Pet intelligent necklace and insect repellent method

Abstract

The application discloses a kind of pet intelligent necklace and pest control method, the pet intelligent necklace includes necklace band, shell cover, wear rope mouth, multi-sensor fusion module, intelligent self-adapting learning module, multi-frequency dynamic control module, wireless communication and cloud interaction module and energy management module;Shell cover both ends are provided with wear rope mouth, for fixing the end of two necklace bands, the other end of two necklace bands is connected by buckle;Multi-frequency dynamic control module, intelligent self-adapting learning module, multi-sensor fusion module, wireless communication and cloud interaction module and energy management module are located inside shell cover.The present application can be aimed at different parasitic species, individual physiological state and environmental change of pet, realize real-time adaptive pest control strategy, can continuously optimize pest control parameter, improve pest control efficiency and reduce energy consumption, significantly enhance the intelligent and personalized level.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to a smart pet collar and a deworming method. Background Technology

[0002] Many ultrasonic flea and tick collars on the market currently use fixed frequencies or are limited to switching between a few frequencies. Such solutions struggle to effectively address the frequency sensitivities of different parasites and varying environmental conditions, leading to inconsistent flea and tick control, or even failure under specific conditions. Furthermore, traditional collars lack comprehensive sensing of the pet's physiological state (such as heart rate and blood oxygen saturation) and environmental parameters (such as temperature, humidity, and light intensity), making personalized, dynamic strategy adjustments impossible. While some flea and tick collars offer Bluetooth or app control, these are often limited to manual adjustment and preset modes, lacking deep learning and adaptive optimization capabilities.

[0003] Chinese invention patent application CN113225718A, entitled "A Bluetooth Pet Mosquito Repellent, Insect Repellent, and Anti-Loss Wearing Device System," can only be controlled via Bluetooth, has limited frequency switching, lacks deep learning capabilities, and consumes a lot of power. Chinese utility model patent CN214853596U, entitled "A Pet Collar with Break Protection and Insect Repellent Effect," can only use a fixed frequency, cannot achieve dynamic modulation, lacks intelligent control, sensor fusion, and cloud analysis, and has limited functionality.

[0004] Therefore, there is an urgent need for an ultrasonic pet deworming collar that can achieve wideband dynamic control, intelligent adaptive learning, and multi-dimensional perception to cope with various parasites, complex environments, and individual pet differences. Summary of the Invention

[0005] The purpose of this invention is to provide a smart pet collar and deworming method. By using broad-spectrum ultrasonic modulation technology, deep learning algorithms, cloud big data, and multi-sensor fusion, the ultrasonic frequency, waveform, and output intensity can be adaptively adjusted according to different parasite types, the pet's own physiological state, and environmental conditions to achieve a highly personalized and efficient deworming solution.

[0006] This invention adopts the following technical solution: a smart pet collar, comprising:

[0007] The collar strap, outer shell, cord threading port, multi-sensor fusion module, intelligent adaptive learning module, multi-frequency dynamic control module, wireless communication and cloud interaction module, and energy management module.

[0008] The outer casing has loops at both ends for securing one end of the two collar straps, and the other ends of the two collar straps are connected by buckles.

[0009] The multi-frequency dynamic control module, intelligent adaptive learning module, multi-sensor fusion module, wireless communication and cloud interaction module, and energy management module are located inside the outer casing.

[0010] The multi-frequency dynamic control module and the intelligent adaptive learning module transmit data wirelessly. The intelligent adaptive learning module and the multi-sensor fusion module transmit data wirelessly. The wireless communication and cloud interaction module transmits data wirelessly to the multi-frequency dynamic control module, the intelligent adaptive learning module, and the multi-sensor fusion module, respectively.

[0011] The energy management module provides power and monitors the battery status in real time.

[0012] Furthermore, the multi-sensor fusion module includes physiological sensors, environmental sensors, motion and position sensors, and an edge computing unit, integrating the sensors; the physiological sensors are used to acquire the pet's physiological state parameters in real time, the environmental sensors are used to collect the environmental parameters of the pet, the motion and position sensors are used to monitor the pet's activity range, movement intensity, and location, and the edge computing unit is used to preprocess the data.

