Intelligent pet trainer control method and related equipment

Abstract

The invention belongs to the technical field of pet training equipment, and relates to an intelligent pet trainer control method and related equipment, and the method comprises the steps: collecting environment data and motion state data of a pet; selecting a target collection resource of the trainer according to the environment data and the motion state data; adjusting performance parameters of the trainer according to the target collection resources to obtain the performance parameters; identifying behavior intention data of the pet, and obtaining a personalized training strategy of the pet through the behavior intention data and a preset behavior prediction model; and adjusting communication parameters and positioning parameters of the trainer according to the performance parameters and the personalized training strategy. According to the invention, the endurance of equipment is prolonged through multi-source energy collection and efficiency adjustment, and the problem of frequent charging is solved. And a personalized strategy is generated by combining the individual behavior characteristics of the pet and the real-time state, so that stress response is reduced. Meanwhile, communication and positioning parameters are dynamically adjusted, energy consumption and function requirements are balanced, the communication positioning reliability in a complex environment is improved, and the equipment intelligence and the use effect are optimized.

Description

Technical Field

[0001] This application relates to the field of pet training equipment technology, and in particular to a control method and related equipment for an intelligent pet trainer. Background Technology

[0002] Existing pet training and control methods have significant shortcomings in technological application. On the one hand, in terms of personalized training, training strategies are relatively rigid, mostly based on a passive logic of triggering punishment when boundaries are crossed. They fail to dynamically adjust training programs based on the pet's behavioral characteristics and physiological state, making it impossible to adapt to individual differences among pets and easily leading to poor training results or stress reactions in pets. On the other hand, in terms of low power consumption and battery life, the energy supply method is singular, mainly relying on lithium battery charging. Battery life is limited by battery capacity, requiring frequent charging in scenarios such as long-term outdoor training or indoor low-light conditions. Furthermore, the lack of dynamic optimization of energy harvesting methods and device power consumption allocation based on environmental conditions (such as light intensity) and the pet's activity status further exacerbates the battery life shortcoming and affects the continuous use of the device.

[0003] In view of the above, this application is hereby submitted. Summary of the Invention

[0004] The purpose of this application is to propose a control method, device, computer equipment, and storage medium for an intelligent pet trainer, so as to realize personalized training strategies and multiple power supply methods for pets in different scenarios, and solve the problems of single energy supply, high power consumption, and rigid training strategies.

[0005] To address the aforementioned technical problems, this application provides a control method for an intelligent pet trainer, employing the following technical solution: A method for controlling an intelligent pet trainer includes the following steps: Collect environmental data and pet movement data; Based on the environmental data and the motion state data, the target acquisition resources of the trainer are adjusted, wherein the target acquisition resources include light energy resources, mechanical energy resources and thermal energy resources; Based on the target acquired resources, the performance parameters of the trainer are adjusted to obtain the target performance parameters; Identify the pet's behavioral intention data, and predict the pet's behavioral intention using the behavioral intention data and a preset behavior prediction model to obtain a personalized training strategy for the pet. The communication and positioning parameters of the trainer are adjusted according to the target performance parameters and the personalized training strategy.

[0006] Furthermore, adjusting the target acquisition resources of the trainer based on the environmental data and the motion state data includes: When the light intensity of the environmental data is greater than or equal to a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a first acquisition ratio, wherein the acquisition ratio of the light energy resources is greater than the acquisition ratio of the mechanical energy resources, and the acquisition ratio of the light energy resources is greater than the acquisition ratio of the thermal energy resources. When the light intensity of the environmental data is less than a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a second acquisition ratio, wherein the sum of the acquisition ratio of the mechanical energy resources and the acquisition ratio of the thermal energy resources in the second acquisition ratio is greater than the acquisition ratio of the light energy resources.

[0007] Furthermore, the aforementioned performance parameters include the magnetic damping coefficient and the operating voltage. The step of adjusting the performance parameters of the trainer based on the target acquisition resources to obtain the target performance parameters includes: Based on the mechanical energy resources in the target acquisition resources, the magnetic damping coefficient of the trainer is adjusted to obtain the first performance parameter; Based on the light energy resources in the target acquisition resources, the operating voltage of the trainer is adjusted to obtain the second performance parameter; The target performance parameters are obtained based on the first performance parameter and the second performance parameter.

[0008] Furthermore, the aforementioned behavioral intent data includes pet posture images, pet vocal data, and pet physiological data; Identify the pet's behavioral intention data, and using the behavioral intention data and a preset behavior prediction model, predict the pet's behavioral intention to obtain a personalized training strategy for the pet, including: Extract the visual features of the pet posture image, the acoustic features of the pet sound data, and the physiological features of the pet physiological data; The visual features, acoustic features, and physiological features are fused into a behavioral feature vector; The behavioral feature vector is input into the preset behavior prediction model to predict the pet's behavioral intentions; Based on the stated behavioral intent, a personalized training strategy is generated for the pet.

[0009] Furthermore, the aforementioned behavioral intentions include boundary-crossing intentions and stress response intentions. The step of predicting the pet's behavioral intentions using the behavioral intention data and a preset behavioral prediction model to obtain the pet's personalized training strategy includes: If the pet is predicted to have the intention to cross the boundary, a training strategy containing warning and punishment instructions is generated. If the pet is predicted to have the stress intention, a soothing strategy is generated to suspend punishment or reduce the intensity of training.

[0010] Furthermore, adjusting the communication and positioning parameters of the trainer based on the target performance parameters and the personalized training strategy includes: Based on the target performance parameters, determine the current available energy of the trainer; Based on the personalized training strategy, the positioning update frequency and communication response requirements are determined; The communication and positioning parameters of the trainer are adjusted based on the current available energy, the positioning update frequency, and the communication response requirements.

[0011] Furthermore, if there are multiple pets, the personalized training strategy for each pet obtained through the behavioral intention data and the preset behavioral prediction model includes: Collect data on the interaction behavior among multiple pets. Based on the interactive behavior data, the behavioral intent data, and the preset behavior prediction model, personalized training strategies are generated for each of the pets.

[0012] To address the aforementioned technical problems, this application also provides an intelligent pet trainer control device, which employs the following technical solution: A smart pet trainer control device includes: The data acquisition module is used to collect environmental data and pet movement data; An energy acquisition determination module is used to adjust the target acquisition resources of the trainer based on the environmental data and the motion state data, wherein the target acquisition resources include light energy resources, mechanical energy resources and thermal energy resources; An energy supply determination module is used to adjust the performance parameters of the trainer based on the target acquisition resources to obtain the target performance parameters; The strategy module is used to identify the pet's behavioral intention data, and predict the pet's behavioral intentions using the behavioral intention data and a preset behavior prediction model, thereby obtaining a personalized training strategy for the pet. The communication and positioning module is used to adjust the communication and positioning parameters of the trainer according to the target performance parameters and the personalized training strategy.

