An automatic pet feeder and its remote control method
By collecting and analyzing pet image data in an automatic pet feeder, a status recognition model and a feeding amount knowledge graph are constructed, solving the problem of fixed feeding amount in existing technologies. This enables intelligent feeding control based on the individual pet's condition, preventing cross-infection of diseases and ensuring precise feeding.
Patent Information
- Application Number
- CN202511912772.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-18
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2045-12-18
AI Technical Summary
Existing automatic pet feeders fail to consider a pet's emotional state, hunger level, and age, resulting in a fixed amount of food that may lead to obesity or illness in pets. Furthermore, they cannot accurately control the feeding needs of multi-pet households.
By collecting pet image data through camera devices, analyzing their mental state, hunger state, and individual age, a pet status recognition model and feeding amount knowledge graph are constructed to achieve intelligent control of feeding amount. An automatic sensing function is installed on the feeder to turn it on only for designated pets to prevent cross-infection of diseases.
It enables dynamic adjustment of feeding amounts based on individual pet conditions, preventing cross-infection of diseases, meeting the precise feeding needs of multi-pet households, and improving the rationality and safety of feeding.
Smart Images

Figure CN122030283B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of pet feeders, and more particularly to an automatic pet feeder and a remote control method. Background Technology
[0002] Modern pet owners often face challenges in ensuring their pets eat on time due to busy work schedules or travel. To address this, smart automatic pet feeders have emerged. These devices use electronic control systems to automate feeding, solving the "food shortage" problem and helping pets establish regular eating habits. These devices typically consist of a food storage bin, a dispensing mechanism, a control module (Wi-Fi / Bluetooth), a weighing sensor, and an app. Some high-end models also integrate features like cameras, voice interaction, and UV sterilization. However, current automatic pet feeders often fail to consider the pet's emotional state, hunger level, or age. The fixed feeding amount can lead to obesity and disease in pets. Furthermore, these feeders often cannot be controlled to turn on only the desired pet. Summary of the Invention
[0003] This invention overcomes the shortcomings of the prior art and provides an automatic pet feeder and a remote control method.
[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: The first aspect of this invention provides an automatic pet feeder, comprising: The remote terminal and housing have a support frame inside the housing, and a plurality of feeding mechanisms and a camera device arranged in a linear array are installed on the support frame. The feeding mechanism includes a grain storage tank, a first drive motor is installed at the lower outlet of the grain storage tank, the output end of the first drive motor is connected to a first rotating shaft, a bearing is installed at the other end of the first rotating shaft, the bearing is fixed on the conveying cylinder, and a grain outlet pipe is provided below the conveying cylinder. The camera device collects image data of the pet, and by analyzing the image data, it obtains data on the pet's mental state, hunger state, age, and category. Based on the pet's mental state, hunger state, age, and category, it dynamically analyzes the required food data. The feeding mechanism is intelligently controlled based on the required grain data.
[0005] Furthermore, in the automatic pet feeder, a feeding frame is provided directly below the food dispensing tube, and a quality detection sensor is installed inside the feeding frame. The quality detection sensor acquires real-time quality data, and the feeding mechanism is intelligently controlled based on the required food data and the real-time quality data.
[0006] Furthermore, in the automatic pet feeder, a food storage lid is also installed on the top of the food storage container.
[0007] Furthermore, in the automatic pet feeder, a second drive motor is also installed on the top of the food storage tank. The output end of the second drive motor is connected to a second rotating shaft, and several stirring rods are installed on the second rotating shaft.
[0008] Furthermore, in the automatic pet feeder, several spiral blades are welded onto the first rotating shaft.
[0009] A second aspect of the present invention provides a remote control method for an automatic pet feeder, applicable to any of the automatic pet feeders described in any one of the claims, comprising the following steps: Collect image data on the mental state, hunger state, and individual age of pets of different types; A pet status recognition model is constructed based on the image data of the mental state, hunger state, and individual age of the pets of different pet types, and real-time pet image data is collected by the camera device on the automatic pet feeder. Based on the pet status recognition model, the model identifies the pet type, mental state, hunger state, and individual age of the pet in real-time pet image data, and constructs a knowledge graph of pet feeding amount. The system combines the pet feeding knowledge graph with real-time pet image data, including pet type, mental state, hunger status, and age, to generate the current pet feeding amount, and provides prompts and control via a remote terminal.
