Pet GPS positioning tracking fence alarm method and system

Abstract

The invention discloses a pet GPS positioning tracking fence alarm method and system, and relates to the technical field of pet tracking, and the method comprises the steps: receiving first positioning data sent by a target pet tracking device; identifying target change information caused by environmental interference according to the first positioning data; identifying a signal interference area according to the frequency and the duration of the target change information; according to the electronic fence area and signal quality information sent by the multiple target pet tracking devices, a signal loss area is identified, and the signal quality information comprises the number of connectable satellites, signal intensity or positioning data disturbance values; if the pet is located in a signal poor area, the moving direction of the pet is obtained, and the signal poor area comprises a signal interference area or a signal loss area; and if the moving direction points to the boundary of the electronic fence or the poor signal area, generating a pet missing early warning. According to the invention, pet GPS positioning tracking and alarm triggering are realized, and the positioning tracking accuracy and the alarm timeliness are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pet tracking, and in particular to a pet GPS positioning tracking fence alarm method and system. BACKGROUND

[0002] In modern pet management, electronic fence systems are used to restrict the activities of pets within a certain area to prevent them from accidentally leaving. Such systems usually rely on tracking devices worn by pets that can periodically obtain global positioning system (GPS) positioning information. After receiving these positioning data, the electronic fence system compares it with the pre-set electronic fence boundary to determine whether the current location of the pet is still within the allowed activity area. Once it is detected that the pet has crossed the boundary, an alarm will be triggered to notify the pet owner to take intervention measures.

[0003] However, in practical applications, electronic fence systems face challenges. For example, in public spaces such as parks, when temporary activities are held and large metal structures (such as stages, tents) are set up, these temporary facilities will significantly interfere with global positioning system signals, causing the received signals by the pet positioning device to have multipath effects, which in turn causes the positioning data (including position, speed, and direction) to have instantaneous and irregular accuracy decreases, making it difficult for the system to accurately deduce the true motion trajectory of the pet, thereby affecting the accuracy of the out-of-bound risk prediction, which may result in delayed or false alarms, and the alarm is not timely.

[0004] At the same time, these large temporary structures can also physically form new, temporary global positioning system signal blind areas that are not pre-identified by the system. When the pet moves towards the boundary formed by the large metal structure and enters these new temporary signal blind areas, the system will lose the ability to track the pet. The system fails to timely and accurately issue an alarm due to dynamic changes in the environment and a decrease in data quality, resulting in the pet entering an area that cannot be tracked without the owner's knowledge, greatly increasing the difficulty and risk of finding, and the positioning tracking accuracy is low.

[0005] In summary, the technical problems in the related art need to be improved. SUMMARY

[0006] The main purpose of the embodiments of the present application is to provide a pet GPS positioning tracking fence alarm method and system that can combine signal interference areas and signal loss areas to achieve pet GPS positioning tracking and alarm triggering, improving the positioning tracking accuracy and the timeliness of the alarm.

[0007] In one aspect, the embodiments of the present application provide a pet GPS positioning tracking fence alarm method, comprising the following steps:

[0008] receive first positioning data sent by a target pet tracking device;

[0009] identify target change information caused by environmental interference according to the first positioning data, the target change information including position jump information and direction mutation information;

[0010] identify a signal interference area according to a frequency and a duration of the target change information;

[0011] identify a signal loss area according to signal quality information sent by the target pet tracking device, the signal quality information including a number of connectable satellites, signal strength, or positioning data disturbance value;

[0012] if the pet is in a poor signal area, obtain a moving direction of the pet, the moving direction being collected by a geomagnetic sensor built in the target pet tracking device, the poor signal area including the signal interference area or the signal loss area;

[0013] if the moving direction points to an electronic fence boundary or the poor signal area, generate a pet disconnection warning.

[0014] In some embodiments, the identifying the signal loss area according to the signal quality information sent by the target pet tracking device includes:

[0015] divide the electronic fence area to obtain a plurality of grid areas;

[0016] perform signal judgment on each grid area respectively within a preset detection duration;

[0017] if each signal quality information in the grid area is signal quality decrease, determine the grid area as the signal loss area.

