresponding to the target user operation 540 as the target pet's actual demand type 560. Further explanation of the second sample location information, the second sample behavior information, and the demand type can be found in [reference needed]. Figures 2-4 And its related descriptions.

[0153] Candidate user actions refer to the user actions that are available for selection. For more information on user actions, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0154] In some embodiments, the processor can determine candidate user operations in multiple ways. For example, the processor can preset multiple user operations and randomly select at least one as a candidate user operation. Another example is that the processor can obtain at least one demand type corresponding to the second sample location information and second sample behavior information of the target pet based on a demand model, and determine the user operation corresponding to at least one demand type as at least one candidate user operation. More information on demand models and demand types can be found in [link to relevant documentation]. Figure 2 And its related descriptions.

[0155] Sample satisfaction refers to the satisfaction level corresponding to the candidate user's actions. For more information on satisfaction and satisfaction models, please refer to [link / reference needed]. Figure 2 And its related descriptions.

[0156] Target user actions refer to user actions determined based on candidate user actions.

[0157] In some embodiments, the processor may determine the target user operation 540 based on at least one sample satisfaction level 530 in a variety of ways. For example, the processor may sort at least one candidate user operation 513 according to at least one sample satisfaction level 530 and determine the candidate user operation with the highest sample satisfaction level as the target user operation 540.

[0158] In some embodiments, the processor can use the target pet's actual need type 560 determined by the satisfaction model 520 as a second label to train the initial need model 570. An exemplary training process includes inputting a second training sample (i.e., second sample location information 511 and second sample behavior information 512) and the second label (i.e., the target pet's actual need type 560) into the initial need model 570 to train and obtain a need model 580. For further explanation on training a need model based on the initial need model, see [link to documentation]. Figure 4 And its related descriptions.

[0159] The satisfaction level of the target pet is determined using a satisfaction model as described in some embodiments of this specification. This model then determines the target user's actions and the actual needs of the sample pet. It comprehensively considers multiple factors and uses intelligent analysis through a model to obtain accurate sample satisfaction levels. Based on this satisfaction level, reasonable target user actions and actual needs types are determined. By using a relatively easy-to-train satisfaction determination model to obtain highly accurate second labels, and then training the aforementioned needs model, the training difficulty and cost of the needs model are reduced, training efficiency is improved, and the accuracy of the needs model's output results is guaranteed.

[0160] This specification also provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes the control method for the pet smart sensor collar as described in any of the above embodiments.

[0161] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0162] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0163] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0164] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0165] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0166] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0167] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and are considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A control method for a smart sensor collar for pets, characterized in that, The pet smart sensor collar includes at least a sound detection unit, a positioning detection unit, an activity detection unit, a physiological detection unit, a processor, and a light unit; the method includes: The sound information of the target pet is obtained based on the sound detection unit; The location information of the target pet is obtained based on the location detection unit; The activity information of the target pet is obtained based on the activity detection unit; The physiological information of the target pet is obtained based on the physiological detection unit; The behavioral information of the target pet is determined based on the sound information, the activity information, and the physiological information; The location information and behavioral information are processed based on a demand model to determine the demand type of the target pet. The demand model is a machine learning model. Control the lighting unit to display the required type; When the required type is to provide a corresponding service for the target. Obtain user actions; The location information, behavioral information, and user actions are processed based on a satisfaction model to determine the target pet's satisfaction with the user actions. The satisfaction model is a machine learning model.

2. The method as described in claim 1, characterized in that, The demand model was obtained in the following way: Obtain a first training sample set, which includes a first training sample and its first label. The first training sample includes the first sample location information and the first sample behavior information of the sample pet. The first label is the actual need type of the sample pet. Based on the first training sample set, train an initial machine learning model to determine the initial requirement model; Based on the initial requirement model, the requirement model is determined.

3. The method as described in claim 2, characterized in that, The process of determining the demand model based on the initial demand model includes: Obtain a second training sample set, which includes a second training sample and its second label. The second training sample includes the second sample location information and the second sample behavior information of the target pet. The second label is the actual need type of the target pet. The initial demand model is trained based on the second training sample set, and the demand model is determined.

4. The method as described in claim 3, characterized in that, The methods for obtaining the second tag include: The second sample location information, the second sample behavior information, and at least one candidate user operation of the target pet are processed based on the satisfaction model to determine the sample satisfaction of the target pet with the at least one candidate user operation. The satisfaction model is a machine learning model. The target user's actions are determined based on the satisfaction level of at least one of the aforementioned samples; The demand type corresponding to the target user's operation is determined as the actual demand type of the target pet.

5. The method as described in claim 1 or 4, characterized in that, The satisfaction model was obtained through the following method: Obtain a third training sample set, which includes a third training sample and its third label. The third training sample includes the third sample location information, third sample behavior information and at least one sample user operation of the target pet. The third label is the satisfaction level of at least one sample user operation. The initial satisfaction model is trained based on the third training sample set, and the satisfaction model is determined.

6. The method as described in claim 1, characterized in that, The acquisition of the target pet's sound information based on the sound detection unit includes: Initial sound information is obtained based on the sound detection unit; The sound information of the target pet is extracted from the initial sound information.

7. The method as described in claim 1, characterized in that, The method further includes: Based on the behavioral information and the basic information of the target pet, the health status of the target pet is determined.

8. A smart sensor collar for pets, characterized in that, The smart pet collar includes at least a sound detection unit, a positioning detection unit, an activity detection unit, a physiological detection unit, a processor, and a light unit; The sound detection unit is used to detect the sound information of the target pet; The positioning detection unit is used to detect the positioning information of the target pet; The activity detection unit is used to detect the activity information of the target pet; The physiological detection unit is used to detect the physiological information of the target pet; The processor is used for The behavioral information of the target pet is determined based on the sound information, the activity information, and the physiological information; The location information and behavioral information are processed based on a demand model to determine the demand type of the target pet. The demand model is a machine learning model. Control the lighting unit to display the required type; When the required type is to provide a corresponding service for the target. Obtain user actions; The location information, behavioral information, and user actions are processed based on a satisfaction model to determine the target pet's satisfaction with the user actions. The satisfaction model is a machine learning model.

9. A computer-readable storage medium storing computer instructions, wherein when a computer reads the computer instructions in the storage medium, the computer executes the control method for the pet smart sensor collar as described in claims 1-7.

Images