[0010] Furthermore, in the remote control method for the automatic pet feeder, a pet status recognition model is constructed based on the pet's mental state, hunger state, and individual age image data for different pet types. Specifically: Image data of the mental state, hunger state, and individual age of pets of different types are used as model inputs, and the mental state, hunger state, and individual age of pets are used as model outputs. A pet state recognition model is constructed based on deep learning. The learning rate, number of training iterations, and conditions for stopping training of the pet state recognition model are set. The pet state recognition model is trained based on the learning rate and number of training iterations. After reaching the specified number of training iterations, determine whether the real-time pet state recognition model has reached the conditions for stopping training. When the real-time pet state recognition model reaches the conditions for stopping training, stop training and save the model parameters of the pet state recognition model. After reaching the specified number of training iterations, if the real-time pet state recognition model does not meet the conditions for stopping training, the number of training iterations is increased, and the pet state recognition model continues to be trained.
[0011] Furthermore, in the remote control method for the automatic pet feeder, the pet's type, mental state, hunger state, and individual age are identified from real-time pet image data based on the pet state recognition model. Specifically: The real-time pet image data is input into the pet state recognition model for recognition, and the pet type, pet mental state, pet hunger state, and pet individual age status image data are obtained from the real-time pet image data. The output results include the pet's type, mental state, hunger state, and individual age from the real-time pet image data.
[0012] Furthermore, in the remote control method for automatic pet feeders, a knowledge graph of pet feeding amounts is constructed, specifically as follows: The optimal feeding amount for different pet types under the conditions of pet's mental state, hunger state, and individual age is determined, and the pet type, pet's mental state, pet's hunger state, and pet's individual age are used as nodes in the first graph. Using the optimal feeding amount as a node in the second graph, a directed descriptive relationship is constructed. Based on the directed descriptive relationship, the nodes in the first graph and the nodes in the second graph are connected to form a topological structure graph, thus constructing a knowledge graph of pet feeding amount. The topological structure graph is input into the pet feeding amount knowledge graph for knowledge representation, and the pet feeding amount knowledge graph is constructed.
[0013] Furthermore, in the remote control method for the automatic pet feeder, the current pet feeding amount is generated by combining the pet feeding knowledge graph with real-time pet image data, including the pet's type, mental state, hunger status, and age. This information is then controlled via a remote terminal. Specifically: The pet type, mental state, hunger state, and individual age of the pet in the real-time pet image data are input into the pet feeding knowledge graph for knowledge retrieval. By querying, we can obtain real-time pet image data, including the pet's type, mental state, hunger level, and optimal feeding amount for the pet's individual age. The current feeding amount for the pet is generated based on the pet's type, mental state, hunger state, and the optimal feeding amount for the pet's age in the real-time pet image data. A prompt message is generated based on the current amount of food the pet is fed, and sent to a remote terminal, which then performs prompt control based on the prompt message.
[0014] This invention addresses the shortcomings of the prior art and has the following beneficial effects: This invention places pet food in a food storage container. When the food needs to be removed, a first drive motor drives a first rotating shaft, which uses blades to transport the food from the storage container to a feeding frame. A camera captures images of the pets, and this data is analyzed to obtain information on the pet's mental state, hunger level, age, and pet type. Based on this data, the required food quantity is dynamically determined, and the feeding mechanism is intelligently controlled. Each food dispensing port can also be equipped with an automatic sensor, opening only for designated pets and closing when the wrong pet is present. The port closes promptly after each pet finishes eating to prevent uneaten food from being consumed by the wrong pet, thus preventing cross-infection of diseases. Furthermore, it allows for separate control of the food dispensing amount and frequency per day, meeting the needs of multi-pet households for precise feeding. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.
[0016] Figure 1 A schematic diagram of the overall structure of the automatic pet feeder is shown. Figure 2 A partial structural diagram of an automatic pet feeder is shown; Figure 3 A cross-sectional structural diagram of an automatic pet feeder is shown. Detailed Implementation
[0017] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner. Therefore, they only show the components related to the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.
[0018] In the description of this application, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the scope of protection of this application. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.
[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this application based on the specific circumstances.
[0020] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.
