Pet feeding method and pet feeder based on mobile communication network

Abstract

This disclosure relates to a pet feeding method and pet feeder based on a mobile communication network. The method includes: processing the pet's facial feature vectors and nasal print feature vectors using a pre-trained attention mechanism model to determine the pet's identity; acquiring various feature information of the pet and corresponding correction factors to generate target energy; determining the dispensing amount for each independent feeding chamber based on the target energy, the pet's original nutrient database, the inventory information of multiple independent feeding chambers, and the target nutrient composition matrix for each feeding chamber; releasing the corresponding substance into a mixing chamber according to the dispensing amount; and interacting with the user about the feeding information via a mobile communication network. This achieves a targeted, personalized, precise, and healthy feeding effect.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent pet feeding technology, and in particular to a pet feeding processing method and pet feeder based on a mobile communication network. Background Technology

[0002] With the development of the pet economy and the fast pace of modern life, pet owners have limited time. How to feed their pets more conveniently while also understanding their pets' condition and interacting with them has become an emerging problem and demand.

[0003] In existing technologies, for example, video communication can be achieved through a network camera IPC and a local area network Wi-Fi. Users can remotely view the pet's status through a mobile app and manually trigger the pet feeding device to feed the pet. Alternatively, users can place the device in a fixed location according to the pet's habits and set it to release a fixed amount of food at a preset time through a built-in clock chip.

[0004] This shows that most pet feeding devices have pre-set programs, limited functionality, and cannot meet truly personalized needs. Furthermore, they heavily rely on Wi-Fi, and in some locations, poor signal can prevent control. Summary of the Invention

[0005] To solve the above-mentioned technical problems, or at least partially solve them, this disclosure provides a pet feeding processing method and pet feeder based on a mobile communication network, which can improve the effect of targeted, personalized, precise, and healthy feeding based on the pet's real-time multi-dimensional dynamic information, and allow real-time interaction between the feeding status and the user.

[0006] This disclosure provides a pet feeding method based on a mobile communication network. The method is applied to a pet feeder equipped with a mobile communication module. The pet feeder includes a camera, multiple independent sealed feeding chambers, and a mixed feeding system. Each feeding chamber is equipped with an independent food dispensing mechanism and a food level sensor. The mixed feeding system includes a mixing chamber, a stirring mechanism, a discharge port, and a feeding tray. The method includes: calculating the fusion weight between facial feature vectors and nasal print feature vectors collected from the current pet via the camera using a pre-trained attention mechanism model; processing the facial feature vectors and nasal print feature vectors based on the fusion weights to generate an identity feature vector; performing similarity matching between the identity feature vector and a pre-registered pet identity feature database to determine the pet's identity; acquiring pre-stored breed tags, age, weight, and health information corresponding to the pet's identity; acquiring activity feature information collected by an electronic collar corresponding to the pet's identity; and acquiring historical feeding feature information recorded by the pet feeder corresponding to the pet's identity; and according to a preset formula... The growth energy formula corresponding to the label is used to calculate the initial energy based on the age and weight. The initial energy is then corrected to generate the target energy based on the first correction factor corresponding to the health characteristic information, the second correction factor corresponding to the activity characteristic information, and the third correction factor corresponding to the historical feeding characteristic information. The original nutritional database corresponding to the breed label and the age is obtained. The inventory information of multiple independent feeding chambers configured in the pet feeder corresponding to the breed label and the target nutrient matrix of each feeding chamber are also obtained. Based on the target energy, the original nutritional database, the inventory information of the multiple independent feeding chambers, and the target nutrient matrix of each feeding chamber, the dispensing amount corresponding to each independent feeding chamber is determined. Each independent feeding chamber is controlled to release the corresponding substance into the mixing chamber according to the dispensing amount. The substances are mixed evenly at a preset speed and time. The discharge port is controlled to deliver the mixture to the food bowl for the pet to eat. The pet eating video captured by the camera is sent to the client corresponding to the pet feeder via the mobile communication module for the user to view the feeding status.

[0007] This disclosure also provides a pet feeder, which includes a camera and a mobile communication module, as well as multiple independent sealed feeding chambers and a mixed feeding system. Each feeding chamber is equipped with an independent food dispensing mechanism and a storage sensor. The mixed feeding system includes a mixing chamber, a stirring mechanism, a discharge port, and a feeding tray. The pet feeder includes: an identity recognition module, used to calculate the fusion weight between facial feature vectors and nasal print feature vectors collected by the camera from the current pet through a pre-trained attention mechanism model, process the facial feature vectors and nasal print feature vectors based on the fusion weights to generate an identity feature vector, and perform similarity matching between the identity feature vector and a pre-registered pet identity feature database to determine the pet's identity identifier; a first acquisition module, used to acquire pre-stored breed tags, age, weight, and health feature information corresponding to the pet's identity identifier, as well as activity feature information collected by an electronic collar corresponding to the pet's identity identifier, and historical feeding feature information recorded by the pet feeder corresponding to the pet's identity identifier; and an energy generation module, used to generate energy according to a preset growth energy corresponding to the breed tag. The formula calculates initial energy based on the age and weight, and corrects the initial energy to generate target energy based on a first correction factor corresponding to the health characteristic information, a second correction factor corresponding to the activity characteristic information, and a third correction factor corresponding to the historical feeding characteristic information. A second acquisition module acquires the original nutritional database corresponding to the breed label and the age, and acquires the inventory information of multiple independent feeding chambers configured in the pet feeder corresponding to the breed label, as well as the target nutrient matrix of each feeding chamber. A determination module determines the dispensing amount corresponding to each independent feeding chamber based on the target energy, the original nutritional database, the inventory information of the multiple independent feeding chambers, and the target nutrient matrix of each feeding chamber. A feeding module controls each independent feeding chamber to release the corresponding substance into the mixing chamber according to the dispensing amount, mixes the substances evenly at a preset speed and time, controls the discharge port to dispose of the mixture into the food bowl for the pet to eat, and sends the pet eating video captured by the camera to the client corresponding to the pet feeder via the mobile communication module for the user to view the feeding situation.

[0008] This disclosure also provides an electronic device, the electronic device comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement a pet feeding processing method based on a mobile communication network as performed by the pet feeder provided in this disclosure.

[0009] This disclosure also provides a computer-readable storage medium storing a computer program for executing a pet feeding processing method based on a mobile communication network, as performed by the pet feeder provided in this disclosure.