[0013] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution: A computer device includes a memory and a processor, the memory storing computer-readable instructions, the processor executing the computer-readable instructions to implement the steps of the intelligent pet trainer control method as described above.

[0014] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below: A computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of the intelligent pet trainer control method as described above.

[0015] Compared with the prior art, the embodiments of this application have the following main advantages: The intelligent pet trainer control method disclosed in this application effectively overcomes the limitations of existing pet training equipment that relies on a single energy source by collecting environmental data, pet movement status data, and behavioral intention data, combined with the determination of multi-source energy sources. This improves energy utilization efficiency, solves the problem of frequent charging in scenarios such as long-term outdoor training and low-light indoor environments, and extends the device's battery life. Furthermore, it generates personalized training strategies based on behavioral intention data and a preset behavior prediction model, breaking away from traditional rigid training logic. This adapts to the individual behavioral characteristics and real-time states of pets (such as boundary crossing intentions and stress states), helping to improve pet command obedience rates and reduce the incidence of stress reactions. In addition, by dynamically adjusting communication and positioning parameters according to target performance parameters and personalized training strategies, a precise balance between energy consumption and functional requirements can be achieved. This avoids functional waste during high battery periods and prevents core function failure during low battery periods, while also improving the reliability of communication and positioning in complex environments, thus optimizing the intelligence level and actual usage effect of the pet training equipment. Attached Figure Description

[0016] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is an exemplary system architecture diagram provided in this application that can be applied thereto; Figure 2 This is a flowchart of a control method for an intelligent pet trainer provided in this application; Figure 3 This is a diagram showing the positioning and communication parameters of a control method for an intelligent pet trainer provided in this application. Figure 4This is a schematic diagram of the structure of a smart pet trainer control device provided in this application; Figure 5 This is a schematic diagram of the structure of a computer device provided in this application. Detailed Implementation

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0021] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0022] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0023] Terminal devices 101, 102, and 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptops, and desktop computers, etc.

[0024] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on terminal devices 101, 102, and 103.

[0025] It should be noted that the intelligent pet trainer control method provided in this application embodiment is generally executed by a terminal device, and correspondingly, the intelligent pet trainer control device is generally installed in the terminal device.

[0026] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0027] Continue to refer to Figure 2 A flowchart of an embodiment of the intelligent pet trainer control method according to this application is shown. The intelligent pet trainer control method includes the following steps: Step S201: Collect environmental data and pet movement status data.

[0028] In this embodiment, the intelligent pet trainer control method operates on an electronic device (e.g., Figure 1 The terminal device shown can send or receive data via wired or wireless connection. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultrawideband) connections, and other currently known or future wireless connection methods.

[0029] Traditional energy technologies are disconnected from the environment, relying solely on lithium batteries without adjusting energy harvesting methods based on ambient light and temperature. This leads to wasted energy in bright outdoor conditions and insufficient battery life in low-light indoor environments. Furthermore, training strategies are out of sync with the pet's condition, relying solely on passive punishment for boundary crossings without considering the pet's activity intensity or behavioral intent (e.g., whether the approach to the enclosure was unintentional). This results in rigid training and frequent stress responses. Therefore, it is necessary to acquire environmental data, pet activity data, and behavioral intent data, match them with appropriate energy harvesting methods, and generate training strategies tailored to the individual pet's condition, avoiding blind training without data support.

[0030] In this embodiment, two types of data need to be collected: the physical parameters of the external environment in which the trainer is located (environmental data) and the dynamic characteristics and behavioral tendencies of the pet itself (movement state data).

[0031] Specifically, environmental data can be collected through the environmental perception module on the trainer. The module may include a light sensor (to obtain ambient light intensity), a temperature and humidity sensor (to obtain ambient temperature and humidity), an electromagnetic environment sensor (to obtain the intensity of surrounding electromagnetic interference), etc. The collection dimensions include key environmental parameters that "affect energy collection efficiency" and "affect communication and positioning reliability".

[0032] Motion status data can be collected through the motion detection module on the trainer. The module may include a three-axis accelerometer and a gyroscope. The three-axis accelerometer is used to detect the pet's motion acceleration and step frequency, and the gyroscope is used to detect the pet's movement direction, posture changes, etc., thereby obtaining the pet's motion status characteristics such as "stationary / active", "activity intensity", and "movement direction".

[0033] Step S202: Adjust the target acquisition resources of the trainer according to the environmental data and the motion state data, wherein the target acquisition resources include light energy resources, mechanical energy resources and thermal energy resources.

[0034] It should be noted that solar energy resources refer to energy forms that can be obtained by the trainer from ambient light through a compatible energy harvesting unit and converted into electrical energy, such as solar energy. Its availability is directly related to the intensity of ambient light and can serve as the trainer's primary or secondary energy source. Mechanical energy resources refer to energy forms that can be obtained by the trainer from the pet's movement processes (such as walking, running, and activity) through a compatible energy harvesting unit and converted into electrical energy. Its availability is directly related to the pet's activity state (activity intensity, frequency, etc.) and belongs to movement-related energy. Thermal energy resources refer to energy forms that can be obtained by the trainer from the temperature difference between the pet's own temperature and the ambient temperature through a compatible energy harvesting unit and converted into electrical energy. Its availability is directly related to the magnitude of the temperature difference between the pet and the environment and belongs to temperature difference-driven energy.

[0035] During the use of the trainer, if relying solely on batteries, the battery life is limited by battery capacity. Frequent charging is necessary for long-term outdoor training, and there is no additional energy supply in low-light indoor environments. The three energy harvesting methods each have their own scenario adaptability: solar energy relies on sunlight (highly efficient in bright outdoor light), mechanical energy relies on pet movement (highly efficient when the pet is active), and thermal energy relies on temperature differences (the pet's body temperature is stable compared to the ambient temperature, unaffected by sunlight or movement). Combining energy harvesting methods can achieve scenario complementarity. For example, in bright outdoor light scenarios, solar and mechanical energy resources can be used, taking advantage of both sunlight and pet movement. In low-light indoor scenarios, mechanical and thermal energy resources can be used, utilizing pet movement and stable temperature differences to address the problem of insufficient solar energy supply.

[0036] In this embodiment, the current scene type is determined based on environmental data. The feasibility of light energy resources can be determined by light intensity, and the feasibility of thermal energy resources can be determined by ambient temperature. Taking light intensity as an example, it can be divided into scenes with sufficient light (such as sunny days outdoors) and scenes with insufficient light (such as indoors or cloudy days). Taking ambient temperature as an example, it can be divided into scenes with significant temperature differences (such as large temperature differences between indoors and outdoors in winter) or scenes with small temperature differences (being in one scene for a long time with little temperature change).