[0021] like Figures 1 to 3 As shown, the first aspect of the present invention provides an automatic pet feeder, comprising: The remote terminal (computer terminal, mobile phone terminal, etc., which can be connected by installing communication equipment in the pet automatic feeder), the housing 1, the support frame 2 is provided inside the housing 1, and a plurality of feeding mechanisms 3 arranged in a linear array and a camera device 4 are installed on the support frame 2. The feeding mechanism 3 includes a grain storage tank 301. A first drive motor 302 is installed at the lower outlet of the grain storage tank 301. The output end of the first drive motor 302 is connected to a first rotating shaft 303. A bearing 304 is installed at the other end of the first rotating shaft 303. The bearing 304 is fixed on the conveying cylinder 305. A grain outlet pipe 306 is provided below the conveying cylinder 305. It should be noted that when a user places pet food into the food storage container 301, and needs to remove the pet food, the user drives the first drive motor 302, which in turn drives the first rotating shaft 303. The rotating blades 311 on the first rotating shaft 303 then transport the food from the food storage container 301 to the feeding frame 307.
[0022] It should be noted that the shape and number of feeding frames 307 are not shown in detail in this embodiment. One or more feeding frames 307 can be set below each feed pipe 306. The shape of the feeding frame 307 can be circular, square, oval, etc., and those skilled in the art can make free adjustments according to the actual situation.
[0023] The camera device 4 collects image data of the pet, and analyzes the image data to obtain data on the pet's mental state, hunger state, age, and category. Based on the pet's mental state, hunger state, age, and category, it dynamically analyzes the required food data. The feeding mechanism 3 is intelligently controlled based on the required grain data.
[0024] It's worth noting that each food dispenser can also be equipped with an automatic sensor function, opening only for designated pets and remaining closed for incorrectly designated pets. The sensor closes promptly after each pet finishes eating to prevent uneaten food from being consumed by the wrong pet, thus preventing cross-infection of diseases among pets. Furthermore, it allows for separate control of the food dispensing amount and the frequency of dispensing per day, meeting the needs of multi-pet households requiring precise feeding.
[0025] Furthermore, in the automatic pet feeder, a feeding frame 307 is provided directly below the food dispensing pipe 306, and a quality detection sensor is provided inside the feeding frame 307. The quality detection sensor obtains real-time quality data, and the feeding mechanism is intelligently controlled based on the required food data and the real-time quality data.
[0026] The feeding box 307 is not just designed as shown in the figure. The feeding box 307 can be set as a remote control switch, such as setting an opening and closing door on the feeding box 307. This can prevent the feeding box from being eaten by the wrong pet after each pet has finished eating.
[0027] It should be noted that by analyzing the required grain data based on dynamic analysis, the first drive motor 302 is driven, which in turn drives the first rotating shaft 303. The blades 311 on the first rotating shaft 303 are used to transport the grain from the grain storage tank 301 to the feeding frame 307. When the required grain data is reached, the feeding mechanism 3 stops working.
[0028] Furthermore, in the automatic pet feeder, a food storage lid 312 is also installed on the top of the food storage container 301.
[0029] Furthermore, in the automatic pet feeder, a second drive motor 308 is also installed on the top of the food storage tank 301. The output end of the second drive motor 308 is connected to a second rotating shaft 309, and a plurality of stirring rods 310 are installed on the second rotating shaft 309.
[0030] It should be noted that by driving the second drive motor 308, the second rotating shaft 309 can be driven, and the grain inside can be stirred by the stirring rod 310 to achieve the mixing effect. This effect can be provided when feeding two or more kinds of grain.
[0031] Furthermore, in the automatic pet feeder, a plurality of spiral blades 311 are welded onto the first rotating shaft 303.
[0032] A second aspect of the present invention provides a remote control method for an automatic pet feeder, applicable to any of the automatic pet feeders described in any one of the claims, comprising the following steps: Collect image data on the mental state, hunger state, and individual age of pets of different types; A pet status recognition model is constructed based on the image data of the mental state, hunger state, and individual age of the pets of different pet types, and real-time pet image data is collected by the camera device on the automatic pet feeder. Based on the pet status recognition model, the model identifies the pet type, mental state, hunger state, and individual age of the pet in real-time pet image data, and constructs a knowledge graph of pet feeding amount. The system combines the pet feeding knowledge graph with real-time pet image data, including pet type, mental state, hunger status, and age, to generate the current pet feeding amount, and provides prompts and control via a remote terminal.