[0010] The technical solution provided in this disclosure has the following advantages compared with the prior art: The pet feeding method based on a mobile communication network provided in this disclosure calculates the fusion weight between facial feature vectors and nasal print feature vectors collected from the pet via a camera using a pre-trained attention mechanism model. Based on this fusion weight, the facial feature vectors and nasal print feature vectors are processed to generate an identity feature vector. This identity feature vector is then matched with a pre-registered pet identity feature database to determine the pet's identity. Existing technologies for pet identification rely on simple facial recognition. However, pets' movement and posture characteristics are complex, and they cannot communicate using pet language. Furthermore, there are numerous pet breeds, and incomplete facial data can lead to misidentification, such as misidentifying certain dog breeds as cats. The technical solution provided in this disclosure, by simultaneously collecting multiple features and using an attention mechanism for identification, can more selectively capture suitable feature types, thereby more accurately identifying the pet's identity and laying the foundation for precise feeding.

[0011] By acquiring pre-stored breed tags, age, weight, and health characteristics corresponding to the pet's identity, as well as activity characteristics collected from the electronic collar corresponding to the pet's identity and historical feeding characteristics recorded by the pet feeder, an initial energy is generated based on the pet's age and weight according to a preset growth energy formula corresponding to the breed tag. This initial energy is then corrected to generate a target energy based on a first correction factor corresponding to the pet's health characteristics, a second correction factor corresponding to the pet's activity characteristics, and a third correction factor corresponding to the pet's historical feeding characteristics. However, existing technologies rely on simple, mechanical configurations; for example, feeding cats cat food in a quantity set by the owner. In contrast, the technical solution provided in this disclosure determines the feeding energy based on real-time multi-dimensional characteristics and factor adjustments to these multi-level features. This energy value not only matches the pet type (cat, dog, etc.) but also precisely matches the pet currently in the user's home that needs food, providing a customized energy level and laying the foundation for accurate subsequent feeding.

[0012] By acquiring the original nutritional database corresponding to the breed label and age, the system obtains the inventory information of multiple independent feeding chambers configured with the pet's breed label in the pet feeder, as well as the target nutrient composition matrix of each feeding chamber. Based on the target energy, the original nutritional database, the inventory information of the multiple independent feeding chambers, and the target nutrient composition matrix of each feeding chamber, the system determines the dispensing amount corresponding to each independent feeding chamber. It controls each independent feeding chamber to release the corresponding substance into the mixing chamber according to the dispensing amount, mixes the substances evenly at a preset speed and time, and controls the dispensing port to deliver the mixture to the food bowl for the pet to eat. The system also transmits video of the pet eating, captured by a camera, to the corresponding client of the pet feeder via a mobile communication module for the user to view the feeding status. In existing technologies, feeding mainly involves a single type of food such as cat or dog food. This food is a pre-configured, fixed formula that only differentiates between pet categories, without specifying the exact nutritional needs of a particular breed within that pet category or the appropriate proportions. Compared with existing technologies, the technical solution provided in this disclosure calculates the most suitable feed ratio for a pet based on the target energy, the nutritional database corresponding to the pet's sub-breed, and the nutritional composition matrix of each feeding compartment. This results in a customized nutritional meal, achieving truly precise feeding. Users are not veterinarians, and this method achieves personalized intelligent feeding more efficiently and intelligently. Furthermore, the signal coverage of the mobile communication network is sufficient for users to remotely and stably receive video for interaction, without relying on the limitations of uneven home Wi-Fi signal strength. Attached Figure Description

[0013] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0014] Figure 1 This disclosure provides an information interaction system for a pet feeder and a client based on a mobile communication network, as provided in this embodiment. Figure 2 A schematic flowchart of a pet feeding method based on a mobile communication network provided in an embodiment of this disclosure; Figure 3 Another schematic flowchart of a pet feeding method based on a mobile communication network provided in an embodiment of this disclosure; Figure 4 This is a schematic diagram of the structure of a pet feeder according to an embodiment of the present disclosure; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0015] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0016] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0017] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0018] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0019] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0020] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0021] The method will be described below with reference to specific embodiments. Figure 1This disclosure provides an information interaction system for a pet feeder and its client based on a mobile communication network. The system includes a pet feeder 11 and a client 12, which interact via a cloud server 13 within the mobile communication network. The pet feeder includes a mobile communication module and a camera. The mobile communication module can be configured with matching mobile network components based on the development of mobile networks and the required network speed. For example, it can be a 4G mobile network communication module using LTE Cat.1 or Cat.4 standards, ensuring the pet feeder can autonomously connect to the network in any location with 4G coverage, eliminating reliance on fixed broadband.

[0022] Specifically, based on the plug-and-play communication link established by the 4G network, after the pet feeder is powered on, the 4G mobile network communication module automatically reads the pre-installed SIM card information to access the operator's network, obtains a public IP address, and actively registers with the cloud server platform to establish a long-term communication channel. Subsequently, video streams, audio streams, and control commands that need to interact with the user client are transmitted through this encrypted channel. It should be noted that the user client can be configured on devices such as the user's mobile phone, iPad, or smartwatch that can provide the corresponding technical services, and the user can choose according to their needs. Information transmission between the user client and the pet feeder is conducted through a cloud server connected to the mobile communication network.

[0023] The camera installed on this pet feeder can be selected according to the needs of the scene, such as the required accuracy and processing speed, whether night vision monitoring is emphasized, and whether a dual-lens configuration is needed. The pet feeder is also equipped with a microphone and speaker, which can then collect video and audio information of the pet, as well as other multimedia information that needs processing, and play audio and video information sent by the user through the client, or pre-set multimedia information.

[0024] This pet feeder features multiple independently sealed feeding chambers and a mixing feeding system. Each feeding chamber is equipped with an independent food dispensing mechanism and a level sensor. The mixing feeding system includes a mixing chamber, a stirring mechanism, a dispensing port, and a feeding tray. Specifically, each chamber is independently sealed and equipped with a desiccant to prevent moisture absorption and spoilage. A weight sensor or photoelectric sensor is installed at the bottom of each chamber to monitor the remaining food level in real time, and an anti-jamming design is implemented. The dispensing port of each chamber is customized according to the size of the food particles, for example, it can be a funnel-shaped design to prevent splashing. After each dispensing, the dispensing volume is weighed and calibrated to ensure that the actual dispensing amount deviates from the target value by less than a preset threshold. The outlet of each chamber connects to the mixing chamber, which is equipped with a corresponding stirring mechanism, such as a spiral stirrer, suitable for mixing powder and granules through vibration. Mixing is achieved through high-frequency vibration or by utilizing airflow. Furthermore, a temperature sensor is included to detect the temperature of the mixing chamber; if the temperature exceeds 40°C, mixing is paused or cooling is initiated. Additionally, the pet feeder is equipped with a cleaning mechanism for automatic cleaning and heating drying after each feeding. A feeding bowl is installed at the outlet of the mixing chamber. The discharge port puts the mixture into the feeding bowl. A weight sensor is installed at the bottom of the feeding bowl to detect the amount of mixture consumed in the feeding bowl.