[0018] In some embodiments, the identifying the target change information caused by environmental interference according to the first positioning data includes:

[0019] obtain instantaneous acceleration data of the pet, the instantaneous acceleration data being collected by an inertial sensor built in the target pet tracking device;

[0020] calculate a short-time theoretical displacement vector according to the instantaneous acceleration data;

[0021] calculate a real-time displacement vector according to the first positioning data;

[0022] compare the short-time theoretical displacement vector and the real-time displacement vector to obtain a consistency deviation, the consistency deviation including a direction deviation or a size deviation;

[0023] determine a consistency deviation threshold value;

[0024] if the consistency deviation is greater than the consistency deviation threshold value, take position change data in the real-time displacement vector as the position jump information, and take direction change data of the short-time theoretical displacement vector and the real-time displacement vector as the direction mutation information.

[0025] In some embodiments, the calculating a short-time theoretical displacement vector according to the instantaneous acceleration data comprises:

[0026] converting the instantaneous acceleration data from a device coordinate system to a geographic coordinate system according to gyroscope data;

[0027] integrating the converted instantaneous acceleration data according to a preset time step to obtain instantaneous speed change;

[0028] integrating the instantaneous speed change to obtain instantaneous displacement change;

[0029] generating the short-time theoretical displacement vector according to the instantaneous displacement change.

[0030] In some embodiments, the determining a consistency deviation threshold value comprises:

[0031] obtaining an acceleration change rate of the pet;

[0032] calculating motion state change smoothness according to the acceleration change rate;

[0033] determining a threshold adjustment amount according to the motion state change smoothness;

[0034] determining the consistency deviation threshold value according to the threshold adjustment amount and environmental noise.

[0035] In some embodiments, the determining a threshold adjustment amount according to the motion state change smoothness comprises:

[0036] obtaining a first vertical direction acceleration change feature of the target pet tracking device within a preset time window;

[0037] obtaining a horizontal direction acceleration change feature of the target pet tracking device within a preset time window;

[0038] identifying a complex motion mode of the pet according to the first vertical direction acceleration change feature and the horizontal direction acceleration change feature;

[0039] determining the threshold adjustment amount according to the complex motion mode and the motion state change smoothness.

[0040] In some embodiments, the determining the threshold adjustment amount according to the motion state change smoothness comprises:

[0041] obtaining first inertial sensing data of the target pet tracking device, and second positioning data and second inertial sensing data of the surrounding pet tracking device;

[0042] calculating a relative parameter between the target pet tracking device and the surrounding pet tracking device according to the first positioning data, the first inertial sensing data, the second positioning data and the second inertial sensing data, the relative parameter comprising a relative distance, a relative speed and a directional relationship;

[0043] identifying an interactive behavior pattern of the pet according to the relative parameter;

[0044] determining the threshold adjustment amount according to the interactive behavior pattern and the motion state change smoothness.

[0045] In some embodiments, the identifying the interactive behavior pattern of the pet according to the relative parameter comprises:

[0046] obtaining first vertical direction acceleration data of the target pet tracking device within a preset time window;

[0047] obtaining second vertical direction acceleration data of the surrounding pet tracking device within the preset time window;

[0048] identifying a target motion pattern of the pet in the vertical direction according to the first vertical direction acceleration data and the second vertical direction acceleration data, the target motion pattern comprising a synchronous motion pattern or an asynchronous motion pattern;

[0049] identifying the interactive behavior pattern according to the relative parameter and the target motion pattern.

[0050] In some embodiments, the identifying the target motion pattern of the pet in the vertical direction according to the first vertical direction acceleration data and the second vertical direction acceleration data comprises:

[0051] performing time sequence alignment on the first vertical direction acceleration data and the second vertical direction acceleration data;

[0052] calculating a vertical direction motion correlation between the first vertical direction acceleration data and the second vertical direction acceleration data after time sequence alignment;

[0053] if the vertical direction motion correlation is greater than a preset correlation threshold, determining that the target motion pattern is a synchronous motion pattern;

[0054] If the vertical direction motion correlation is less than a preset correlation threshold, it is determined that the target motion mode is an asynchronous motion mode.