[0033] It should be noted that the amount of food a pet needs varies depending on its type, its mental state, its hunger level, and its age. This method allows for tailored feeding practices based on these specific circumstances. Furthermore, in the remote control method for the automatic pet feeder, a pet status recognition model is constructed based on the pet's mental state, hunger state, and individual age image data for different pet types. Specifically: Image data of the mental state, hunger state, and individual age of pets of different types are used as model inputs, and the mental state, hunger state, and individual age of pets are used as model outputs. A pet state recognition model is constructed based on deep learning. The learning rate, number of training iterations, and conditions for stopping training of the pet state recognition model are set. The pet state recognition model is trained based on the learning rate and number of training iterations. After reaching the specified number of training iterations, determine whether the real-time pet state recognition model has reached the conditions for stopping training. When the real-time pet state recognition model reaches the conditions for stopping training, stop training and save the model parameters of the pet state recognition model. After reaching the specified number of training iterations, if the real-time pet state recognition model does not meet the conditions for stopping training, the number of training iterations is increased, and the pet state recognition model continues to be trained.
[0034] It should be noted that the conditions for stopping training include parameters such as prediction accuracy, model parameters, and loss function convergence value. This method can be used to build a pet state recognition model to identify the real-time state of pets.
[0035] Furthermore, in the remote control method for the automatic pet feeder, the pet's type, mental state, hunger state, and individual age are identified from real-time pet image data based on the pet state recognition model. Specifically: The real-time pet image data is input into the pet state recognition model for recognition, and the pet type, pet mental state, pet hunger state, and pet individual age status image data are obtained from the real-time pet image data. The output results include the pet's type, mental state, hunger state, and individual age from the real-time pet image data.
[0036] Furthermore, in the remote control method for automatic pet feeders, a knowledge graph of pet feeding amounts is constructed, specifically as follows: The optimal feeding amount for different pet types under the conditions of pet's mental state, hunger state, and individual age is determined, and the pet type, pet's mental state, pet's hunger state, and pet's individual age are used as nodes in the first graph. Using the optimal feeding amount as a node in the second graph, a directed descriptive relationship is constructed. Based on the directed descriptive relationship, the nodes in the first graph and the nodes in the second graph are connected to form a topological structure graph, thus constructing a knowledge graph of pet feeding amount. The topological structure graph is input into the pet feeding amount knowledge graph for knowledge representation, and the pet feeding amount knowledge graph is constructed.
[0037] It should be noted that this method can construct a knowledge graph of pet feeding amounts to predict the optimal feeding amount for the same pet type under different conditions, such as the pet's mental state, hunger state, and individual age, thereby improving the rationality of feeding.
[0038] Furthermore, in the remote control method for the automatic pet feeder, the current pet feeding amount is generated by combining the pet feeding knowledge graph with real-time pet image data, including the pet's type, mental state, hunger status, and age. This information is then controlled via a remote terminal. Specifically: The pet type, mental state, hunger state, and individual age of the pet in the real-time pet image data are input into the pet feeding knowledge graph for knowledge retrieval. By querying, we can obtain real-time pet image data, including the pet's type, mental state, hunger level, and optimal feeding amount for the pet's individual age. The current feeding amount for the pet is generated based on the pet's type, mental state, hunger state, and the optimal feeding amount for the pet's age in the real-time pet image data. A prompt message is generated based on the current amount of food the pet is fed, and sent to a remote terminal, which then performs prompt control based on the prompt message.
[0039] It should be noted that this method allows for remote control of automatic pet feeders, enabling users to experience the joy of feeding their pets.
[0040] In addition, this method also includes: Speakers are installed in automatic pet feeders, and cameras are used to capture the pet's expression, ear position, tail wagging amplitude, and body posture. A temporal convolutional network is introduced to analyze the pet's expression, ear position, tail wagging amplitude, and body posture. Through analysis, three states—anxiety, excitement, and normal—are identified. A high-sensitivity MEMS microphone is integrated to collect pet vocalizations. By classifying the pet vocalizations, hunger, attention-seeking, and pain are identified. The emotional state is then comprehensively judged by combining the emotional state, hunger, attention-seeking, and pain. When the system detects that the pet is anxious, it can automatically play a pre-recorded soothing voice message or calming music from the owner, with the volume automatically adjusted according to the ambient noise. Based on emotion recognition results, for mild anxiety, an audio soothing mode is activated, playing the owner's voice or soothing music at a specific frequency. For moderate anxiety, a brief interactive game is triggered using the laser projection device. For persistent anxiety, the system can release a fixed amount of snacks as positive reinforcement. The laser projection device has a wavelength of 650nm and is Class II safety grade.