[0037] Judging a pet's activity level based on its movement data can be done by measuring movement intensity, such as cadence and acceleration. For example, three activity levels can be set: high, medium, and low. High activity level (pet running, frequent walking), medium-low activity level (pet walking slowly, occasionally still), and low activity level (pet remaining still for extended periods).

[0038] Based on the current scenario and pet activity level, select energy sources to ensure a stable and sufficient energy supply. For example, in scenarios with ample sunlight and high pet activity, prioritize solar and mechanical energy resources to increase energy utilization. In scenarios with insufficient sunlight and high pet activity, prioritize thermal and mechanical energy resources.

[0039] Step S203: Based on the target acquisition resources, adjust the performance parameters of the trainer to obtain the target performance parameters.

[0040] Even with a determined multi-source energy acquisition combination, energy waste can still occur if performance parameters are not adjusted. For example, when acquiring solar energy, fixed operating parameters (such as a fixed operating voltage) under different lighting conditions will result in the inefficient use of solar energy. Similarly, when acquiring mechanical energy, fixed mechanical parameters (such as a fixed magnetic damping coefficient) under different motion intensities will lead to low mechanical energy conversion efficiency. Furthermore, without monitoring the energy supply status, it's impossible to determine current energy sufficiency, causing subsequent communication and positioning modules to consume power unnecessarily (e.g., continuing high-frequency positioning even with low battery). Therefore, it is necessary to adjust the performance parameters of the trainer to improve energy utilization.

[0041] In this embodiment, the energy conversion efficiency is adjusted for each energy harvesting method in the energy harvesting combination. When harvesting solar energy resources, operating parameters (such as operating voltage and current) can be adjusted using technologies such as power tracking based on the light intensity in the environmental data to ensure maximum light energy conversion efficiency. For example, when the light intensity changes, the output voltage of the unit is dynamically adjusted so that the unit always operates at the maximum power point, avoiding efficiency loss caused by fixed parameters.

[0042] When collecting mechanical energy resources, the energy capture structure parameters (such as mechanical damping coefficient and spring stiffness) can be adjusted according to the movement intensity in the pet's movement data to ensure maximum mechanical energy conversion efficiency. For example, when the pet's movement intensity is high, the structural parameters can be adjusted to enhance vibration energy capture; when the movement intensity is low, the parameters can be adjusted to reduce structural losses.

[0043] When collecting thermal energy resources, the temperature difference structure (such as heat conduction path and generator arrangement) can be optimized based on the temperature difference in environmental data (pet body temperature and ambient temperature) to ensure maximum thermal energy conversion efficiency. For example, when the temperature difference increases, the heat conduction path is optimized to reduce heat loss, and when the temperature difference decreases, the operating mode of the generator is adjusted to maintain the basic conversion efficiency.

[0044] Finally, the real-time output power, energy conversion efficiency, and state of charge (SOC, reflecting the remaining energy storage) of each energy acquisition unit are collected in real time. The various performance parameters adjusted above are integrated to obtain the target performance parameters.

[0045] Step S204: Identify the pet's behavioral intention data, and predict the pet's behavioral intention using the behavioral intention data and a preset behavioral prediction model to obtain a personalized training strategy for the pet.

[0046] Relying solely on passive logic that triggers punishment for boundary crossings, such as immediately administering an electric shock after a pet crosses a boundary, fails to consider individual differences, behavioral intentions, and psychological states among different pets. This leads to low training efficiency and frequent stress reactions in pets. However, behavioral intention data reflects the pet's true state (e.g., anxious barking + high heart rate indicates stress). By using a pre-set behavioral prediction model, this data can be transformed into a quantifiable behavioral objective. Combined with personalized strategies generated based on individual differences, this improves training efficiency and reduces stress.

[0047] It should be noted that the behavior prediction model refers to an algorithm model that has been pre-trained using pet behavior sample data. After the pet behavior feature vector is input into the behavior prediction model, the behavior prediction model outputs the pet's behavioral intention and state (such as going out of bounds, stress, or normal). The model type can include lightweight deep learning models (such as LSTM networks), traditional machine learning models (such as support vector machines and decision trees), and other learning models.

[0048] In this embodiment, behavioral intent data can be collected through a behavior perception module on the trainer. This module may include a sound sensor to acquire sound signals such as barking and whimpering from the pet. It may also include physiological sensors to acquire physiological signals such as the pet's heart rate and skin impedance. Furthermore, it may include visual sensors to acquire information such as the pet's body posture. Data reflecting the pet's behavioral intent (such as anxiety or boundary-crossing tendencies) is captured from multiple dimensions of sound, physiology, and posture.

[0049] In this embodiment, the behavioral intention data collected in step S201 is preprocessed and input into a preset behavior prediction model. The model outputs the pet's current behavioral intention and state (model prediction result). For example, if the model predicts that the pet is barking anxiously at high frequency and its heart rate is significantly increased, it is determined that the pet is in a state of stress; if the model predicts that the pet is active normally and its heart rate is stable, it is determined that the pet has no abnormal intention.

[0050] By combining the prediction results of the behavior prediction model with the pet's basic information (historical training data, which can be pre-stored in the trainer or obtained through the communication module), targeted training strategies are generated, including training methods (such as sound warnings, vibration interventions, and electric shock interventions), training intensity (such as vibration intensity and electric shock intensity), and training intervals (such as the time interval between two interventions). For example, if a boundary-crossing intention is predicted and the pet is a puppy, a strategy of sound warning (priority), low-intensity vibration (backup), and extended training intervals is generated to avoid electric shocks. If a stress state is predicted and the pet is any type, a strategy of suspending punishment and enabling soothing cues (such as low-frequency sounds) is generated to avoid exacerbating stress.

[0051] Step S205: Adjust the communication parameters and positioning parameters of the trainer according to the target performance parameters and the personalized training strategy.

[0052] It should be noted that, on the one hand, traditional GPS positioning is easily affected by obstructions, and 433MHz communication is unreliable in complex electromagnetic environments. On the other hand, the communication and positioning modules are high-power units of the trainer, and traditional technologies cannot adjust parameters according to the power supply status (such as maintaining high-frequency positioning when the battery is low), which leads to a further reduction in battery life, or the functions are not fully utilized when the battery is high (such as not enabling real-time communication).

[0053] In this embodiment, the acquired target performance parameters are analyzed to determine the current energy sufficiency level of the trainer, which can be categorized into three levels: high sufficiency, medium sufficiency, and low sufficiency. The generated personalized training strategy is analyzed to determine the current functional requirements for communication and positioning. For example, these can be categorized into three levels: high requirement, medium requirement, and low requirement. High-requirement training strategies require real-time tracking of pet location (e.g., outdoor boundary crossing training, multi-pet collaborative training) and real-time transmission of training data (e.g., posture images, physiological data), requiring high-frequency positioning, real-time communication, and high-precision positioning modules. Medium-requirement training strategies require periodic updates of pet location (e.g., indoor basic training, fixed-point potty training) and timed data transmission (e.g., daily training reports), requiring medium-frequency positioning, timed communication, and medium-precision positioning modules. Low-requirement training strategies only need to ensure core fence warnings (e.g., low battery scenarios, nighttime stationary scenarios), without requiring real-time data transmission, requiring low-frequency positioning and only retaining core communication and basic precision positioning modules.