[0013] Furthermore, physiological sensors include heart rate sensors, blood oxygen sensors, and body temperature sensors; environmental sensors include temperature sensors, humidity sensors, light sensors, and barometric pressure sensors; and motion and position sensors include GPS sensors and accelerometers.

[0014] Furthermore, the collar strap is made of flexible, waterproof, and tear-resistant material; the outer shell is made of impact-resistant material.

[0015] Furthermore, the wireless methods include Bluetooth, Wi-Fi, or NB-IoT.

[0016] Furthermore, this invention also proposes a method for deworming a pet smart collar, comprising:

[0017] S1. Use a multi-sensor fusion module to monitor the pet's physiological and behavioral parameters and the parameters of the pet's environment, and preprocess these parameters to obtain preprocessed parameters.

[0018] S2. The preprocessed parameters are processed using the intelligent adaptive learning module to obtain the insect repelling strategy, which includes the insect repelling demand level and the optimal insect repelling parameters, and the insect repelling parameter adjustment command is sent to the multi-frequency dynamic control module.

[0019] S3. Based on the deworming parameter adjustment command, the multi-frequency dynamic control module adjusts the deworming parameters in the frequency range of 10kHz to 120kHz according to the type of parasite and environmental conditions to achieve deworming of pets.

[0020] Information obtained from the S4 multi-sensor fusion module, intelligent adaptive learning module, and multi-frequency dynamic control module is transmitted to the wireless communication and cloud interaction module. It interacts with the cloud platform and mobile application in real time via wireless means to realize data uploading, analysis, and remote control, update the pest control strategy, and transmit it to the intelligent adaptive learning module and multi-frequency dynamic control module.

[0021] Furthermore, in step S1, preprocessing includes signal filtering, data cleaning, normalization, and feature extraction.

[0022] Furthermore, in step S2, within a set time period, the parameters of the deep learning model are trained using big data from the cloud. These parameters include weights and hyperparameters. The obtained pest control strategy is compared with the actual effect, the model parameters are optimized, and the trained deep learning model is obtained.

[0023] The preprocessed parameters are input into the trained deep learning model. The parameters are then transmitted to the hidden layer through the input layer to learn the relationship between the parameters. The activation function is used to perform non-linear mapping, and the insect repelling strategy is obtained through the output layer.

[0024] The optimal insect repellent parameters include ultrasonic frequency, waveform, and output intensity; the waveform includes continuous wave, pulse wave, and mixed wave, and the output intensity includes levels 1 to 5.

[0025] Furthermore, in step S3, the multi-frequency dynamic control module employs frequency sweeping, frequency hopping, pulse modulation, and waveform superposition techniques to cover the parasite's sensitive spectrum.

[0026] When the sensitive frequency corresponding to the parasite is between 30kHz and 60kHz, frequency sweeping or frequency hopping is performed within this range; when the pet is in grass, pulse modulation and waveform superposition techniques are used.

[0027] Furthermore, in step S4, the adjusted deworming parameters in the multi-frequency dynamic control module and the pet's physiological parameters and environmental parameters obtained by the multi-sensor fusion module are compared and evaluated using cloud-based functions. When the pet's heart rate is >160 beats / minute or <50 beats / minute, it indicates an abnormal heart rate, and the output intensity is adjusted. When the ambient temperature of the pet's environment is >30℃ or <10℃ and the ambient humidity is >60% or <30%, it indicates a significant change in ambient temperature and humidity, and the ultrasonic frequency is adjusted. Thus, the deworming strategy is updated.

[0028] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0029] 1. This invention improves the deworming efficiency against specific parasites by approximately 20%-30%.

[0030] 2. This invention utilizes intelligent adaptive learning and low-power strategies to reduce energy consumption by approximately 10%-15% within the same working time, thereby extending device battery life.