[0055] In another aspect, an embodiment of the present application provides a pet GPS positioning tracking fence alarm system, comprising:

[0056] A data receiving module is configured to receive first positioning data sent by a target pet tracking device.

[0057] A change information identifying module is configured to identify target change information caused by environmental interference according to the first positioning data, wherein the target change information includes position jump information and direction mutation information.

[0058] A signal interference identifying module is configured to identify a signal interference area according to a frequency and a duration of the target change information.

[0059] A signal loss identifying module is configured to identify a signal loss area according to an electronic fence area and signal quality information sent by a plurality of target pet tracking devices, wherein the signal quality information includes a number of connectable satellites, signal strength or positioning data disturbance value.

[0060] A direction acquiring module is configured to acquire a moving direction of the pet if the pet is in a poor signal area, wherein the moving direction is acquired by a geomagnetic sensor built in the target pet tracking device, and the poor signal area includes the signal interference area or the signal loss area.

[0061] A pre-warning generating module is configured to generate a pet disconnection pre-warning if the moving direction points to an electronic fence boundary or the poor signal area.

[0062] The embodiments of the present application have at least the following beneficial effects: the embodiments of the present application first receive first positioning data sent by a target pet tracking device, identify target change information caused by environmental interference according to the first positioning data, then identify a signal interference area according to a frequency and a duration of the target change information, and identify a signal loss area according to an electronic fence area and signal quality information sent by a plurality of target pet tracking devices, acquire a moving direction of the pet if the pet is in a poor signal area, and generate a pet disconnection pre-warning if the moving direction points to an electronic fence boundary or the poor signal area, so that pet GPS positioning tracking and alarm triggering can be realized in combination with the signal interference area and the signal loss area, and the positioning tracking accuracy and the alarm timeliness are improved.

[0063] Additional features and advantages of the application will be set forth in the description that follows, and in part will be apparent from the description, or can be learned by practice of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the description and claims. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 This is a flowchart of a pet GPS location tracking fence alarm method according to an embodiment of the present invention;

[0066] Figure 2 This is a schematic diagram of a pet GPS positioning and tracking fence alarm system according to an embodiment of the present invention. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements.

[0068] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0069] The Global Positioning System (GPS) is a high-precision radio navigation positioning system based on artificial Earth satellites. GPS provides accurate geographical location, vehicle speed, and precise time information in most parts of the world and in near-Earth space.

[0070] In the related art, in modern pet management, electronic fence systems have become a common tool for limiting the activities of pets within a certain area and preventing them from accidentally leaving. Such systems usually rely on tracking devices worn by pets that can periodically obtain global positioning system (GPS) positioning information. After the system receives these positioning data, it compares them with the pre-set electronic fence boundary to determine whether the current location of the pet is still within the allowed activity area. Once the pet is detected crossing the boundary, the system will trigger an alarm to notify the pet owner to take intervention measures. This basic working mode, that is, periodically obtaining positioning information and making boundary judgments, constitutes the core operating logic of most current pet electronic fence systems. In order to improve the timeliness of the warning, some schemes not only alarm when the pet reaches the boundary, but also introduce a prediction mechanism. When the system detects that the pet is moving towards the boundary at a faster speed, it will make a risk assessment in advance. This assessment usually calculates a predicted trajectory, and if this trajectory will cross the fence boundary in a short time, the system will trigger an alarm in advance, giving the owner more reaction time. This early warning mechanism aims to give the owner enough reaction time to take intervention measures before the pet actually crosses the boundary or enters a signal blind area where it cannot be tracked, thereby effectively preventing the pet from getting lost.

[0071] However, in actual application, such systems will face some inherent challenges. For example, in a typical urban park environment, such electronic fence systems are widely used. The park usually contains open lawns, winding paths, and some fixed landscape facilities such as pavilions and sparse trees. At the beginning of deployment, the system will accurately draw the electronic fence boundary of the park through on-site survey and information accumulation, and identify those permanently poor signal reception areas due to terrain shielding or building structure, such as under a dense forest at the edge of the park or an underground passage entrance. The system will optimize the early warning logic for these known poor signal areas to ensure that an alarm is issued before the pet approaches these areas.