[0025] Figure 2 This is a flowchart illustrating a pet feeding method based on a mobile communication network provided in an embodiment of this disclosure. This method can be applied to... Figure 1 The pet feeder shown is operated by its processor, as follows: Figure 2 As shown, the method includes: Step 101: Using a pre-trained attention mechanism model, calculate the fusion weight between the facial feature vector and the nasal print feature vector collected by the camera for the current pet. Based on the fusion weight, process the facial feature vector and the nasal print feature vector to generate an identity feature vector. Then, perform similarity matching between the identity feature vector and a pre-registered pet identity feature database to determine the pet's identity identifier.

[0026] In one embodiment of this disclosure, when a pet approaches the feeder, a camera captures the pet's facial and nasal print images. To reduce computational complexity and improve recognition accuracy, the captured images may be preprocessed by normalizing the pixels. Then, the preprocessed facial image is input into a pre-trained first neural network, which, after undergoing multiple convolutional and pooling operations, outputs a facial feature vector F_face. Simultaneously, the preprocessed nasal print image is input into a pre-trained second neural network, which outputs a nasal print feature vector F_nose.

[0027] A pre-trained attention mechanism model dynamically calculates the fusion weight between facial feature vectors and nasal texture feature vectors. Specifically, first, the facial feature vector and nasal texture feature vector are concatenated into a combined vector. Then, the fusion weight α is calculated using the attention mechanism model. This attention mechanism model is a single-layer fully connected network whose parameters include a weight matrix W and a bias term b, both of which are pre-learned using a large number of training samples. For example, z = W·V_concat + b, where W is a 1×256 weight matrix and b is a scalar bias. After calculating the z value, the sigmoid function is used to map z to the interval between 0 and 1 to obtain the fusion weight α, i.e., α = sigmoid(z) = 1 / (1 + e^{-z}).

[0028] In a specific example, assuming the currently acquired facial image is clear but the nasal texture image is blurry, the attention mechanism model calculates z=3.8. Therefore, α= 1 / (1 + e^{-3.8})= 0.978. This α value indicates that, in the current case, the facial feature vector has higher credibility and should be given greater weight. Finally, a fused identity feature vector F_id is generated based on the fusion weight α, for example, F_id = α·F_face + (1-α)·F_nose. Substituting the above numerical calculations, it shows that the identity feature vector F_id is numerically closer to the high-quality facial feature vector, effectively suppressing the interference of low-quality nasal texture features.

[0029] The identity feature vector F_id is matched with a pre-registered pet identity feature database for similarity. This database stores the pet identity ID and corresponding standard feature vector for each pet. For example, the pet identity ID is "Xiaohuang" and the corresponding standard feature vector is [0.248, 1.510, -0.750, ...], and the pet identity ID is "Xiaohei" and the corresponding standard feature vector is [1.230, -0.560, 0.780, ...]. Optionally, this embodiment uses cosine similarity as the matching metric to calculate the similarity between the identity feature vector F_id and the standard feature vectors of each pet. For example, after calculation, the similarity between Xiaohuang's standard feature vector and the identity feature vector F_id is 0.98, and the similarity between Xiaohei's standard feature vector and the identity feature vector F_id is 0.23. Xiaohuang's similarity is greater than the preset threshold, so the current pet's pet identity ID is determined to be Xiaohuang's ID.

[0030] The identity recognition method in this embodiment, through attention mechanism fusion, can dynamically adjust the weights of facial and nasal features based on the quality of the currently acquired image, compared to traditional single-modal recognition or fixed-weight fusion. When the quality of the facial image is good, α approaches 1, with facial recognition as the primary focus; when the quality of the nasal image is good, α approaches 0, with nasal recognition as the primary focus, thus ensuring recognition accuracy.

[0031] Step 102: Obtain the pre-stored breed tag, age, weight and health characteristics information corresponding to the pet's identity, as well as the activity characteristics information collected by the electronic collar corresponding to the pet's identity and the historical feeding characteristics information recorded by the pet feeder corresponding to the pet's identity.

[0032] In one embodiment of this disclosure, the pet feeder retrieves pre-entered basic information from a cloud database or local storage based on the pet's identification, including: breed tag, age, weight, and health characteristics. For example, the current pet's breed tag is Labrador Retriever, its age is its birth date or current age of 5 months, its weight is the most recently measured weight of 28 kg, and its health characteristics include one or more combinations of past medical history, allergy history, medication history, and nutritional supplement preferences. It should be noted that in practical applications, users can enter the above information during initial registration via a client application, and subsequent modifications and updates are also supported.

[0033] Furthermore, the activity characteristic information collected by the electronic collar corresponding to the pet's identification tag is obtained. It should be noted that the activity characteristic information collected by the electronic collar can be transmitted to the pet feeder via a mobile communication network through interaction with a user client. Alternatively, the electronic collar can communicate directly with the pet feeder, periodically sending activity characteristic information carrying the pet's identification tag to the feeder. The electronic collar worn by the pet has built-in sensors such as accelerometers and gyroscopes to collect the pet's activity characteristic information in real time. To more clearly illustrate the activity characteristic information, examples include, but are not limited to: steps, activity duration, activity intensity, sleep duration, sleep quality, scratching frequency, abnormal barking frequency, licking frequency, heart rate, respiratory rate, and body temperature.

[0034] Furthermore, historical feeding characteristic information corresponding to the pet's identity can be retrieved from the information stored locally on the pet feeder or uploaded to the cloud database. This historical feeding characteristic information is generated by statistically analyzing the feeding records of the most recent N days, and includes, but is not limited to, information such as: average daily food intake, average daily feeding speed, number of times the pet refused to eat, average daily water intake, and weight change trend.

[0035] Step 103: Calculate the initial energy based on the age and weight according to the preset growth energy formula corresponding to the variety label. Then, correct the initial energy to generate the target energy based on the first correction factor corresponding to the health characteristic information, the second correction factor corresponding to the activity characteristic information, and the third correction factor corresponding to the historical eating characteristic information.

[0036] In one embodiment of this disclosure, the corresponding growth energy formula is first selected based on the pet's breed label, and the initial energy is calculated based on age and weight. Taking a 5-month-old Labrador Retriever weighing 28kg as an example, the growth energy formula corresponding to the Labrador Retriever is selected as follows: Initial energy A = k × BW^b×c×a, where k is the basic energy of the pet category, b is the power of the weight, c is the sub-breed factor, such as the Labrador Retriever factor, k, b, and c are all pre-acquired constants, BW is the weight, and a is the age. The initial energy is obtained after processing each value.

[0037] The first correction factor is determined based on health characteristic information. In this embodiment, it is determined whether the individual is healthy based on health characteristic information. If the individual is healthy, the basic health correction factor is set to 1. If the individual is not healthy, the severity of the disease is determined based on health characteristic information. For example, a mild disease is set between 1 and 1.1, a moderate disease is set between 1.1 and 1.3, and a severe disease is set between 1.3 and 1.5. The specific value should be set according to the specific situation. For example, a skin disease is set to 1.05, and postoperative healing is set to 1.4.