[0031] 3. The multi-sensor fusion proposed in this invention can maintain stable pest control effects under varying climate and environmental conditions without frequent user intervention. Furthermore, this invention is the first to combine the pet's physiological condition, environmental parameters, and pest control, ensuring both personalized pest control for optimal results and maintaining the pet's best health. Attached Figure Description

[0032] Figure 1 This is an overall structural diagram of the smart pet collar of the present invention.

[0033] Figure 2 This is a diagram of the internal modules of the smart pet collar of the present invention.

[0034] Figure 3 This is a flowchart illustrating the overall implementation of the present invention. Detailed Implementation

[0035] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0036] To achieve the above objectives, this invention proposes a smart pet collar, such as... Figure 1 As shown, it includes:

[0037] The collar strap, outer shell, cord threading port, multi-sensor fusion module, intelligent adaptive learning module, multi-frequency dynamic control module, wireless communication and cloud interaction module, and energy management module.

[0038] The outer casing has loops at both ends for securing one end of the two collar straps, and the other ends of the two collar straps are connected by buckles.

[0039] The multi-frequency dynamic control module, intelligent adaptive learning module, multi-sensor fusion module, wireless communication and cloud interaction module, and energy management module are located inside the outer casing.

[0040] like Figure 2 As shown, the multi-frequency dynamic control module and the intelligent adaptive learning module transmit data wirelessly, the intelligent adaptive learning module and the multi-sensor fusion module transmit data wirelessly, and the wireless communication and cloud interaction module transmit data wirelessly to the multi-frequency dynamic control module, the intelligent adaptive learning module, and the multi-sensor fusion module, respectively.

[0041] The energy management module provides power and monitors the battery status in real time.

[0042] The multi-sensor fusion module includes physiological sensors, environmental sensors, motion and position sensors, and an edge computing unit, integrating the sensors. The physiological sensors are used to acquire the pet's physiological state parameters in real time, the environmental sensors are used to collect the environmental parameters of the pet, the motion and position sensors are used to monitor the pet's activity range, movement intensity, and location, and the edge computing unit is used to preprocess the data.

[0043] Physiological sensors include heart rate sensors, blood oxygen sensors, and body temperature sensors; environmental sensors include temperature sensors, humidity sensors, light sensors, and barometric pressure sensors; motion and position sensors include GPS sensors and accelerometers.

[0044] The collar strap is made of flexible, waterproof, and tear-resistant material; the outer shell is made of impact-resistant material.

[0045] The wireless connection can be Bluetooth, Wi-Fi, or NB-IoT.

[0046] A method of using a smart pet collar to repel parasites, such as Figure 3 As shown, it includes:

[0047] S1. Utilize a multi-sensor fusion module to comprehensively monitor the pet's physiological and behavioral parameters and the parameters of the pet's environment, and preprocess these parameters to obtain preprocessed parameters.

[0048] Edge computing units are used to preprocess the parameters monitored by sensors to reduce data latency and improve decision response speed.

[0049] The preprocessing includes signal filtering, data cleaning, normalization, and simple feature extraction.

[0050] Signal filtering is used to eliminate electromagnetic interference or invalid noise; data cleaning is used to remove obviously unreasonable or missing data; normalization is used to represent data such as temperature and humidity in the same dimension or relative proportion, which facilitates subsequent model calculations; simple feature extraction is used to extract corresponding features to determine the current state, such as extracting acceleration peaks or frequency components from motion sensor data to determine activity level.

[0051] S2. The preprocessed parameters are processed using an intelligent adaptive learning module to obtain a pest control strategy. This strategy includes the pest control demand level and optimal pest control parameters, and an adjustment command for the pest control parameters is sent to the multi-frequency dynamic control module. The specific content is as follows:

[0052] Within a set time period (such as daily or weekly), the parameters of the deep learning model are trained using big data in the cloud to make the deworming strategy more suitable for the current pet's habits and environment. These parameters include weights and hyperparameters. The obtained deworming strategy is compared with the actual effect, the model parameters are optimized, and the trained deep learning model is obtained.