[0072] However, as a public space, the environment of the park is not constant. Imagine that on a certain weekend, a large community event is being prepared in the park, such as a music festival or a temporary exhibition. For this event, staff will set up a series of large temporary structures in an area of the park, usually in the open area near the edge. These structures may include tall metal stage skeletons, large tents with metal supports, temporary broadcast towers, or even metal supports for large display screens. The characteristic of these temporary facilities is that they are not part of the park's regular landscape and therefore were not considered in the system's initial environmental map and signal poor area identification.

[0073] When pets move around these temporary structures, these large metal objects have a significant impact on the global positioning system signals. They reflect, scatter, and even partially absorb satellite signals, causing the signal path to become complex, resulting in a so-called multipath effect. This means that the signal received by the pet positioning device is no longer a single direct signal, but contains a large number of delayed signals reflected back from the surrounding metal structures. This phenomenon can cause the position information calculated by the positioning device to deviate momentarily, resulting in irregular jumps in the position point in a short period of time, or abnormal fluctuations in the calculation of speed and direction. These data anomalies are not complete signal loss, but rather precision problems caused by signal quality degradation, making the pet movement data received by the system unreliable.

[0074] In this case of impaired data quality, the original out-of-bound risk prediction algorithm of the system faces serious challenges. Since the input position, speed, and direction data itself has a large degree of uncertainty, the system may not be able to accurately infer the true motion trajectory of the pet. For example, the pet may be moving steadily towards the boundary at a constant speed, but due to signal interference, the data received by the system shows that its speed fluctuates rapidly, and even the direction deviates temporarily. This can cause the prediction mechanism to make false judgments: it may mistakenly believe that the pet is slowing down or turning, thus underestimating the risk of crossing the boundary and delaying the issuance of an alarm; or it may not generate a reliable prediction result due to excessive data fluctuations, thus choosing not to issue an alarm to avoid false positives.

[0075] Further complexity lies in the fact that these temporarily erected large structures, in addition to causing signal interference, can themselves physically form new, temporary signal shadow areas. For example, a huge metal stage backdrop or a dense exhibition tent can form a temporary, unpreviously identified global positioning system signal blind area in its interior or immediately behind it. When the pet moves in the direction of the boundary formed by these temporary structures under signal interference and eventually enters these new temporary blind areas, the system will completely lose the ability to track the pet.

[0076] Therefore, before the pet enters these signal blind areas formed by temporary structures, which are not pre-set by the system, the system cannot effectively predict the risk of crossing the boundary based on accurate speed and direction information, as the pet's movement data has become unreliable in the interference area. This means that at the most critical moment when early warning is needed, the system fails to issue timely and accurate alarms due to dynamic changes in the environment and a decline in data quality, causing the pet to enter an area that cannot be tracked without the owner's awareness, greatly increasing the difficulty and risk of finding, and reducing the accuracy of positioning and tracking and the timeliness of the alarm.

[0077] In the pet GPS positioning tracking fence alarm system, when a dynamic and local GPS signal interference caused by a large-scale temporary metal structure (such as a mobile stage or an exhibition tent) appears in the pet activity area, the challenge of instantaneous and irregular accuracy reduction of positioning data (including position, speed and direction) and the formation of a temporary signal blind area not previously identified by the system needs to be overcome, so that the pet's imminent out-of-bound risk can be accurately identified and an alarm can be sent in time to avoid the pet from getting lost even if the motion data is not reliable.

[0078] Therefore, the embodiments of the present application can more accurately and earlier predict the pet's risk of disconnection by introducing the identification of target change information (including position jump information and direction mutation information) caused by environmental interference, the identification of signal interference areas and signal loss areas, and the judgment of the pet's moving direction in the poor signal area, thereby effectively solving the problems of early warning delay or misjudgment caused by dynamic changes in the environment and signal quality reduction in the prior art.

[0079] The embodiments of the present application will be specifically explained below in combination with the drawings:

[0080] Figure 1 is an optional flowchart of a pet GPS positioning tracking fence alarm method provided by the embodiments of the present application, Figure 1 The method in can include but is not limited to steps S101 to S106.