[0054] Based on the energy availability level and functional requirements level, communication and positioning parameters are dynamically adjusted. Adjustments to communication parameters include communication frequency (e.g., number of communications per second), communication protocol (e.g., Bluetooth, 433MHz, BeiDou short message service), and data transmission rate (e.g., amount of data transmitted per transmission). Adjustments to positioning parameters include positioning frequency (e.g., how often to locate), positioning accuracy level (e.g., meter-level, ten-meter-level), and positioning module type (e.g., GPS, BeiDou, Bluetooth positioning).

[0055] For example, in a training scenario using a receiver RX and a transmitter TX as trainers, when the distance between the receiver RX (pet) and the transmitter TX (owner) is less than 50 meters (a customizable threshold), under both high demand and medium abundance levels, only 433MHz communication is enabled, and the distance is roughly determined by the 433MHz signal strength (RSSI value). UWB is temporarily disabled to reduce power consumption. When the distance between the receiver RX (pet) and the transmitter TX (owner) is ≥50 meters or close to the threshold range (e.g., 48 meters), and both high demand and high abundance levels are applied, the UWB module will be automatically activated, enabling high-precision ranging to ensure accurate positioning.

[0056] This application effectively overcomes the limitations of existing pet training equipment that relies on a single energy source by collecting environmental data, pet movement status data, and behavioral intention data, combined with the determination of multi-source energy collection combinations and adjustment of energy conversion efficiency. This improves energy utilization efficiency and solves the problem of frequent charging in scenarios such as long-term outdoor training and indoor low-light conditions, extending the device's battery life. Simultaneously, it generates personalized training strategies based on behavioral intention data and preset behavior prediction models, breaking away from traditional rigid training logic. This adapts to the individual behavioral characteristics and real-time states of pets (such as boundary crossing intentions and stress states), helping to improve pet command obedience rates and reduce the incidence of stress reactions. Furthermore, by dynamically adjusting communication and positioning parameters according to target performance parameters and personalized training strategies, a precise balance between energy consumption and functional requirements can be achieved. This avoids functional waste during high battery periods and prevents core function failure during low battery periods, while also improving the reliability of communication and positioning in complex environments, comprehensively optimizing the intelligence level and actual usage effect of pet training equipment.

[0057] In some optional implementations of this embodiment, adjusting the target acquisition resources of the trainer based on the environmental data and the motion state data includes: When the light intensity of the environmental data is greater than or equal to a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a first acquisition ratio, wherein the acquisition ratio of the light energy resources is greater than the acquisition ratio of the mechanical energy resources, and the acquisition ratio of the light energy resources is greater than the acquisition ratio of the thermal energy resources. When the light intensity of the environmental data is less than a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a second acquisition ratio, wherein the sum of the acquisition ratio of the mechanical energy resources and the acquisition ratio of the thermal energy resources in the second acquisition ratio is greater than the acquisition ratio of the light energy resources.

[0058] In this embodiment, based on the efficiency characteristics of the solar energy resource unit (e.g., the solar unit efficiency is 20%, and the output power reaches 1.2W when the light intensity is ≥30000 lux), a threshold for light intensity (e.g., 30000 lux) is preset. This threshold is the critical point between efficient and inefficient output of solar energy resources. When the efficiency is above the threshold, the efficiency of solar energy resources is ≥18%, and when the efficiency is below the threshold, the efficiency is <10%.

[0059] When the light intensity exceeds a preset threshold (e.g., 50,000 lux on a sunny outdoor day), the output power of the solar energy resource unit is high (1.2W). The collection ratio of solar, thermal, and mechanical energy resources is adjusted to a first collection ratio, where the collection ratio of solar energy resources is greater than that of mechanical and thermal energy resources. For example, if the first collection ratio is 8:1:1, then the collection ratio of solar, thermal, and mechanical energy resources is adjusted to 8:1:1. That is, when collecting resources, solar energy resources are mainly collected, making them account for 80% of all target resources collected in the end, while the collection ratios of mechanical (dependent on pet movement) and thermal (dependent on the temperature difference between body temperature and ambient temperature) resources are reduced to 10% each. This can compensate for the power fluctuations of solar energy caused by instantaneous shading (such as tree shade) and reduce the use of devices for collecting mechanical and thermal energy resources, further reducing the overall power consumption of the trainer.

[0060] When the light intensity is less than a preset threshold (e.g., 500 lux indoors), the output power of the light energy resource unit is <0.2W, resulting in extremely low efficiency. The collection ratio of light, heat, and mechanical energy resources is adjusted to a second collection ratio, where the sum of the collection ratios of mechanical and heat resources is greater than the collection ratio of light resources. For example, if the second collection ratio is 1:5:4, then the collection ratio of light, heat, and mechanical energy resources is adjusted to 1:5:4. That is, when collecting resources, mechanical and heat resources are primarily collected, making their collection ratios in the final collected target resources 50% and 40%, respectively, while the collection ratio of light resources is reduced to 10%.

[0061] The environmental perception module of the trainer collects light data periodically (e.g., every hour). If the light intensity is detected to be close to the threshold (e.g., ±1000 lux) for three consecutive times, the threshold is automatically fine-tuned (e.g., the threshold can be adjusted to 35000 lux in areas with strong light at high altitudes) to avoid misjudgments caused by fixed thresholds.

[0062] For example, in an outdoor park, the environmental data is a light intensity of 50,000 lux (higher than the preset threshold of 30,000 lux), and the pet's activity is running (step frequency 2.5Hz). The energy harvesting system primarily uses a flexible solar cell (100mm² cadmium telluride thin-film battery) as the main power source, outputting 1.2W, which is stored in a 3.7V / 1000mAh lithium battery via an MPPT circuit. A mechanical energy resource unit (magnetic circuit switch structure) serves as the auxiliary power source, outputting 0.8W to supplement the solar power. A thermal energy resource unit (pet body temperature 38℃ - ambient temperature 25℃) outputs 0.1W as a backup. The final total energy supply reaches 2.1W, meeting the power consumption requirements for GPS positioning (1 time / 30s) and continuous Bluetooth 5.1 connection, while maintaining the lithium battery SOC above 90%.

[0063] For example, in an indoor playground, the environmental data is a light intensity of 800 lux (below the preset threshold of 30,000 lux), and the pet's activity is walking (step frequency 1Hz). The light energy collection ratio is adjusted to 5%, the mechanical energy collection ratio to 75% (output 0.3W), and the thermal energy collection ratio to 20% (output 0.1W), with both working together to power the system. The final total energy supply reaches 0.4W, meeting the power consumption requirements for Bluetooth positioning (accuracy 2 meters) and low-intensity vibration training, while the lithium battery SOC remains above 60%, and the static power consumption is reduced to 0.5uA.