[0053] The preprocessed parameters are input into the trained deep learning model. The parameters are then transmitted to the hidden layer through the input layer to learn the relationship between the parameters. The activation function is used to perform non-linear mapping, and the insect repelling strategy is obtained through the output layer.

[0054] The optimal insect repellent parameters include ultrasonic frequency (such as 25kHz, 50kHz, 80kHz, etc.), waveform (continuous wave, pulse wave, mixed wave, etc.), and output intensity (such as adjustable from level 1 to 5).

[0055] As usage time accumulates, the system can continuously update its strategies, achieving progressive optimization from the initial plan to personalized customization, thereby improving the accuracy and efficiency of pest control.

[0056] S3. Based on the deworming parameter adjustment command, the multi-frequency dynamic control module adjusts the deworming parameters in the frequency range of 10kHz to 120kHz according to the type of parasite and environmental conditions to achieve deworming of pets.

[0057] Frequency sweeping, frequency hopping, pulse modulation, and waveform superposition techniques are used to cover the sensitive spectrum of parasites, so as to adapt to the sensitive frequency bands of various parasites and continuously optimize the parasite repellency efficiency.

[0058] When the sensitive frequency corresponding to the parasite is between 30kHz and 60kHz, frequency sweeping or frequency hopping is performed within this range. When the pet is in grassy or damp areas, the output intensity is appropriately increased, and pulse modulation and waveform superposition techniques are used. Combining different waveforms and frequency sweeping methods can achieve highly efficient interference and expulsion of the parasite's auditory or sensory organs.

[0059] When a pet experiences significant stress, the system can automatically reduce output intensity or shorten the treatment cycle to balance pet comfort and deworming effectiveness. Based on periodic monitoring data, the frequency, waveform, and intensity are continuously fine-tuned to ensure optimal deworming results at different times and in different environments.

[0060] Information obtained from the S4 multi-sensor fusion module, intelligent adaptive learning module, and multi-frequency dynamic control module is transmitted to the wireless communication and cloud interaction module. It interacts with the cloud platform and mobile application in real time via wireless means such as Bluetooth, Wi-Fi, or NB-IoT to achieve data upload. The cloud performs a comprehensive analysis of the overall pet health, environmental risks, and parasite outbreak trends. Combining the biological characteristics of parasites and individual differences of pets, it updates the deworming strategy and transmits it to the intelligent adaptive learning module and multi-frequency dynamic control module.

[0061] The system compares and evaluates the adjusted deworming parameters in the multi-frequency dynamic control module with the pet's physiological parameters and environmental parameters obtained by the multi-sensor fusion module in the cloud, thereby updating the deworming strategy. It can flexibly change the frequency and power output in different environments and pet states, effectively improving the deworming effect.

[0062] The comprehensive analysis utilizes the computing power of cloud platforms (such as big data analytics, model training and prediction), specifically including:

[0063] Behavioral pattern recognition: Through multi-dimensional data, identify the pet's activity patterns, such as whether it is at rest, moving, or resting.

[0064] Health status assessment: Based on physiological data (such as heart rate, blood oxygen level, etc.), determine whether the pet is in a state of health risk or whether there is stress caused by environmental changes.

[0065] Environmental adaptability analysis: By analyzing real-time environmental data such as temperature and humidity, we determine whether the pet is currently in a suitable environment for deworming. The process of updating the deworming strategy includes:

[0066] Changes in a pet's physiological state: If data such as heart rate or blood oxygenation become abnormal (e.g., heart rate >160 beats / minute or <50 beats / minute), the cloud will assess whether the pet has health problems (e.g., excessive stress, fever, etc.) and adjust the deworming strategy accordingly. For example, when a pet is anxious or under a lot of stress, the intensity of ultrasound can be reduced to avoid overstimulation.

[0067] Environmental changes: Significant changes in ambient temperature and humidity (e.g., ambient temperature >30℃ or <10℃, ambient humidity >60% or <30%) may affect the activity and sensitivity of parasites. The cloud system will determine the appropriate deworming frequency range based on environmental data and transmit the new frequency settings to the collar.