[0081] Step S101, receiving first positioning data sent by a target pet tracking device;

[0082] Step S102, identifying target change information caused by environmental interference according to the first positioning data, the target change information including position jump information and direction mutation information;

[0083] Step S103, identifying a signal interference area according to the frequency and duration of the target change information;

[0084] Step S104, identifying a signal loss area according to the electronic fence area and signal quality information sent by a plurality of target pet tracking devices, the signal quality information including the number of connectable satellites, signal strength or positioning data disturbance value;

[0085] Step S105, if the pet is in a poor signal area, acquiring the pet's moving direction, the moving direction being collected by a geomagnetic sensor built in the target pet tracking device, the poor signal area including the signal interference area or the signal loss area;

[0086] Step S106, if the moving direction points to the electronic fence boundary or the poor signal area, generating a pet disconnection warning.

[0087] The steps S101 to S106 shown in the embodiments of the present application can realize pet GPS positioning tracking and alarm triggering in combination with signal interference areas and signal loss areas, thereby improving positioning tracking accuracy and alarm timeliness.

[0088] In some embodiments, steps S101 to S106, the implementation environment of the present embodiment generally includes a pet tracking device, a data receiving and processing server (or cloud platform), and a user terminal (such as a mobile phone APP). The pet tracking device is responsible for data acquisition and sending, the server is responsible for data receiving, processing, analysis, and early warning generation, and the user terminal is used to receive early warning information and manage electronic fences.

[0089] First positioning data sent by a target pet tracking device can be received. For example, a GPS module can be built into the target pet tracking device, which periodically (for example, once every second or once every five seconds) acquires the current position, speed, timestamp, and other first positioning data of the pet, and sends these first positioning data to the data receiving server through a wireless communication module (such as a cellular network, Bluetooth, or Wi-Fi). The data receiving server can be configured with a corresponding communication interface and data caching mechanism to ensure that it can stably and efficiently receive data streams from multiple pet tracking devices. For example, the server can use a message queue to queue the received positioning data for processing, avoiding data loss or processing delays.

[0090] Then, target change information caused by environmental interference is identified according to the first positioning data, wherein the target change information includes position jump information and direction mutation information. In actual application, GPS signals are easily affected by environmental factors such as buildings, trees, and water bodies, resulting in multipath effects or signal shielding, causing the positioning data to fluctuate greatly instantaneously, i.e., “position jump information” or “direction mutation information”. In order to identify these abnormalities, the system can use multiple methods. For example, a position change threshold and a direction change threshold can be set. When the displacement or direction change calculated from the first positioning data received continuously twice exceeds the pre-set threshold in a very short time (for example, within 1 second), and such change is inconsistent with the normal movement pattern of the pet, it can be judged that there is position jump information or direction mutation information. Another implementation way is that the system can perform smoothing processing on the received first positioning data, for example, using a moving average filter or a Kalman filter. By comparing the difference between the original data and the smoothed data, if the difference exceeds a certain range, it is considered that there is target change information caused by environmental interference.

[0091] According to the frequency and duration of the target change information, the signal interference area is identified. For example, the number of position jump information and direction mutation information appearing in a unit time in a certain geographic area (i.e. frequency) and the duration of each anomaly can be counted. If the frequency of target change information in a certain area is high and the duration is long, it indicates that there is persistent environmental interference in this area, which can be marked as a signal interference area. Specifically, the system can divide the electronic fence area into multiple grids, and count the target change information in each grid area. When the frequency of target change information in a certain grid area exceeds the preset threshold and the average duration also exceeds the preset threshold within a preset observation period, the grid area is identified as a signal interference area.