[0064] This application addresses the issues of wasted solar energy in high-light scenarios and reliance on inefficient solar energy in low-light scenarios, improving energy utilization efficiency in solar-dominated scenarios and energy supply stability in low-light scenarios. It achieves automatic switching between high-light and low-light scenarios through illumination thresholds, eliminating the need for manual user adjustments, reducing operational complexity, and adapting to various scenarios such as homes, outdoors, and basements.

[0065] In some optional implementations of this embodiment, the performance parameters include the magnetic damping coefficient and the operating voltage. The above-mentioned adjustment of the trainer's performance parameters based on the target acquisition resources to obtain the target performance parameters includes: Based on the mechanical energy resources in the target acquisition resources, the magnetic damping coefficient of the trainer is adjusted to obtain a first performance parameter; based on the light energy resources in the target acquisition resources, the operating voltage of the trainer is adjusted to obtain a second performance parameter; based on the first performance parameter and the second performance parameter, the target performance parameter is obtained.

[0066] It should be noted that in traditional technologies, the magnetic damping coefficient of the mechanical energy resource unit is fixed, maintaining high damping even when the pet is stationary, leading to energy loss. The operating voltage of the solar energy resource unit is fixed, making it impossible to match the maximum power point with changes in illumination, resulting in low solar energy utilization. Motion-adaptive damping and a maximum power point tracking (MPPT) algorithm are used to address the problem of energy conversion efficiency being disconnected from motion and illumination.

[0067] Specifically, the harvesting of mechanical energy relies on the movement of a magnet relative to a coil, cutting magnetic field lines. The magnetic damping coefficient is a core physical parameter used to adjust the resistance of the magnet's movement relative to the coil, determining the resistance to the magnet's motion. If the damping coefficient is too high, the magnet cannot move effectively when the pet moves slightly, resulting in low energy conversion efficiency. If the damping coefficient is too low, the magnet moves excessively when the pet moves vigorously, and energy cannot be fully captured.

[0068] The system can collect pet movement data (acceleration, cadence) using a three-axis accelerometer to identify three states: stationary, walking, and running. Stationary (acceleration < 0.1g, cadence = 0), walking (acceleration 0.1-0.5g, cadence 1-2Hz), and running (acceleration > 0.5g, cadence > 2Hz).

[0069] The magnetic damping coefficient is adjusted according to the motion state data. For example, it is set to 0.1 N•s / m for running (low damping, to ensure the magnet moves fully and captures a large amount of mechanical energy), 0.05 N•s / m for walking (medium damping, to balance the motion amplitude and energy capture), and the mechanical energy harvesting is turned off for stationary state (infinite damping coefficient to avoid static energy consumption). The adjusted magnetic damping coefficient and the output power obtained after adjusting the magnetic damping coefficient are used as the first performance parameters.

[0070] Specifically, when collecting solar energy resources, the output power of a solar cell exhibits a single-peak curve as the operating voltage changes, indicating a unique maximum power point. This point can be tracked using the MPPT algorithm to ensure the solar cell always operates at its maximum power, avoiding power loss caused by a fixed voltage. For example, the initial operating voltage range is set to 1.8-5V (covering the voltage output range of a cadmium telluride thin-film battery). Output voltage (U) and current (I) are collected every 100ms, and the output power P = U × I is calculated. The voltage ΔU is slightly increased (e.g., by 0.1V). If the power P increases, it indicates the current voltage is below the maximum power point, and the voltage is increased further. If the power P decreases, it indicates the current voltage is above the MPP, and the voltage ΔU is decreased. When the power change is <5% after three consecutive adjustments, the current voltage is determined to be the MPP, and this voltage is maintained until changes in sunlight cause power fluctuations >10%, at which point perturbation tracking is restarted. The adjusted operating voltage and the power fluctuation parameters obtained after adjusting the operating voltage are used as the second performance parameter.

[0071] Finally, the first and second performance parameters of the two types of units are integrated to form the target performance parameters, such as "solar output 1.2W, vibration output 0.8W, current power = 92%, conversion efficiency 85%".

[0072] This application improves the conversion efficiency of mechanical energy resources by dynamically adjusting the magnetic damping coefficient, resulting in no additional energy consumption in static scenarios and reducing energy waste in low-activity scenarios. By adjusting the performance of light energy resources, the utilization rate of light energy is improved, ensuring stable output power in high-light scenarios.

[0073] In some optional implementations of this embodiment, the behavioral intent data includes pet posture images, pet sound data, and pet physiological data; identifying the pet's behavioral intent data, and predicting the pet's behavioral intent using the behavioral intent data and a preset behavioral prediction model to obtain a personalized training strategy for the pet, including: Visual features of the pet's posture image, acoustic features of the pet's sound data, and physiological features of the pet's physiological data are extracted; the visual features, acoustic features, and physiological features are fused into a behavioral feature vector; the behavioral feature vector is input into the preset behavior prediction model to predict the pet's behavioral intention; and a personalized training strategy for the pet is generated based on the behavioral intention.

[0074] It should be noted that relying solely on single posture data (such as crossing boundaries) to determine behavior, without integrating sound and physiological data, will lead to low accuracy in intent prediction. Personalized training strategies are generated by combining behavioral intent data with a pre-set behavior prediction model. This behavioral intent data includes pet posture images, sound data, and physiological data. Through feature extraction, fusion, and model input, the system predicts the pet's behavioral intent, providing a basis for personalized strategies.

[0075] In this embodiment, pet posture image data can be captured by the image acquisition unit (such as a miniature camera) built into the trainer to capture the pet's limb postures (such as facing the fence, arching its back, and tucking its tail). Feature extraction can be performed using a 3D CNN (three-dimensional convolutional neural network) to extract visual features. Taking three consecutive frames of images as input, the convolutional layer extracts the posture change trend (such as gait frequency and movement direction), and outputs a 256-dimensional visual feature vector (reflecting the dynamic features of the posture).

[0076] Pet sound data can be collected via microphone to capture pet sounds (barking, whimpering, quiet). Feature extraction can be performed using MFCC (Mel-frequency cepstral coefficients) to extract acoustic features, converting the sound signal into a Mel-frequency spectrum, calculating 13-dimensional MFCC coefficients (reflecting changes in sound frequency and amplitude), and outputting a 13-dimensional acoustic feature vector (distinguishing between anxious barking and normal vocalizations).

[0077] Pet physiological data can be collected using skin impedance sensors (reflecting heart rate variability) to calculate heart rate values ​​(beats / min). Feature extraction can be performed by extracting the mean heart rate and standard deviation of heart rate (reflecting heart rate stability) to output a 2D physiological feature vector (distinguishing between high heart rates during stress and normal heart rates).