[0038] The second correction factor is determined based on the activity feature information. In this embodiment, the pet's activity status is divided into five levels based on the activity feature information. For example, taking the step count feature as an example, the step count is compared with the step count range corresponding to different levels to determine the pet's activity status level. Alternatively, the dimensions of different features can be combined to determine the level. Based on the determined activity status level, the pre-set second correction factor is obtained.

[0039] The third correction factor is determined based on historical feeding characteristics. In this embodiment, the average daily feeding energy of the pet is obtained based on historical feeding characteristics. The average daily feeding energy is compared with the initial energy. Different deviation levels are determined based on the difference, and a pre-set third correction factor corresponding to the deviation level is obtained.

[0040] Furthermore, based on the first, second, and third correction factors, the initial energy is corrected to generate the target energy. It should be noted that the specific calculation process can be selected according to application needs. To more clearly illustrate how to obtain the target energy, an example is provided below, but is not limited to: Target Energy B = Initial Energy A × First Correction Factor × Second Correction Factor × Third Correction Factor. Through this multi-layered correction mechanism, accurate calculation of the pet's energy needs is achieved, comprehensively considering life stage, health status, real-time activity level, and historical feeding trends, thus preparing data for the accuracy of subsequent individualized feeding.

[0041] Step 104: Obtain the original nutrition database corresponding to the breed label and the age, and obtain the inventory information of multiple independent feeding chambers configured in the pet feeder corresponding to the breed label and the target nutrient composition matrix of each feeding chamber. In one embodiment of this disclosure, a raw nutrition database corresponding to breed labels and age is obtained. Specifically, the pet feeder can send a query request for breed labels and age to a cloud server via a mobile communication module, and obtain the query results returned by the cloud server. The cloud server includes raw nutrition databases corresponding to different pets at different stages. For example, a Labrador Retriever at five months old is determined to be a puppy / large breed dog based on its age and breed. The corresponding raw nutrition database includes the recommended nutrient intake per 1000 kcal of energy, as shown in Table 1 below. It should be noted that the nutrients shown in Table 1 are only examples, and the specific content of the database shall prevail. Table 1

[0042] The pet feeder is equipped with multiple independent, sealed feeding compartments, each storing different types of nutrients. These compartments include, but are not limited to, staple food compartments, protein compartments, and vitamin compartments. The configuration of these compartments corresponds to the breed label to ensure that the nutritional needs of that breed of pet are met. Each feeding compartment has a weight sensor installed at its bottom to obtain the inventory level. If the sensor detects that the inventory level in any compartment is below a preset threshold, such as 10%, the pet feeder sends a replenishment reminder message with the compartment's identifier to the user's client via the communication module.

[0043] Next, it is necessary to obtain the target nutrient composition matrix for each feeding bin. Specifically, the target nutrient composition matrix is ​​explained as follows: each feeding bin corresponds to a nutrient composition vector, which represents the content of each nutrient per gram of the substance in that bin. The vectors of all feeding bins are combined to form the target nutrient composition matrix A. To illustrate the target nutrient composition matrix more clearly, an example is shown in Table 2. The dimension of matrix A is 5 × 8, where m is the number of feeding bins and n is the number of species to be monitored.

[0044] Table 2

[0045] Step 105: Determine the output quantity corresponding to each independent feeding bin based on the target energy, the original nutrient database, the inventory information of the multiple independent feeding bins, and the target nutrient composition matrix of each feeding bin.

[0046] Specifically, different calculation methods can be selected based on actual application needs to determine the output quantity corresponding to each independent feeding bin based on the target energy, original nutrient database, inventory information of multiple independent feeding bins, and target nutrient composition matrix of all feeding bins obtained in the above steps. It should be noted that the algorithm can use neural networks or other models to determine the precise feeding amount. To more clearly illustrate how to determine the output quantity corresponding to each independent feeding bin, this disclosure provides some implementation methods, but is not limited to these methods, as follows: Based on the target energy B calculated in step 103 and the number of times the pet eats per day, determine the energy required for the current meal. For example: Assume a target energy of 4637 kcal / day, 3 meals per day, and 1546 kcal per meal. Then, based on the original nutrient database (Table 1) and the target nutrient composition matrix (Table 2), obtain the nutrient requirement vector for this meal, as shown in Table 3. An example of the nutrient requirement calculation process is as follows: Table 3

[0047] The output quantity of each feeding bin is set as the solution function, and the inventory of each feeding bin is used as the constraint. The minimum variance method is used to iteratively solve the equation based on the target nutrient composition matrix and energy (as shown in Table 2) of the feeding bin, with the objective function being close to the nutrient requirement (Table 3). After iterative calculation, the output quantity corresponding to each independent feeding bin is determined. Assuming the calculated results are 320g of staple food in bin 1, 15g of protein powder in bin 2, 4.5g of calcium and phosphorus powder in bin 3, and 0.2g of vitamin powder in bin 4, the feed is mixed according to the output quantities of these independent feeding bins.

[0048] Step 106: Control each independent feeding chamber to release the corresponding substance into the mixing chamber according to the discharge volume. The stirring mechanism mixes the substances evenly at a preset speed and time. The discharge port is controlled to deliver the mixture into the food bowl for the pet to eat. The video of the pet eating captured by the camera is sent to the client corresponding to the pet feeder through the mobile communication module for the user to view the feeding situation.

[0049] In one embodiment of this disclosure, the pet feeder controls the rotation of the screw motor in compartment 1 to release 320g of staple food into the mixing chamber, the rotation of the micro-screw motor in compartment 2 to release 15g of protein powder into the mixing chamber, the rotation of the micro-screw motor in compartment 3 to release 4.5g of calcium and phosphorus powder into the mixing chamber, and the rotation of the micro-screw motor in compartment 4 to release 0.2g of vitamin powder into the mixing chamber, based on the dispensing volume of each compartment. This activates the stirring mechanism to mix the substances evenly at a preset speed and time, and then controls the discharge port to open, dispensing the mixture into the food bowl for the pet to eat. Simultaneously, a video of the pet eating, captured by a camera, is sent to the corresponding client of the pet feeder via a mobile communication module for the user to view the feeding status.

[0050] Optionally, in one embodiment of this disclosure, before mixing the various substances evenly according to a preset rotation speed and time, it is first detected whether the substance exiting the chamber is a preset temperature-sensitive substance. Temperature-sensitive substances refer to nutrients or drugs that will denature, inactivate, decompose, or deteriorate under conditions above a specific temperature, such as vitamin C which rapidly oxidizes and decomposes above 60°C. If it is a temperature-sensitive substance, the temperature of the mixing chamber is adjusted to match the temperature-sensitive substance, exemplified by, but not limited to, active cooling, active heating, and other cooling processes. The temperature adaptation processing provided by this disclosure can better assist intelligent pet feeding devices that need to mix multiple materials, and is particularly suitable for precision nutrition feeding scenarios containing temperature-sensitive substances such as vitamins and probiotics.