[0092] According to the signal quality information sent by the electronic fence area and the plurality of target pet tracking devices, the signal loss area is identified, wherein the signal quality information is an index for measuring the reliability of positioning data, which can include the number of connectable satellites (the more satellites, the higher the positioning accuracy), signal strength (the stronger the signal, the less likely to be disturbed) or positioning data disturbance value (the greater the data fluctuation, the more unstable the positioning). The signal loss area is a blind area for pet tracking, and identifying these areas is crucial for early warning. The signal quality information sent by multiple pet tracking devices at different time points can be used for comprehensive judgment. For example, when the number of connectable satellites reported by multiple pet tracking devices in a certain area is generally lower than the preset threshold, or the signal strength is generally lower than the preset threshold, or the positioning data disturbance value is generally higher than the preset threshold, the area is identified as a signal loss area. At the same time, the electronic fence area can be divided into grids, and the signal quality information can be continuously monitored in each grid area. If the signal quality information received in a certain grid area shows a decrease in signal quality (e.g. zero connectable satellites or extremely low signal strength) over a period of time, the grid area is determined to be a signal loss area.

[0093] If the pet is in a signal poor area, the moving direction of the pet is obtained, wherein the moving direction is collected by a geomagnetic sensor built-in the target pet tracking device, and the signal poor area includes a signal interference area or a signal loss area. When the pet enters the signal poor area, the traditional location-based early warning method will fail because the positioning data can be unreliable or even missing. At this time, obtaining the moving direction of the pet becomes the key supplementary information for judging the risk of disconnection. The geomagnetic sensor built-in the target pet tracking device can sense the earth's magnetic field in real time and calculate the current direction of the device according to the earth's magnetic field, thereby inferring the moving direction of the pet. For example, the geomagnetic sensor can periodically collect geomagnetic field data and convert these data into the heading angle of the device in the geographical coordinate system through an internal algorithm. When the system determines that the current position of the pet is in the identified signal interference area or signal loss area, the target pet tracking device uploads the moving direction data collected by the geomagnetic sensor.

[0094] If the moving direction points to the boundary of the electronic fence or the signal poor area, a pet disconnection warning is generated. In the case that the pet is in a signal poor area and the positioning data is unreliable, the moving direction of the pet provides an important risk indication. The moving direction of the pet can be continuously monitored, and if the moving direction of the pet continues to point to the boundary direction of the electronic fence, even if accurate positioning data cannot be obtained at this time, it indicates that the pet has the risk of crossing the boundary. Similarly, if the moving direction of the pet points to the depth of the identified signal poor area (especially the signal loss area), it indicates that the pet may be entering a more difficult to track area and has the risk of disconnection. In this case, the system will immediately generate a pet disconnection warning and notify the pet owner through the user terminal (such as a mobile phone APP). For example, the system can preset a fan-shaped area pointing to the boundary of the electronic fence, and if the moving direction of the pet falls into the fan-shaped area and lasts for a period of time, the early warning is triggered. Alternatively, the moving direction of the pet can be judged according to the shape and position of the signal poor area whether it is moving towards the center or deeper of these areas.

[0095] It can be understood that the target pet tracking device refers to a device worn on a pet, which can acquire and send positioning data and / or inertial sensor data, geomagnetic sensor data, and other information. The device usually has a built-in GPS module, inertial sensors (such as accelerometers, gyroscopes), and geomagnetic sensors for real-time monitoring of the pet's position, motion state, and orientation. The target change information refers to abnormal changes in position or direction in the positioning data due to environmental interference (such as multipath effect, signal obstruction). The target change information includes position jump information and direction mutation information. The position jump information refers to a large change in the pet's position in a short time that is not physically possible, and the direction mutation information refers to a sharp change in the pet's movement direction in a short time that is not physically possible. The signal interference area refers to an area where the GPS signal is disturbed by obstacles in the environment (such as buildings, metal structures), resulting in a decrease in positioning data accuracy or abnormalities. The signal loss area refers to an area where the GPS signal is completely blocked or attenuated, causing the pet tracking device to be unable to acquire valid positioning data, which is commonly referred to as a signal blind area. The electronic fence area refers to a pre-set virtual geographic area for limiting the pet's activity range.

[0096] Through the above technical solutions, the embodiment introduces the identification of target change information caused by environmental interference, dynamic identification of signal interference areas and signal loss areas, and the use of geomagnetic sensors to obtain the pet's moving direction for auxiliary judgment in signal poor areas, to construct a more robust and intelligent pet GPS positioning and tracking fence alarm method. This allows the system to more accurately predict the risk of pet disconnection in complex and variable environments, significantly improving the practicality and reliability of the pet electronic fence system, and effectively solving the pain points of early warning delay or misjudgment in the prior art.