[0068] Insufficient parasite control effect: If the feedback data received by the cloud platform indicates that the current frequency range cannot effectively eliminate the target parasites (e.g., poor parasite detection or behavioral pattern feedback), the parasite control parameters (such as frequency, pulse pattern, etc.) will be optimized through a deep learning model to ensure the best parasite control effect.

[0069] External interference: Special circumstances affecting the environment or the pet (such as excessively high temperatures or sudden vigorous exercise by the pet) may affect the effectiveness of the deworming treatment. In such cases, the cloud platform will dynamically adjust the deworming strategy to maintain its effectiveness.

[0070] Frequency adjustment: Based on the analysis of the cloud platform, the frequency range of the ultrasonic insect repellent is automatically adjusted, for example, the optimal frequency is selected between 80kHz and 120kHz.

[0071] Pulse mode adjustment: If the deworming effect is found to be insufficient, the frequency of frequency switching or the intensity of pulse modulation may be increased to enhance the interference effect on parasites.

[0072] Deworming intensity adjustment: The deworming intensity is automatically adjusted according to the pet's health status and environmental changes to avoid causing discomfort to the pet.

[0073] The pet collar continuously uploads real-time data to the cloud, which then adjusts its strategies accordingly. For example, if the cloud detects changes in the pet's health, it may need to reassess the deworming frequency and parameters and make adjustments in real time.

[0074] Pet owners can manually control the collar via a mobile app. For example, they can adjust the frequency range, turn the flea and tick treatment mode on or off, or view statistics on the effectiveness of the treatment.

[0075] The S5 energy management module provides power and monitors battery status in real time. It automatically shortens its operating cycle when the battery level drops below 20% or the voltage drops below 3.6V.

[0076] Example:

[0077] When the temperature and humidity sensors detect a temperature between 25 and 30°C and a humidity between 60 and 80%, it is determined that the current environment is suitable for the breeding of fleas and ticks. In this case, the ultrasonic frequency is adjusted to the range of 30 kHz to 50 kHz, the output waveform is a pulse wave, and the output intensity is increased to level 3 to 4.

[0078] When the light sensor or GPS location determines that the pet is in an outdoor environment with direct sunlight and the temperature is >35℃, it may cause some parasites to become less active. In this case, the output intensity should be reduced to level 1-2 or the frequency bandwidth should be lowered to reduce energy consumption.

[0079] If deep learning analysis determines that the current multi-parameter comprehensive result poses a high risk to a specific mite, then the frequency sweep range is concentrated in the range of 80kHz to 100kHz, and the number of pulse modulation frequency switching times is increased to improve the interference effect on this type of parasite.

[0080] The specific content with higher risks includes:

[0081] If a pet's heart rate is consistently higher than normal (e.g., above 160 bpm) and accompanied by other health indicators (e.g., decreased blood oxygen saturation, <90%), it may indicate that the pet is under some kind of health stress, which increases the risk of mites.

[0082] When blood oxygen levels remain below the normal range (e.g., below 90%) and the pet is inactive, it may indicate the presence of parasites.

[0083] Based on their life habits, mites are generally more active in warm and humid environments. Therefore, the risk of mites is higher when the temperature is close to or above 25°C and the humidity is above 60%.

[0084] If a pet stays in grassy or damp areas for an extended period and does not show significant changes in activity (low activity level), the risk of mites is high.

[0085] If the risk is high, add two pulse switching cycles to ensure the frequency is effective against mites.

[0086] If the cloud analysis results indicate a high risk of mites, four pulse frequency switches can be added to ensure that the frequency covers a wider range of mite frequencies.

[0087] In situations with extremely high mite risk (such as when the pet's surrounding environment is suitable for mite survival and the pet is clearly unwell), the system can increase the frequency switching by 6 times to maximize the deworming effect.

[0088] The adjustment of the pulse frequency is automated by the cloud platform based on real-time data feedback, ensuring effective interference with mites and reducing excessive stimulation to pets.