[0051] Optionally, in one embodiment of this disclosure, after controlling the discharge port to deliver the mixture into the food bowl for the pet to eat, the method further includes: collecting the actual amount of food consumed, the pet's eating speed, the pet's eating status, and obtaining the pet's activity data within a preset time period after eating through an electronic collar, and inputting it into a preset adaptive learning model to obtain the corresponding real-time nutrition database. Based on the real-time nutrition database, the method determines whether to adjust the original nutrition database. If adjustment is required, the adjustment information is sent to the client through the mobile communication module for the user to view.

[0052] Specifically, after each feeding, the pet feeder's built-in food bowl weight sensor collects the pet's actual food intake. The camera analyzes the pet's eating speed and state, including but not limited to normal eating, hesitant eating, and refusal to eat. Simultaneously, the system uses an electronic collar to acquire pet activity data over a preset time period after feeding, including steps, activity intensity, and activity duration.

[0053] The aforementioned multi-dimensional data is input into a preset adaptive learning model. This model can dynamically evaluate the degree of matching between the currently used nutrition database (i.e., the original nutrition database) and the pet's actual needs based on the pet's multi-dimensional data, and output a real-time nutrition database, which contains nutrient recommendation values ​​dynamically matched to the multi-dimensional data collected after the pet's feeding.

[0054] The system compares the real-time nutrition database with the original nutrition database, calculates the deviation of the recommended amount of each nutrient, and if the deviation of a certain nutrient exceeds a preset threshold, it determines that the original nutrition database needs to be adjusted. The adjustment information includes the name of the nutrient to be adjusted, the current recommended amount, the suggested adjustment amount, and the reason for the adjustment. The pet feeder sends this adjustment information to the user's client via a mobile communication module for the user to view and confirm. The user can view the adjustment suggestion in the client. If the user accepts the adjustment, the system will update the pet's original nutrition database, making the subsequent feeding plan more in line with the pet's actual nutritional needs. Therefore, this closed-loop feedback mechanism provided in this embodiment correlates actual food intake with post-feeding activity levels, enabling the system to identify the balance between energy intake and expenditure, avoid long-term energy excess or deficiency, continuously optimize the nutrition model, and achieve more precise adaptive feeding.

[0055] In summary, the pet feeding method based on a mobile communication network provided in this disclosure obtains the original nutritional database corresponding to the breed label and age, acquires the inventory information of multiple independent feeding chambers configured in the pet feeder corresponding to the pet's breed label, and the target nutrient composition matrix of each feeding chamber. Based on the target energy, the original nutritional database, the inventory information of the multiple independent feeding chambers, and the target nutrient composition matrix of each feeding chamber, the dispensing amount corresponding to each independent feeding chamber is determined. Each independent feeding chamber is controlled to release the corresponding substance into the mixing chamber according to the dispensing amount. The substances are mixed evenly at a preset rotation speed and time. The mixture is then dispensed into the food bowl for the pet to eat from. The video of the pet eating, captured by a camera, is sent to the corresponding client of the pet feeder via a mobile communication module for the user to view the feeding status. In the prior art, feeding mainly involves a single type of food such as cat food or dog food. This food is a pre-configured, fixed formula that only differentiates between pet categories, without specifying the exact nutritional needs of a particular breed within that pet category or the appropriate proportions.

[0056] Compared to existing technologies, the technical solution provided in this disclosure calculates the most suitable nutrient ratio for a pet based on the target energy level, a nutritional database corresponding to the pet's sub-breed, and the nutrient composition matrix of each feeding compartment. This results in a customized nutritional meal, achieving truly precise feeding. Users are not veterinarians, and personalized intelligent feeding is achieved more efficiently and intelligently. By combining a standard nutritional database with the nutrient composition matrix of multiple feeding compartments, precise proportions of nutrients tailored to the individual nutritional needs of pets are achieved. Compared to existing technologies that dispense food according to a fixed formula, this embodiment can dynamically adjust the proportions of various nutrients based on the pet's real-time status, significantly improving the accuracy and scientific nature of feeding. Furthermore, the signal coverage of the mobile communication network is sufficient for users to stably receive video for remote interaction, without relying on the limitations of uneven home Wi-Fi signal strength.

[0057] Optionally, in one embodiment of this disclosure, the pet feeder is further configured with a mobile chassis and a control system, the control system being configured to plan a movement path based on the pet's location information and control the pet feeder to move to the vicinity of the pet.

[0058] Specifically, in the above embodiments, before calculating the fusion weight between the facial feature vector and the nasal print feature vector collected by the camera from the current pet using a pre-trained attention mechanism model, the method further includes: In response to receiving a feeding instruction sent by the client via the mobile communication network, or in response to a trigger instruction for a feeding timer preset on the pet feeder, the system plays a user-preset feeding call voice. If the pet is detected to have arrived at the preset area within a preset time, the system inputs the pet's facial image captured by the camera into a pre-trained first neural network to extract facial feature vectors, and inputs the pet's nose print image captured by the camera into a pre-trained second neural network to extract nose print feature vectors for subsequent processing. The specific processing flow and implementation method refer to the specific execution process described in the above embodiments, and will not be repeated here.

[0059] If the pet is not detected to have reached the preset area within the preset time, an abnormal pet feeding information will be sent to the client via the mobile communication module. The user client can obtain the user's location information through the electronic collar. If the pet feeder needs to locate and check on the pet, it will send the pet's location information to the pet feeder via the mobile communication network.

[0060] After receiving the location information collected by the pet's electronic collar from the client, the system retrieves a pre-scanned home map through the control system, calculates its own coordinates in the indoor map in real time, and plans the optimal path from its own coordinates to the pet's location based on the marked passable areas, obstacles, and restricted areas on the map. It then controls the pet feeder to find the pet based on the optimal path. Once it reaches the pet, it collects the pet's video information and sends it to the client through the communication module, prompting the user to feed the pet. Based on the user's feedback on the feeding instructions, it decides whether to feed the pet.

[0061] In another embodiment of this disclosure, Figure 3 This is another schematic flowchart illustrating a pet feeding method based on a mobile communication network provided in an embodiment of this disclosure. This method can be applied to... Figure 1 The pet feeder shown is operated by its processor, as follows: Figure 3 As shown, the method includes: Step 201: Calculate the initial energy based on the age and weight according to the preset growth energy formula corresponding to the variety label, and input the health characteristic information, activity characteristic information and historical feeding characteristic information into a pre-trained anomaly detection neural network; Specifically, firstly, the corresponding growth energy formula is selected based on the pet's breed label, and the initial energy is calculated based on age and weight. Then, the collected health characteristics information, such as past medical history and current vital signs data, activity characteristics information, such as scratching frequency, licking frequency, steps, and activity intensity, and historical feeding characteristics information, such as food intake, feeding speed, and number of times food was refused, are used to form a multi-dimensional feature vector, which is then input into a pre-trained anomaly detection neural network.