[0097] In some embodiments, in step S104, identifying the signal loss area according to the electronic fence area and the signal quality information sent by the plurality of target pet tracking devices can include but is not limited to the following steps:

[0098] Divide the electronic fence area to obtain a plurality of grid areas;

[0099] In a preset detection time, perform signal judgment on each grid area respectively;

[0100] If each signal quality information in the grid area is signal quality decline, the grid area is determined as a signal loss area.

[0101] In some embodiments, due to only macroscopic judgment, the identification of signal loss area may not be fine enough or there may be misjudgment, thereby affecting the accuracy and timeliness of the pet lost warning. Therefore, it is necessary to finely divide and judge each one of the electronic fence area to improve the accuracy and reliability of the identification. The electronic fence area can be divided into a plurality of grid areas first. For example, a larger electronic fence area can be subdivided into a plurality of smaller, independent geographic units, i.e. grid areas. These grid areas can be regular, such as square or rectangular, or irregularly divided according to the actual terrain or signal coverage characteristics. The purpose is to realize the localization and fine monitoring of the signal condition in the electronic fence area, so as to more accurately locate the specific position of the poor signal.

[0102] In the preset detection duration, signal judgment is performed on each grid area respectively, which can continuously collect and analyze the signal quality information sent by the target pet tracking device in each grid area. The preset detection duration can be set according to actual needs, for example, several minutes or several hours, which aims to exclude transient or accidental signal fluctuations and ensure the judgment of signal quality to have certain continuity and stability.

[0103] If each signal quality information in the grid area is signal quality decline, the grid area is determined as signal loss area. This means that only when all the monitored target pet tracking devices in the grid area show a downward trend or are lower than the acceptable threshold in the preset detection duration, the grid area is identified as a signal loss area. This strict judgment condition aims to avoid misjudging the entire area as a signal loss area due to a single device failure or temporary signal interference, thereby improving the accuracy and robustness of the identification.

[0104] The embodiment realizes the localized monitoring of the signal condition by refining the electronic fence area into a plurality of grid areas. This fine division enables the system to independently evaluate the signal quality for each small area, rather than making a general judgment of the entire large area. Further, continuous signal judgment is performed on each grid area in the preset detection duration, and all signal quality information in the area is required to show a downward trend, which effectively filters transient signal fluctuations or local, non-universal signal problems. Therefore, only when there is continuous and universal signal quality decline in a certain grid area, it will be accurately identified as a signal loss area. This mechanism ensures that the identification of the signal loss area is based on stable and reliable data, avoiding misjudgment due to accidental signal problems.

[0105] To more clearly illustrate the technical solutions, specific examples are used in the following for explanation. Assume that an electronic fence area covers a courtyard with an area of 1000 square meters. In order to accurately identify the signal loss area, the courtyard is divided into 100 grid areas of 10 square meters. The system continuously monitors the signal quality information, such as the number of connectable satellites and signal strength, sent by the target pet tracking device located in each grid area. Within a preset detection time length (for example, 5 minutes), if all target pet tracking devices in a certain grid area continuously report that the number of connectable satellites is less than 3 and the signal strength is less than -100 dBm, the grid area will be determined by the system as a signal loss area. For example, in a corner of the courtyard, due to the shielding of tall buildings, GPS signals cannot be effectively received for a long time. Through this grid-based and continuous judgment mechanism, the grid area in the corner will be accurately identified as a signal loss area. When the pet enters this area and moves towards the fence boundary, even if the GPS positioning data is unavailable, the system can generate a disconnection warning in a timely manner based on the geomagnetic sensor data and the identified signal loss area information.

[0106] Through the above technical solutions, the embodiment can realize fine and accurate identification of the signal loss area in the electronic fence area. Compared with the judgment based on only the overall signal quality, the embodiment significantly improves the accuracy of the signal loss area positioning through grid division and multi-dimensional and continuous signal judgment, reducing the risk of false positives and false negatives. This makes the triggering of the pet disconnection warning more timely and reliable, effectively improving the level of pet safety protection and providing pet owners with a more worry-free tracking and management experience.