[0041] It's important to note that traditional pet feeders lack emotional interaction features, making it difficult to meet the emotional needs of pets. This system alleviates separation anxiety and enhances the emotional bond between humans and pets through multimodal emotional computing and remote interaction. Owners can remotely view their pet's status and communicate via voice through an app. The feeder has a built-in high-quality speaker (frequency response 100Hz-16kHz) and a noise-canceling microphone array, using audio beamforming technology to suppress ambient noise. For unstable network conditions, the system employs adaptive bitrate adjustment (adjustable from 150kbps to 1.2Mbps) to ensure uninterrupted basic voice communication. When pet anxiety is detected, the system automatically plays pre-recorded soothing voice messages or calming music, with the volume automatically adjusting to ambient noise. Furthermore, the system captures the pet's expression, ear position, tail wagging amplitude, and body posture through a camera. Temporal convolutional networks are used to analyze continuous behavioral sequences, identifying three emotional states: anxiety (e.g., continuous pacing, excessive grooming), excitement (e.g., high-frequency tail vibration, staring at the feeder), and normal. Meanwhile, a high-sensitivity MEMS microphone is integrated to collect pet sounds, and a pre-trained audio classification model (based on ResNet architecture) is used to distinguish categories (such as hunger, seeking attention, pain) and comprehensively judge the emotional state.
[0042] In addition, this method also includes: Based on digital twin technology, a full life cycle digital profile is created for each pet, continuously collecting data on diet, weight, and activity levels. Long short-term memory networks are used to analyze time-series data, establish a correlation model between eating habits, weight changes, and health indicators, and form a digital twin. The digital twin is automatically updated daily. When a significant deviation is detected that exceeds the preset deviation value, the update frequency of the digital twin is increased to obtain comprehensive data on the pet's exercise volume and excretion frequency. This comprehensive data is then used to assess the risk of obesity and indigestion. Based on the aforementioned risks of obesity and indigestion, a reinforcement learning algorithm is used, with pet body shape scores and blood indicators as reward signals, to optimize feeding frequency and food quantity. When the risk of obesity and indigestion exceeds the preset risk index, the feeding frequency and amount are optimized; when the risk of obesity and indigestion exceeds the preset risk index, the current feeding frequency and amount are maintained.
[0043] It should be noted that the risks of obesity and indigestion include low, medium, and high risks. This method enables a shift from passive feeding to proactive health management by creating a virtual mapping model of the pet, providing personalized dietary adjustments and health risk warnings. The digital twin updates automatically daily. When a significant deviation is detected, such as a sudden 30% reduction in food intake, a high-frequency data collection mode is activated, increasing from once every 12 hours to once every 2 hours. When the system detects simultaneous decrease in water intake and weight loss, a three-level warning signal is generated and pushed to the owner via the app. Furthermore, a reinforcement learning algorithm is used, with the pet's body shape score and blood indicators as reward signals, to optimize feeding frequency and amount. For pets prone to obesity, the system implements a "slow weight loss plan," gradually reducing daily food intake over four weeks (e.g., 5% per week) while increasing the proportion of dietary fiber to avoid stress reactions caused by sudden dieting. The system can also interface with veterinary platforms to automatically adjust the prescription diet ratio for pets with chronic diseases.
[0044] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.
[0045] The embodiments described above are merely specific implementations of this application, used to illustrate the technical solutions of this application, and are not intended to limit it. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.