[0062] Step 202: If no abnormalities are detected in the health feature information, the activity feature information, and the historical eating feature information, then the corresponding first correction factor, the second correction factor, and the third correction factor are output. Specifically, this anomaly detection neural network uses a large amount of healthy pet sample data to learn the feature distribution patterns under normal conditions. When the input feature vector falls within the normal distribution range, the neural network outputs a no-anomaly judgment and simultaneously outputs a first correction factor corresponding to health feature information, a second correction factor corresponding to activity feature information, and a third correction factor corresponding to historical feeding feature information. For the specific implementation of subsequent processing based on the above correction factors, please refer to the process described in the above embodiments, which will not be repeated here.

[0063] Step 203: If one or more abnormal feature information is detected, output health abnormality information, and / or activity abnormality information, and / or eating abnormality information, and generate a standardized consultation request carrying the pet's identity identifier, and send it to the preset cloud veterinary service platform through the mobile communication module; Specifically, when the anomaly detection neural network detects that any one or more of the following deviations from the normal range—health characteristics, activity characteristics, or historical feeding characteristics—it outputs a corresponding anomaly label, health anomaly information, and / or activity anomaly information, and / or feeding anomaly information. The pet feeder then packages this anomaly information along with the pet's identification, health data from the past N days, and the current video clip to generate a standardized consultation request conforming to a preset format, which is then sent to a preset cloud-based veterinary service platform via a mobile communication module.

[0064] Step 204: Receive treatment information returned by the veterinary service platform through the mobile communication network. The treatment information includes at least: the level of urgency, the feeding amount adjustment coefficient, and the type and dosage of therapeutic nutrients. Specifically, after receiving a consultation request, the veterinary service platform, through an online veterinarian or AI-assisted diagnosis, returns structured treatment information via the mobile communication network. This information includes at least the urgency level (e.g., Level 1, Level 2), a feeding adjustment factor, and the type and dosage of therapeutic nutrients to be added (e.g., increasing vitamin B1 by 10mg daily for three consecutive days). The pet feeder then takes appropriate action based on the urgency level.

[0065] Step 205: If the urgency level is detected to be less than or equal to the preset threshold, the initial energy is corrected according to the feeding amount adjustment coefficient to generate the target energy, and the original nutrient database is adjusted according to the type and dosage of the therapeutic nutrient to generate the target nutrient database. Specifically, if the urgency level is less than or equal to a preset threshold, such as level two or below, it indicates that the abnormal situation is controllable. The pet feeder corrects the initial energy according to the feeding amount adjustment coefficient to generate the target energy, and performs feeding according to the adjusted plan. It also adjusts the original nutrient database (as shown in Table 2 above) according to the type and dosage of therapeutic nutrients. After correcting Table 2, a target nutrient database is generated. It should be noted that in the above embodiment, if there are other ways to correct Table 2 to generate a new target nutrient database, the subsequent processing will be carried out according to the new nutrient database.

[0066] Step 206: If the detection indicates that the urgency level is greater than a preset threshold, the treatment information is sent to the client via the mobile communication module, and the user is prompted whether to feed the patient.

[0067] Specifically, if the urgency level exceeds a preset threshold, such as level one, it indicates that the abnormal situation may require user intervention or further guidance from a veterinarian. In this case, the pet feeder does not automatically adjust the feeding; instead, it sends treatment information to the client via a mobile communication module and notifies the user via a pop-up window or notification that a pet health abnormality has been detected, requesting veterinary advice. The user can view detailed veterinary treatment recommendations in the client and choose whether to feed the pet according to the recommendations. If the user confirms, the feeding will proceed according to the provided adjustment plan; if the user refuses, no feeding will be performed.

[0068] This embodiment achieves real-time monitoring of pet health status through an anomaly detection neural network. Under normal circumstances, the system continuously outputs corrective factors for precise feeding; under abnormal circumstances, it automatically triggers a closed-loop remote veterinary consultation to obtain professional treatment advice. Furthermore, through an urgency-based grading mechanism, the system ensures timely response to emergencies while delegating high-risk decision-making authority to the user for confirmation. This effectively solves the problems of existing technologies that only monitor without intervention or blindly intervene automatically, thus improving adaptive safety.

[0069] To achieve the above embodiments, this disclosure also proposes a pet feeder. Figure 4 This is a schematic diagram of a pet feeder according to an embodiment of the present disclosure, as shown below. Figure 4 As shown, the pet feeder 100 includes a camera, multiple independently sealed feeding chambers, and a mixing feeding system. Each feeding chamber is equipped with an independent food dispensing mechanism and a food level sensor. The mixing feeding system includes a mixing chamber, a stirring mechanism, a dispensing port, and a food tray. The pet feeder 100 also includes: The identity recognition module 1001 is used to calculate the fusion weight between the facial feature vector and the nasal print feature vector collected by the camera of the current pet through a pre-trained attention mechanism model, process the facial feature vector and the nasal print feature vector based on the fusion weight to generate an identity feature vector, and perform similarity matching between the identity feature vector and a pre-registered pet identity feature database to determine the pet's identity identifier. The first acquisition module 1002 is used to acquire pre-stored breed tag, age, weight and health characteristics information corresponding to the pet identification, as well as activity characteristics information collected by the electronic collar corresponding to the pet identification and historical feeding characteristics information recorded by the pet feeder corresponding to the pet identification. The energy generation module 1003 is used to calculate the initial energy based on the age and weight according to the preset growth energy formula corresponding to the variety label, and to correct the initial energy to generate target energy based on the first correction factor corresponding to the health characteristic information, the second correction factor corresponding to the activity characteristic information, and the third correction factor corresponding to the historical eating characteristic information. The second acquisition module 1004 is used to acquire the original nutrition database corresponding to the breed label and the age, and to acquire the inventory information of multiple independent feeding chambers configured in the pet feeder corresponding to the breed label and the target nutrient composition matrix of each feeding chamber. The determining module 1005 is used to determine the output quantity corresponding to each independent feeding bin based on the target energy, the original nutrient database, the inventory information of the multiple independent feeding bins and the target nutrient composition matrix of each feeding bin; The feeding module 1006 is used to control each of the independent feeding chambers to release the corresponding substance into the mixing chamber according to the discharge volume, mix the substances evenly with a preset rotation speed and time, control the discharge port to put the mixture into the food bowl for the pet to eat, and send the pet eating video captured by the camera to the client corresponding to the pet feeder through the mobile communication module for the user to view the feeding situation.