[0089] The smart pet collar of this invention is fixed to the pet's neck and worn daily to prevent various parasites such as fleas, ticks, lice, and demodex mites.

[0090] When pets go out (such as in grasslands, woodlands, etc.) or come into contact with other animals, the collar uses a multi-frequency dynamic control module to emit ultrasonic waves of different frequencies and waveforms, effectively interfering with and repelling parasites.

[0091] Meanwhile, the multi-sensor fusion module inside the collar can monitor the pet's body temperature, heart rate, movement status, and environmental temperature and humidity in real time, and transmit the data to the intelligent adaptive learning module and the cloud for analysis, thereby dynamically and adaptively optimizing the deworming parameters.

[0092] The modular design of this invention reduces equipment maintenance costs and provides a convenient path for future continuous iteration of algorithms and hardware. This invention employs low-power chips, intelligent sleep mechanisms, and dynamic power allocation technology to maximize energy savings.

[0093] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A smart pet collar, characterized in that, include: Collar strap, outer shell, cord threading port, multi-sensor fusion module, intelligent adaptive learning module, multi-frequency dynamic control module, wireless communication and cloud interaction module, and energy management module; The outer casing has loops at both ends for securing one end of two collar straps, and the other ends of the two collar straps are connected by buckles. The multi-frequency dynamic control module, intelligent adaptive learning module, multi-sensor fusion module, wireless communication and cloud interaction module, and energy management module are located inside the outer casing; The multi-frequency dynamic control module and the intelligent adaptive learning module transmit data wirelessly, the intelligent adaptive learning module and the multi-sensor fusion module transmit data wirelessly, and the wireless communication and cloud interaction module transmit data wirelessly to the multi-frequency dynamic control module, the intelligent adaptive learning module, and the multi-sensor fusion module, respectively. The energy management module provides power and monitors the battery status in real time. Each module is configured to perform the following processing: S1. Use a multi-sensor fusion module to monitor the pet's physiological and behavioral parameters and the parameters of the pet's environment, and preprocess these parameters to obtain preprocessed parameters; S2. The preprocessed parameters are processed using the intelligent adaptive learning module to obtain a pest control strategy. This strategy includes the pest control demand level and optimal pest control parameters, and an adjustment command for the pest control parameters is sent to the multi-frequency dynamic control module. Specifically: Within a set time period, the parameters of the deep learning model are trained using big data in the cloud. These parameters include weights and hyperparameters. The obtained pest control strategy is compared with the actual effect, the model parameters are optimized, and the trained deep learning model is obtained. The preprocessed parameters are input into the trained deep learning model. The parameters are then transmitted to the hidden layer through the input layer to learn the relationship between the parameters. The activation function is used to perform non-linear mapping, and the insect repelling strategy is obtained through the output layer. The optimal insect repellent parameters include ultrasonic frequency, waveform, and output intensity; waveforms include continuous wave, pulse wave, and mixed wave, and output intensity includes levels 1 to 5. S3. Based on the deworming parameter adjustment command, the multi-frequency dynamic control module adjusts the deworming parameters in the frequency range of 10kHz to 120kHz according to the type of parasite and environmental conditions to achieve deworming of pets. Information obtained from the S4 multi-sensor fusion module, intelligent adaptive learning module, and multi-frequency dynamic control module is transmitted to the wireless communication and cloud interaction module. It interacts wirelessly with the cloud platform and mobile application in real time to achieve data uploading, analysis, and remote control, updating the pest control strategy and transmitting the updated information to the intelligent adaptive learning module and multi-frequency dynamic control module. Specifically: The adjusted deworming parameters in the multi-frequency dynamic control module and the pet's physiological parameters and environmental parameters obtained by the multi-sensor fusion module are compared and evaluated using cloud-based functions. When the pet's heart rate is >160 beats / minute or <50 beats / minute, it indicates an abnormal heart rate, and the output intensity is adjusted. When the ambient temperature of the pet's environment is >30°C or <10°C and the ambient humidity is >60% or <30%, it indicates a significant change in ambient temperature and humidity, and the ultrasonic frequency is adjusted. Thus, the deworming strategy is updated.