[0078] Visual feature vectors (256-dimensional), acoustic feature vectors (13-dimensional), and physiological feature vectors (2-dimensional) are combined and normalized into a 271-dimensional behavioral feature vector, which is then normalized to the [0,1] interval to avoid the influence of differences in the amount of data in different dimensions (such as large visual feature values ​​and small physiological feature values) on the model's judgment.

[0079] The preset behavior prediction model takes a 271-dimensional behavior feature vector as input and outputs the pet's behavioral intent. During the training phase, it can be trained using 1000+ sets of pet behavior samples (out of bounds, stressed, normal). During the inference phase, the fused feature vector is input into the model in real time to obtain the probability distribution of the pet's current behavioral state (e.g., probability of normal posture, probability of vocal anxiety, probability of physiological stress).

[0080] This application avoids the limitations of single-data sets by fusing multimodal data, thus improving prediction accuracy. Based on accurate predictions, it triggers early warnings of boundary violations, improving command obedience. Timely identification of stress states also reduces the incidence of stress reactions. Simultaneously, it avoids ineffective punishments caused by misjudgments (such as triggering electric shocks due to misjudgment of boundary violations), reducing pet fear and increasing training acceptance.

[0081] In some optional implementations of this embodiment, the behavioral intent includes boundary-crossing intent and stress response intent. The step of generating a personalized training strategy for the pet based on the behavioral intent includes: If the pet is predicted to have the intention to cross the boundary, a training strategy containing warning and punishment instructions is generated; if the pet is predicted to have the intention to be stressed, a soothing strategy to suspend punishment or reduce training intensity is generated.

[0082] Specifically, combining the output behavioral feature vector with the model prediction results, if the pet's posture is facing the fence, gait frequency is >2Hz, and there are no stress physiological signals, it is determined that the pet has an intention to cross the boundary. Based on the probability of this intention and the distance between the pet and the fence (the set maximum activity distance for the pet), the level of this intention is divided into three levels: The probability of a minor boundary crossing intention is 50%-70%, and the distance from the fence is >5 meters; the probability of a moderate boundary crossing intention is 70%-90%, and the distance from the fence is 2-5 meters; the probability of a severe boundary crossing intention is >90%, and the distance from the fence is <2 meters.

[0083] In the tiered training strategy, the strategy is divided according to the boundary violation intention levels obtained above. Mild boundary violation intentions trigger only a warning command, moderate boundary violation intentions trigger a warning command and a mild penalty, and severe boundary violation intentions trigger a warning command and a moderate penalty.

[0084] Specifically, if the output behavioral feature vector and model prediction results meet the criteria of "high-frequency anxious barking (frequency 4000-6000Hz) and heart rate > 120 beats / min" or "abnormal posture (tail tucked, back arched) and heart rate > 110 beats / min", the pet is determined to be in a stress-related state. In this case, punishment is immediately suspended, and environmental soothing and training adjustments are carried out.

[0085] For example, in an indoor scenario, the enclosure is 10m x 8m. A 4-month-old puppy enters the sofa restricted area. After triggering a 30% intensity vibration, the model predicts an 88% probability of stress, characterized by high-frequency barking (5500Hz), a heart rate of 135 bpm, and a pacing posture with its tail tucked between its legs. The vibration module is immediately turned off, and 20dB white noise is played. The controller sends a stress alert, playing a voice message such as "Baby, don't be afraid, come down" through the collar. The restricted area intervention is paused, and the training switches to the "sit" command. Upon completion, a warm LED light is triggered. After 3 minutes, the puppy's heart rate drops to 95 bpm, there is no more anxious barking, and it returns to a normal state.

[0086] This application avoids harsh punishment upon triggering a single instance of discipline by employing tiered penalties, and alleviates stress responses through soothing strategies, thus reducing the incidence of stress reactions in pets. The training process is gentler, minimizing fear and resistance in pets. Simultaneously, early warnings and tiered intervention improve pets' obedience to commands and increase their acceptance of training.

[0087] In some optional implementations of this embodiment, adjusting the communication parameters and positioning parameters of the trainer according to the target performance parameters and the personalized training strategy includes: Based on the target performance parameters, determine the current available energy of the trainer; based on the personalized training strategy, determine the positioning update frequency and communication response requirements; based on the current available energy, the positioning update frequency, and the communication response requirements, adjust the communication parameters and positioning parameters of the trainer.

[0088] Traditional technologies use fixed communication and positioning parameters, which leads to energy waste when the battery is high (e.g., operating at high frequencies even when no high-frequency positioning is needed) and core functionality failure when the battery is low (e.g., the camera remains on even when the battery is low). Furthermore, they are not adjusted to meet training strategy requirements, resulting in a mismatch between functionality and needs (e.g., using GPS positioning for indoor training). Therefore, the communication and positioning parameters of the trainer need to be adjusted based on the required positioning update frequency and communication response requirements based on the currently available energy.

[0089] In this embodiment, the currently available energy can be represented by SOC, which refers to the state of charge of the lithium battery, reflecting the percentage (0%-100%) of the remaining power of the trainer relative to its rated capacity. It can be calculated by collecting voltage and current data through the battery management system (BMS) (e.g., 3.7V corresponds to SOC=100%, and 3.2V corresponds to SOC=30%).

[0090] Based on energy supply capacity, the State of Charge (SOC) is divided into three levels, clearly defining the upper limit of energy consumption for each level. High-power level: SOC ≥ 80%, indicating sufficient remaining power for the trainer to support high-power functions such as high-frequency GPS positioning and continuous Bluetooth connection; Medium-power level: 30% ≤ SOC < 80%, indicating moderate remaining power for the trainer, requiring a balance between power consumption and functionality, and disabling unnecessary modules; Low-power level: SOC < 30%, indicating insufficient remaining power for the trainer, retaining only low-power functions such as fence alerts and core communication.

[0091] Based on the personalized training strategy, determine the functional requirements for communication and positioning, and clarify the parameter configuration direction for different requirements. High-requirement training strategies require real-time tracking of pet location (e.g., outdoor boundary crossing training, multi-pet collaborative training) and real-time transmission of training data (e.g., posture images, physiological data). The trainer must have at least high-frequency positioning, real-time communication, and high-precision positioning modules. Medium-requirement training strategies require regular updates to pet location (e.g., indoor basic training, potty training) and timed data transmission (e.g., daily training reports). The trainer must have at least medium-frequency positioning, timed communication, and medium-precision positioning modules. Low-requirement training strategies only need to ensure core fence warnings (e.g., low battery scenarios, nighttime stationary scenarios), without requiring real-time data transmission; low-frequency positioning is sufficient. Therefore, only core communication and basic precision positioning modules can be retained in the trainer.