[0070] The pet feeder provided in this disclosure can execute the pet feeding processing method based on a mobile communication network provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects of the method.

[0071] To achieve the above embodiments, this disclosure also proposes a pet feeder, which is further equipped with a mobile chassis and a control system. The control system is configured to plan a movement path based on the pet's location information and control the pet feeder to move to the vicinity of the pet. The pet feeder also includes: The trigger module is used to play the user-preset feeding call voice in response to receiving a feeding instruction sent by the client through the mobile communication network, or in response to a trigger instruction for a feeding time preset on the pet feeder. The identity recognition module is specifically used to, if a pet is detected to have arrived at a preset area within a preset time, input the pet's facial image captured by the camera into a pre-trained first neural network to extract facial feature vectors, and input the pet's nose print image captured by the camera into a pre-trained second neural network to extract nose print feature vectors. The mobile control module is used to send abnormal pet feeding information to the client via the mobile communication module if the pet is not detected to have reached the preset area within a preset time, receive location information collected by the pet electronic collar sent by the client, control the pet feeder to find the pet according to the location information, collect video information of the pet and send it to the client, and prompt the user whether to feed the pet.

[0072] Furthermore, the energy generation module includes: The detection unit is used to input the health feature information, the activity feature information and the historical eating feature information into a pre-trained anomaly detection neural network. If the health feature information, the activity feature information and the historical eating feature information are all normal, the unit outputs the corresponding first correction factor, the second correction factor and the third correction factor. The first processing unit is used to output health abnormality information, and / or activity abnormality information, and / or eating abnormality information if one or more abnormal feature information is detected, and to generate a standardized consultation request carrying the pet's identity identifier, and send it to a preset cloud veterinary service platform through the mobile communication module. The receiving unit is used to receive treatment information returned by the veterinary service platform through the mobile communication network. The treatment information includes at least: the level of urgency, the feeding amount adjustment coefficient, and the type and dosage of therapeutic nutrients. The second processing unit is used to correct the initial energy according to the feeding amount adjustment coefficient to generate target energy if the detection finds that the urgency level is less than or equal to the preset threshold. The third processing unit is used to send the treatment information to the client through the mobile communication module and prompt the user whether to feed the patient if the detected urgency level is greater than a preset threshold.

[0073] The pet feeder provided in this disclosure can execute the pet feeding processing method based on a mobile communication network provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects of the method.

[0074] To implement the above embodiments, this disclosure also proposes a computer program product, including a computer program / instruction, which, when executed by a processor, implements the pet feeding processing method based on a mobile communication network executed by the pet feeder in the above embodiments.

[0075] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure.

[0076] The following is a detailed reference. Figure 5 The diagram illustrates a structural schematic suitable for implementing the electronic device 500 in the embodiments of this disclosure. The electronic device 500 in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein. Figure 5 The electronic device in the picture could be a pet feeder.

[0077] like Figure 5 As shown, electronic device 500 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from storage device 508 into random access memory (RAM) 503. RAM 503 also stores various programs and data required for the operation of electronic device 500. Processing unit 501, ROM 502, and RAM 503 are interconnected via bus 504. Input / output (I / O) interface 505 is also connected to bus 504.

[0078] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 508 including, for example, magnetic tapes, hard disks, etc.; and communication devices 509. Communication device 509 allows electronic device 500 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 5 An electronic device 500 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0079] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 509, or installed from storage device 508, or installed from ROM 502. When the computer program is executed by processing device 501, it performs the functions defined in the methods of embodiments of this disclosure.

[0080] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0081] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0082] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0083] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform a pet feeding process performed by the pet feeder.

[0084] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0085] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0086] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0087] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0088] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0089] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0090] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

Claims

1. A pet feeding processing method based on a mobile communication network, characterized in that, The method is applied to a pet feeder with a mobile communication module, the pet feeder including a camera and multiple independent feeding compartments, and the method includes: Using a pre-trained attention mechanism model, the fusion weight between the facial feature vector and the nasal print feature vector collected by the camera is calculated. Based on the fusion weight, the facial feature vector and the nasal print feature vector are processed to generate an identity feature vector. The identity feature vector is then matched with a pre-registered pet identity feature database to determine the pet's identity. The system retrieves pre-stored breed tags, age, weight, and health information corresponding to the pet's identification, as well as activity information collected by the electronic collar corresponding to the pet's identification and historical feeding information recorded by the pet feeder corresponding to the pet's identification. The age and weight are calculated to generate initial energy according to the preset growth energy formula corresponding to the variety label. The initial energy is then corrected to generate target energy according to the first correction factor corresponding to the health characteristic information, the second correction factor corresponding to the activity characteristic information, and the third correction factor corresponding to the historical eating characteristic information. Obtain the original nutrition database corresponding to the breed label and the age, and obtain the inventory information of multiple independent feeding chambers configured in the pet feeder corresponding to the breed label and the target nutrient composition matrix of the feeding chambers; Based on the target energy, the original nutrient database, the inventory information of the multiple independent feeding bins, and the target nutrient composition matrix of the feeding bins, the output quantity corresponding to each independent feeding bin is determined; Each independent feeding chamber is controlled to release the corresponding substance into the mixing chamber according to the discharge volume. The stirring mechanism mixes the substances evenly at a preset speed and time. The discharge port is controlled to deliver the mixture into the food bowl for the pet to eat. The video of the pet eating captured by the camera is sent to the client corresponding to the pet feeder through the mobile communication module so that the user can view the feeding situation.

2. The method according to claim 1, characterized in that, The pet feeder is also equipped with a mobile chassis and a control system. The control system is configured to plan a movement path based on the pet's location information and control the pet feeder to move to the vicinity of the pet. Before calculating the fusion weights between the facial feature vector and nasal print feature vector captured by the camera of the current pet using the pre-trained attention mechanism model, the method further includes: In response to receiving a feeding instruction sent by the client through the mobile communication network, or in response to a trigger instruction for a feeding time preset on the pet feeder, the user-preset feeding call voice is played; If a pet is detected to have reached a preset area within a preset time, the pet's facial image captured by the camera is input into a pre-trained first neural network to extract facial feature vectors, and the pet's nose print image captured by the camera is input into a pre-trained second neural network to extract nose print feature vectors. If the pet is not detected to have reached the preset area within the preset time, the mobile communication module sends a pet feeding abnormality information to the client, receives the location information collected by the pet electronic collar sent by the client, controls the pet feeder to find the pet according to the location information, collects the pet's video information and sends it to the client, and prompts the user whether to feed the pet.