Claims
1. A remote control method for an automatic pet feeder, characterized in that, Includes the following steps: Collect image data on the mental state, hunger state, and individual age of pets of different types; A pet status recognition model is constructed based on image data of pets' mental state, hunger state, and individual age status for different pet types, and real-time pet image data is collected through a camera device on an automatic pet feeder. Based on the pet status recognition model, the model identifies the pet type, mental state, hunger state, and individual age of the pet in real-time pet image data, and constructs a knowledge graph of pet feeding amount. The system combines the pet feeding knowledge graph with real-time pet image data, including pet type, mental state, hunger status, and age, to generate the current pet feeding amount, and provides prompts and control via a remote terminal. Constructing a knowledge graph of pet feeding amounts, specifically: The optimal feeding amount for different pet types under the conditions of pet's mental state, hunger state, and individual age is determined, and the pet type, pet's mental state, pet's hunger state, and pet's individual age are used as nodes in the first graph. Using the optimal feeding amount as a node in the second graph, a directed descriptive relationship is constructed. Based on the directed descriptive relationship, the nodes in the first graph and the nodes in the second graph are connected to form a topological structure graph, thus constructing a knowledge graph of pet feeding amount. The topology graph is input into the pet feeding amount knowledge graph for knowledge representation, and the pet feeding amount knowledge graph is constructed. Among them, automatic pet feeders include: The remote terminal and housing have a support frame inside the housing, and a plurality of feeding mechanisms and a camera device arranged in a linear array are installed on the support frame. The feeding mechanism includes a grain storage tank, a first drive motor is installed at the lower outlet of the grain storage tank, the output end of the first drive motor is connected to a first rotating shaft, a bearing is installed at the other end of the first rotating shaft, the bearing is fixed on the conveying cylinder, and a grain outlet pipe is provided below the conveying cylinder. The camera device collects image data of the pet, and by analyzing the image data, it obtains data on the pet's mental state, hunger state, age, and category. Based on the pet's mental state, hunger state, age, and category, it dynamically analyzes the required food data. The feeding mechanism is intelligently controlled based on the required grain data.
2. The remote control method for an automatic pet feeder according to claim 1, characterized in that, A feeding frame is installed directly below the grain outlet pipe, and a quality detection sensor is installed inside the feeding frame. The quality detection sensor acquires real-time quality data, and the feeding mechanism is intelligently controlled based on the required grain data and the real-time quality data.
3. The remote control method for an automatic pet feeder according to claim 1, characterized in that, The grain storage tank is also equipped with a grain storage cover on top.
4. The remote control method for an automatic pet feeder according to claim 1, characterized in that, A second drive motor is also installed on the top of the grain storage tank. The output end of the second drive motor is connected to a second rotating shaft, and several stirring rods are installed on the second rotating shaft.
5. The remote control method for an automatic pet feeder according to claim 1, characterized in that, Several spiral blades are welded onto the first rotating shaft.
6. The remote control method for an automatic pet feeder according to claim 1, characterized in that, A pet status recognition model is constructed based on image data of pets' mental state, hunger state, and individual age, for different pet types. Specifically: Image data of the mental state, hunger state, and individual age of pets of different types are used as model inputs, and the mental state, hunger state, and individual age of pets are used as model outputs. A pet state recognition model is constructed based on deep learning. The learning rate, number of training iterations, and conditions for stopping training of the pet state recognition model are set. The pet state recognition model is trained based on the learning rate and number of training iterations. After reaching the specified number of training iterations, determine whether the real-time pet state recognition model has reached the conditions for stopping training. When the real-time pet state recognition model reaches the conditions for stopping training, stop training and save the model parameters of the pet state recognition model. After reaching the specified number of training iterations, if the real-time pet state recognition model does not meet the conditions for stopping training, the number of training iterations is increased, and the pet state recognition model continues to be trained.
7. The remote control method for an automatic pet feeder according to claim 1, characterized in that, Based on the pet status recognition model, the model identifies the pet type, mental state, hunger state, and individual age of the pet in real-time pet image data. Specifically: The real-time pet image data is input into the pet state recognition model for recognition, and the pet type, pet mental state, pet hunger state, and pet individual age status image data are obtained from the real-time pet image data. The output results include the pet's type, mental state, hunger state, and individual age from the real-time pet image data.
8. The remote control method for an automatic pet feeder according to claim 1, characterized in that, The system combines the pet feeding knowledge graph with real-time pet image data, including pet type, mental state, hunger status, and age, to generate the current pet's feeding amount. This information is then controlled via a remote terminal. The pet type, mental state, hunger state, and individual age of the pet in the real-time pet image data are input into the pet feeding knowledge graph for knowledge retrieval. By querying, we can obtain real-time pet image data, including the pet's type, mental state, hunger level, and optimal feeding amount for the pet's individual age. The current feeding amount for the pet is generated based on the pet's type, mental state, hunger state, and the optimal feeding amount for the pet's age in the real-time pet image data. A prompt message is generated based on the current amount of food the pet is fed, and sent to a remote terminal, which then performs prompt control based on the prompt message.
Citation Information
Patent Citations
Automatic feeder for pets
CN107410065A
Automatic feeding device based on pet face recognition
CN112167074A
Intelligent multi-pet feeder based on image recognition and intelligent feeding system
CN116491436A