[0092] refer to Figure 3 , Figure 3 This application provides a diagram illustrating the positioning and communication parameters of a control method for an intelligent pet trainer. The communication and positioning parameters are adjusted according to different SOC levels and training strategies. For example, if the SOC level is high power consumption and the functional requirement level is medium demand, the communication parameters are adjusted to Bluetooth 5.1 mode with a timed connection to the receiver on the pet (once every 10 seconds). The positioning parameters are adjusted to GPS positioning mode with a positioning frequency of once per minute and an accuracy of 3 meters.

[0093] This application allows for full utilization of functions when the power level of the trainer is high and retention of basic functions when the power level is low, based on the trainer's battery status and different training strategies, thus reducing energy waste. It also improves communication success rate, and the switching between multiple positioning modules enables positioning accuracy of 1-5 meters in complex environments (indoors, high-rise buildings, areas with dense electromagnetic fields).

[0094] In some optional implementations of this embodiment, there are multiple pets, and the step of obtaining a personalized training strategy for the pets through the behavioral intent data and a preset behavioral prediction model includes: Collect interactive behavior data among multiple pets; based on the interactive behavior data, the behavioral intention data, and a preset behavior prediction model, generate personalized training strategies for each of the multiple pets.

[0095] When training multiple pets, by collecting interactive behavior data among the pets (such as chasing, avoiding, and approaching the fence together), and combining it with the individual behavioral characteristics of each pet (such as historical training data), an independent personalized training strategy can be generated for each pet. Priorities can also be set to achieve collaborative training of multiple pets.

[0096] In this embodiment, each pet is equipped with a unique RFID tag. The trainer has a built-in RFID reader (reading distance ≤ 1 meter) to distinguish different pets by reading the tag ID. For example, ID1 corresponds to a Golden Retriever, ID2 corresponds to a Persian cat, and ID3 corresponds to a Teddy dog, thus avoiding identity confusion.

[0097] Collecting environmental data (light, temperature), movement data (step frequency, acceleration), and behavioral intention data (posture, sound, physiology) for each pet separately ensures the independence of the data for each individual. The trainer's built-in wide-angle camera (120° field of view, covering the activity range of multiple pets) captures the interaction behaviors between pets, and target tracking algorithms (such as Deep SORT (Deep Simple Online and Realtime Tracking), a deep learning-based multi-target tracking (MOT) algorithm) are used to label the interaction type and frequency of each pet.

[0098] Specific interaction types can include chasing (pet A follows and approaches pet B), avoiding (pet A moves away from pet B), and joint behavior (pet A and pet B approach the fence together, or remain still together). The number of interactions per unit time (1 minute) is used to obtain the interaction frequency, and an interaction behavior matrix is ​​generated based on the interaction frequency (e.g., "ID1 chases ID2: 5 times / minute, ID2 avoids ID1: 3 times / minute").

[0099] Based on the interaction behavior matrix, the impact of interactions on the training objective is determined, including dominant and consequential behaviors. For example, if ID1 intends to cross the boundary and ID2 follows, then ID1 is the "dominant crosser" and ID2 is the "follower," with ID1's behavior having a greater impact on training. For instance, if ID1 is stressed (anxious barking and ID2 approaches), then ID1's stress may lead to ID2's stress, and calming ID1 should be prioritized.

[0100] Prioritize strategy execution based on pet type and training goals to ensure core needs are met first. For example, the first priority is functional pets (such as guide dogs and search and rescue dogs), where training accuracy must be prioritized to avoid interference from interactions that could affect their work performance. The second priority is active pets (such as Golden Retrievers and Labradors), which are prone to dominant behavior during interactions and require focused intervention to prevent them from inciting other pets to violate rules. The third priority is quiet pets (such as Persian cats and Garfield cats), which are mostly followers during interactions, allowing for gentler strategies to reduce unintended punishments.

[0101] Basic strategies are generated based on each pet's historical training data. For example, the frequency of punishment is reduced for pets with an obedience rate >80%, while the number of warnings is increased for pets with an obedience rate <50%.

[0102] Based on the interaction's impact and priority, adjust the basic strategy to ensure that the strategy is independent and adaptable to the interaction scenario. For the dominant transgressor (such as the ID1 Golden Retriever), use warnings and moderate vibrations to prevent it from leading other pets to cross the boundary; for followers (such as the ID2 Persian Cat), only use warnings without punishment to avoid being mistakenly punished for following. For stressed pets, prioritize soothing strategies to prevent stress transmission.

[0103] This application reduces costs by enabling multi-pet training with a single device (e.g., one controller and multiple pet collars). Independent strategies adapt to different pets, avoiding a one-size-fits-all approach and minimizing differences in training effectiveness. Interactive data identifies the dominant and follower pets, providing only warnings and no punishments to the follower, reducing the likelihood of incorrect punishments and decreasing pet resistance to training.

[0104] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0105] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0106] Further reference Figure 4 As a response to the above Figure 2 The implementation of the method shown in this application provides an embodiment of a smart pet trainer control device, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0107] like Figure 4 As shown, the intelligent pet trainer control device 400 described in this embodiment includes: a data acquisition module 401, an energy acquisition and determination module 402, an energy supply determination module 403, a strategy module 404, and a communication and positioning module 405. Wherein: The data acquisition module 401 is used to collect environmental data and pet movement status data; The energy acquisition determination module 402 is used to adjust the target acquisition resources of the trainer according to the environmental data and the motion state data, wherein the target acquisition resources include light energy resources, mechanical energy resources and thermal energy resources; The energy supply determination module 403 is used to adjust the performance parameters of the trainer based on the target acquisition resources to obtain the target performance parameters; The strategy module 404 is used to identify the pet's behavioral intention data, predict the pet's behavioral intention through the behavioral intention data and a preset behavior prediction model, and obtain a personalized training strategy for the pet. The communication and positioning module 405 is used to adjust the communication parameters and positioning parameters of the trainer according to the target performance parameters and the personalized training strategy.

[0108] In some embodiments of this application, the energy harvesting and determination module 402 is further configured to: When the light intensity of the environmental data is greater than or equal to a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a first acquisition ratio, wherein the acquisition ratio of the light energy resources is greater than the acquisition ratio of the mechanical energy resources, and the acquisition ratio of the light energy resources is greater than the acquisition ratio of the thermal energy resources. When the light intensity of the environmental data is less than a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a second acquisition ratio, wherein the sum of the acquisition ratio of the mechanical energy resources and the acquisition ratio of the thermal energy resources in the second acquisition ratio is greater than the acquisition ratio of the light energy resources.

[0109] In some embodiments of this application, the energy supply determination module 403 is further configured to: Based on the mechanical energy resources in the target acquisition resources, the magnetic damping coefficient of the trainer is adjusted to obtain the first performance parameter; Based on the light energy resources in the target acquisition resources, the operating voltage of the trainer is adjusted to obtain the second performance parameter; The target performance parameters are obtained based on the first performance parameter and the second performance parameter.