3. The method according to claim 2, characterized in that, The step of correcting the initial energy to generate target energy based on a first correction factor corresponding to the health characteristic information, a second correction factor corresponding to the activity characteristic information, and a third correction factor corresponding to the historical eating characteristic information includes: The health feature information, the activity feature information, and the historical eating feature information are input into a pre-trained anomaly detection neural network. If no anomalies are detected in the health feature information, the activity feature information, and the historical eating feature information, the corresponding first correction factor, the second correction factor, and the third correction factor are output. If one or more abnormal feature information is detected, health abnormality information, and / or activity abnormality information, and / or eating abnormality information are output, and a standardized consultation request carrying the pet's identification is generated and sent to the preset cloud veterinary service platform through the mobile communication module. The veterinary service platform receives treatment information returned via the mobile communication network. The treatment information includes at least: the level of urgency, the feeding adjustment coefficient, and the type and dosage of therapeutic nutrients. If the detection determines that the urgency level is less than or equal to the preset threshold, the initial energy is corrected according to the feeding amount adjustment coefficient to generate the target energy; If the detection indicates that the urgency level is greater than a preset threshold, the treatment information is sent to the client via the mobile communication module, and the user is prompted whether to feed the patient.

4. The method according to claim 3, characterized in that, The step of determining the output quantity corresponding to each independent feeding bin based on the target energy, the original nutrient database, the inventory information of the multiple independent feeding bins, and the target nutrient composition matrix of each feeding bin includes: The original nutrient database is adjusted according to the type and dosage of the therapeutic nutrients to generate a target nutrient database; The current feeding energy is determined based on the target energy and feeding time, and the nutrient requirements are determined based on the current feeding energy and the target nutrient database. Based on the inventory information of the multiple independent feeding bins and the target nutrient composition matrix and material density of the feeding bins, an algorithm that approximates the nutrient requirement through minimum variance is used to determine the output quantity corresponding to each of the independent feeding bins.

5. The method according to any one of claims 1-4, characterized in that, Before the preset rotation speed and time are used to mix the substances evenly, the process further includes: The system detects whether the substance exiting the chamber is a preset temperature-sensitive substance. If it is, the temperature of the mixing chamber is adjusted to match the temperature-sensitive substance. After the mixture is dispensed into the food bowl for the pet from the controlled discharge port, the process further includes: The system collects data on actual food intake, pet eating speed, pet eating status, and pet activity data within a preset time period after eating, obtained through an electronic collar. This data is then input into a preset adaptive learning model to obtain the corresponding real-time nutrition database. Based on the real-time nutrition database, it is determined whether to adjust the original nutrition database. If adjustment is required, the adjustment information is sent to the client via the mobile communication module for the user to view.

6. A pet feeder, characterized in that, The pet feeder includes a camera, multiple independent feeding chambers, and a mixed feeding system. Each feeding chamber is equipped with an independent food dispensing mechanism and a food level sensor. The mixed feeding system includes a mixing chamber, a stirring mechanism, a dispensing port, and a food tray. The pet feeder also includes: The identity recognition module is used to calculate the fusion weight between the facial feature vector and the nasal print feature vector collected by the camera of the current pet through a pre-trained attention mechanism model, process the facial feature vector and the nasal print feature vector based on the fusion weight to generate an identity feature vector, and perform similarity matching between the identity feature vector and a pre-registered pet identity feature database to determine the pet's identity identifier. The first acquisition module is used to acquire pre-stored breed tags, age, weight and health characteristics information corresponding to the pet's identity, activity characteristics information collected by the electronic collar corresponding to the pet's identity, and historical feeding characteristics information recorded by the pet feeder corresponding to the pet's identity. An energy generation module is used to calculate and generate initial energy based on the age and weight according to a preset growth energy formula corresponding to the variety label, and to correct the initial energy to generate target energy based on a first correction factor corresponding to the health characteristic information, a second correction factor corresponding to the activity characteristic information, and a third correction factor corresponding to the historical eating characteristic information. The second acquisition module is used to acquire the original nutrition database corresponding to the breed label and the age, and to acquire the inventory information of multiple independent feeding chambers configured in the pet feeder corresponding to the breed label and the target nutrient composition matrix of each feeding chamber. The determination module is used to determine the output quantity corresponding to each independent feeding bin based on the target energy, the original nutrient database, the inventory information of the multiple independent feeding bins, and the target nutrient composition matrix of each feeding bin; The feeding module is used to control each of the independent feeding chambers to release the corresponding substance into the mixing chamber according to the discharge volume, mix the substances evenly with a preset rotation speed and time, and control the discharge port to put the mixture into the food bowl for the pet to eat. The module also sends the pet eating video captured by the camera to the client corresponding to the pet feeder through the mobile communication module so that the user can view the feeding situation.

7. The pet feeder according to claim 6, characterized in that, The pet feeder is also equipped with a mobile chassis and a control system. The control system is configured to plan a movement path based on the pet's location information and control the pet feeder to move to the vicinity of the pet. The pet feeder also includes: The trigger module is used to play the user-preset feeding call voice in response to receiving a feeding instruction sent by the client through the mobile communication network, or in response to a trigger instruction for a feeding time preset on the pet feeder. The identity recognition module is specifically used to, if a pet is detected to have arrived at a preset area within a preset time, input the pet's facial image captured by the camera into a pre-trained first neural network to extract facial feature vectors, and input the pet's nose print image captured by the camera into a pre-trained second neural network to extract nose print feature vectors. The mobile control module is used to send abnormal pet feeding information to the client via the mobile communication module if the pet is not detected to have reached the preset area within a preset time, receive location information collected by the pet electronic collar sent by the client, control the pet feeder to find the pet according to the location information, collect video information of the pet and send it to the client, and prompt the user whether to feed the pet.

8. The pet feeder according to claim 7, characterized in that, The energy generation module includes: The detection unit is used to input the health feature information, the activity feature information and the historical eating feature information into a pre-trained anomaly detection neural network. If the health feature information, the activity feature information and the historical eating feature information are all normal, the unit outputs the corresponding first correction factor, the second correction factor and the third correction factor. The first processing unit is used to output health abnormality information, and / or activity abnormality information, and / or eating abnormality information if one or more abnormal feature information is detected, and to generate a standardized consultation request carrying the pet's identity identifier, and send it to a preset cloud veterinary service platform through the mobile communication module. The receiving unit is used to receive treatment information returned by the veterinary service platform through the mobile communication network. The treatment information includes at least: the level of urgency, the feeding amount adjustment coefficient, and the type and dosage of therapeutic nutrients. The second processing unit is used to correct the initial energy according to the feeding amount adjustment coefficient to generate target energy if the detection finds that the urgency level is less than or equal to the preset threshold. The third processing unit is used to send the treatment information to the client through the mobile communication module and prompt the user whether to feed the patient if the detected urgency level is greater than a preset threshold.

9. An electronic device, characterized in that, The electronic device includes: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the pet feeding processing method based on a mobile communication network performed by the pet feeder according to any one of claims 1-5.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program for executing the pet feeding processing method based on a mobile communication network performed by any of the pet feeders described in claims 1-5.

Images