2. The smart pet collar according to claim 1, characterized in that, The multi-sensor fusion module includes physiological sensors, environmental sensors, motion and position sensors, and an edge computing unit, integrating the sensors. The physiological sensors are used to acquire the pet's physiological state parameters in real time, the environmental sensors are used to collect the environmental parameters of the pet, the motion and position sensors are used to monitor the pet's activity range, movement intensity, and location, and the edge computing unit is used to preprocess the data.

3. The smart pet collar according to claim 2, characterized in that, Physiological sensors include heart rate sensors, blood oxygen sensors, and body temperature sensors; environmental sensors include temperature sensors, humidity sensors, light sensors, and barometric pressure sensors; motion and position sensors include GPS sensors and accelerometers.

4. The smart pet collar according to claim 1, characterized in that, The collar strap is made of flexible, waterproof, and tear-resistant material; the outer shell is made of impact-resistant material.

5. The smart pet collar according to claim 1, characterized in that, The wireless connection can be Bluetooth, Wi-Fi, or NB-IoT.

6. A method for deworming pets using the smart collar of claim 1, characterized in that, include: S1. Use a multi-sensor fusion module to monitor the pet's physiological and behavioral parameters and the parameters of the pet's environment, and preprocess these parameters to obtain preprocessed parameters; S2. The preprocessed parameters are processed using the intelligent adaptive learning module to obtain a pest control strategy. This strategy includes the pest control demand level and optimal pest control parameters, and an adjustment command for the pest control parameters is sent to the multi-frequency dynamic control module. Specifically: Within a set time period, the parameters of the deep learning model are trained using big data in the cloud. These parameters include weights and hyperparameters. The obtained pest control strategy is compared with the actual effect, the model parameters are optimized, and the trained deep learning model is obtained. The preprocessed parameters are input into the trained deep learning model. The parameters are then transmitted to the hidden layer through the input layer to learn the relationship between the parameters. The activation function is used to perform non-linear mapping, and the insect repelling strategy is obtained through the output layer. The optimal insect repellent parameters include ultrasonic frequency, waveform, and output intensity; waveforms include continuous wave, pulse wave, and mixed wave, and output intensity includes levels 1 to 5. S3. Based on the deworming parameter adjustment command, the multi-frequency dynamic control module adjusts the deworming parameters in the frequency range of 10kHz to 120kHz according to the type of parasite and environmental conditions to achieve deworming of pets. Information obtained from the S4 multi-sensor fusion module, intelligent adaptive learning module, and multi-frequency dynamic control module is transmitted to the wireless communication and cloud interaction module. It interacts wirelessly with the cloud platform and mobile application in real time to achieve data uploading, analysis, and remote control, updating the pest control strategy and transmitting the updated information to the intelligent adaptive learning module and multi-frequency dynamic control module. Specifically: The adjusted deworming parameters in the multi-frequency dynamic control module and the pet's physiological parameters and environmental parameters obtained by the multi-sensor fusion module are compared and evaluated using cloud-based functions. When the pet's heart rate is >160 beats / minute or <50 beats / minute, it indicates an abnormal heart rate, and the output intensity is adjusted. When the ambient temperature of the pet's environment is >30°C or <10°C and the ambient humidity is >60% or <30%, it indicates a significant change in ambient temperature and humidity, and the ultrasonic frequency is adjusted. Thus, the deworming strategy is updated.

7. The method for deworming a pet smart collar according to claim 6, characterized in that, In step S1, preprocessing includes signal filtering, data cleaning, normalization, and feature extraction.

8. The method for deworming a pet smart collar according to claim 6, characterized in that, In step S3, the multi-frequency dynamic control module uses frequency sweeping, frequency hopping, pulse modulation, and waveform superposition techniques to cover the sensitive spectrum of the parasite. When the sensitive frequency corresponding to the parasite is between 30kHz and 60kHz, frequency sweeping or frequency hopping is performed within this range. When the pet is in the grass, pulse modulation and waveform superposition techniques are used.

Images