[0110] In some embodiments of this application, the strategy module 404 is further configured to: Extract the visual features of the pet posture image, the acoustic features of the pet sound data, and the physiological features of the pet physiological data; The visual features, acoustic features, and physiological features are fused into a behavioral feature vector; The behavioral feature vector is input into the preset behavior prediction model to predict the pet's behavioral intentions; Based on the stated behavioral intent, a personalized training strategy is generated for the pet.

[0111] In some embodiments of this application, the communication and positioning module 405 is further configured to: Based on the target performance parameters, determine the current available energy of the trainer; Based on the personalized training strategy, the positioning update frequency and communication response requirements are determined; The communication and positioning parameters of the trainer are adjusted based on the current available energy, the positioning update frequency, and the communication response requirements.

[0112] The intelligent pet trainer control device provided in this application extends the device's battery life and solves the problem of frequent charging by collecting and adjusting the efficiency of multiple energy sources. It generates personalized strategies based on the pet's individual behavioral characteristics and real-time status to reduce stress responses. Simultaneously, it dynamically adjusts communication and positioning parameters to balance energy consumption and functional requirements, improves communication and positioning reliability in complex environments, and optimizes the device's intelligence and usage effectiveness.

[0113] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 5 , Figure 5 This is a basic structural block diagram of the computer device in this embodiment.

[0114] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0115] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0116] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for the control method of a smart pet trainer. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0117] In some embodiments, the processor 42 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions of the intelligent pet trainer control method.

[0118] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0119] The computer device provided in this application extends battery life and solves the problem of frequent charging by collecting and adjusting the efficiency of multiple energy sources. It generates personalized strategies based on individual pet behavior characteristics and real-time status to reduce stress responses. Simultaneously, it dynamically adjusts communication and positioning parameters to balance energy consumption and functional requirements, improves communication and positioning reliability in complex environments, and optimizes device intelligence and usage effectiveness.

[0120] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the intelligent pet trainer control method described above.

[0121] The computer-readable storage medium provided in this application has (unique beneficial effects).

[0122] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0123] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A control method for an intelligent pet trainer, characterized in that, Includes the following steps: Collect environmental data and pet movement data; Based on the environmental data and the motion state data, the target acquisition resources of the trainer are adjusted, wherein the target acquisition resources include light energy resources, mechanical energy resources and thermal energy resources; Based on the target acquired resources, the performance parameters of the trainer are adjusted to obtain the target performance parameters; Identify the pet's behavioral intention data, and predict the pet's behavioral intention using the behavioral intention data and a preset behavior prediction model to obtain a personalized training strategy for the pet. The communication and positioning parameters of the trainer are adjusted according to the target performance parameters and the personalized training strategy.

2. The intelligent pet trainer control method according to claim 1, characterized in that, The step of adjusting the target acquisition resources of the trainer based on the environmental data and the motion state data includes: When the light intensity of the environmental data is greater than or equal to a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a first acquisition ratio, wherein the acquisition ratio of the light energy resources is greater than the acquisition ratio of the mechanical energy resources, and the acquisition ratio of the light energy resources is greater than the acquisition ratio of the thermal energy resources. When the light intensity of the environmental data is less than a preset threshold, the acquisition ratio of the light energy resources, the thermal energy resources and the mechanical energy resources is adjusted to a second acquisition ratio, wherein the sum of the acquisition ratio of the mechanical energy resources and the acquisition ratio of the thermal energy resources in the second acquisition ratio is greater than the acquisition ratio of the light energy resources.

3. The intelligent pet trainer control method according to claim 1, characterized in that, The performance parameters include the magnetic damping coefficient and the operating voltage. The step of adjusting the performance parameters of the trainer based on the target acquisition resources to obtain the target performance parameters includes: Based on the mechanical energy resources in the target acquisition resources, the magnetic damping coefficient of the trainer is adjusted to obtain the first performance parameter; Based on the light energy resources in the target acquisition resources, the operating voltage of the trainer is adjusted to obtain the second performance parameter; The target performance parameters are obtained based on the first performance parameter and the second performance parameter.

4. The intelligent pet trainer control method according to claim 1, characterized in that, The behavioral intent data includes pet posture images, pet sound data, and pet physiological data; The process of obtaining a personalized training strategy for the pet using the behavioral intent data and a preset behavioral prediction model includes: Extract the visual features of the pet posture image, the acoustic features of the pet sound data, and the physiological features of the pet physiological data; The visual features, acoustic features, and physiological features are fused into a behavioral feature vector; The behavioral feature vector is input into the preset behavior prediction model to predict the pet's behavioral intentions; Based on the stated behavioral intent, a personalized training strategy is generated for the pet.

5. The intelligent pet trainer control method according to claim 4, characterized in that, The behavioral intentions include boundary-crossing intentions and stress response intentions. The process of generating a personalized training strategy for the pet based on these behavioral intentions includes: If the pet is predicted to have the intention to cross the boundary, a training strategy containing warning and punishment instructions is generated. If the pet is predicted to have the stress intention, a soothing strategy is generated to suspend punishment or reduce the intensity of training.

6. The intelligent pet trainer control method according to claim 1, characterized in that, The step of adjusting the communication and positioning parameters of the trainer according to the target performance parameters and the personalized training strategy includes: Based on the target performance parameters, determine the current available energy of the trainer; Based on the personalized training strategy, the positioning update frequency and communication response requirements are determined; The communication and positioning parameters of the trainer are adjusted based on the current available energy, the positioning update frequency, and the communication response requirements.

7. The intelligent pet trainer control method according to claim 1, characterized in that, There are multiple pets. The process of predicting the pets' behavioral intentions using the behavioral intention data and a preset behavioral prediction model to obtain personalized training strategies for the pets includes: Collect data on the interaction behavior among multiple pets. Based on the interactive behavior data, the behavioral intent data, and the preset behavior prediction model, personalized training strategies are generated for each of the pets.

8. A control device for an intelligent pet trainer, characterized in that, include: The data acquisition module is used to collect environmental data and pet movement data; An energy acquisition determination module is used to adjust the target acquisition resources of the trainer based on the environmental data and the motion state data, wherein the target acquisition resources include light energy resources, mechanical energy resources and thermal energy resources; An energy supply determination module is used to adjust the performance parameters of the trainer based on the target acquisition resources to obtain the target performance parameters; The strategy module is used to identify the pet's behavioral intention data, and predict the pet's behavioral intentions using the behavioral intention data and a preset behavior prediction model, thereby obtaining a personalized training strategy for the pet. The communication and positioning module is used to adjust the communication and positioning parameters of the trainer according to the target performance parameters and the personalized training strategy.

9. A computer device, characterized in that, The device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the intelligent pet trainer control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the intelligent pet trainer control method as described in any one of claims 1 to 7.

Images