[0107] In some embodiments, in step S102, identifying the target change information caused by environmental interference according to the first positioning data can include but is not limited to the following steps:

[0108] Step S201, acquiring instantaneous acceleration data of the pet, the instantaneous acceleration data being collected by an inertial sensor built in the target pet tracking device;

[0109] Step S202, calculating a short-time theoretical displacement vector according to the instantaneous acceleration data;

[0110] Step S203, calculating a real-time displacement vector according to the first positioning data;

[0111] Step S204, comparing the short-time theoretical displacement vector and the real-time displacement vector to obtain a consistency deviation, the consistency deviation including a direction deviation or a size deviation;

[0112] Step S205, determining a consistency deviation threshold;

[0113] If the consistency deviation is greater than the consistency deviation threshold, the position change data in the real-time displacement vector is taken as the position jump information, and the direction change data between the short-time theoretical displacement vector and the real-time displacement vector is taken as the direction mutation information.

[0114] In some embodiments, the instantaneous acceleration data of the pet can be obtained first, where the instantaneous acceleration data refers to the acceleration change experienced by the target pet tracking device within a very short time, which is collected by the inertial sensor built-in the target pet tracking device. The inertial sensor usually includes an accelerometer and a gyroscope, which can monitor the motion state of the device in three-dimensional space in real time and provide high-frequency and high-precision motion data.

[0115] Then, the short-time theoretical displacement vector is calculated according to the instantaneous acceleration data. The short-time theoretical displacement vector reflects the displacement inferred according to the inertial motion trend of the pet within a short time. For example, by integrating the acceleration data, the instantaneous speed change can be obtained, and by integrating the instantaneous speed change, the instantaneous displacement change can be obtained, and then the short-time theoretical displacement vector is generated. The real-time displacement vector is calculated according to the first positioning data. The real-time displacement vector represents the actual moving distance and direction of the pet between two consecutive positioning points.

[0116] The short-time theoretical displacement vector and the real-time displacement vector are compared to obtain the consistency deviation. The consistency deviation includes a direction deviation or a size deviation. The direction deviation refers to the difference in direction between the two displacement vectors, and the size deviation refers to the difference in displacement distance between the two displacement vectors. This comparison aims to find abnormal positioning data caused by external environmental interference (such as GPS signal drift, multipath effect, etc.). In order to accurately determine whether there is environmental interference, it is necessary to determine the consistency deviation threshold. The threshold is used to define the boundary between normal motion and abnormal positioning data. The consistency deviation threshold can be dynamically or statically determined according to the motion state of the pet, the environmental noise level, and historical data analysis, etc.

[0117] If the consistency deviation is greater than the consistency deviation threshold, it indicates that the positioning data may have been disturbed by the environment. The position change data in the real-time displacement vector can be taken as the position jump information, which indicates that the pet has an abnormal position change within a short time that does not conform to its inertial motion law. At the same time, the direction change data between the short-time theoretical displacement vector and the real-time displacement vector is taken as the direction mutation information, which indicates that the actual positioning direction of the pet is significantly different from the direction inferred based on inertial motion. In this way, the position and direction abnormalities caused by environmental interference can be effectively distinguished from normal pet motion.

[0118] By the technical solution, the accuracy and reliability of pet positioning and tracking can be improved. By introducing inertial sensor data to check GPS positioning data, false position jumps and direction mutations caused by environmental interference (such as GPS signal drift, multipath effect, signal shielding, etc.) can be effectively filtered out or identified, and unnecessary fence alarms or disconnection warnings caused by misjudgment can be avoided. This makes the system more accurately reflect the real motion trajectory and state of the pet, thereby improving the intelligent level of pet safety management, reducing the false alarm rate, and improving the user experience.

[0119] In some embodiments, in step S202, calculating the short-time theoretical displacement vector according to the instantaneous acceleration data can include but is not limited to the following steps:

[0120] Converting the instantaneous acceleration data from the device coordinate system to the geographic coordinate system according to the gyroscope data;

[0121] Integrating the instantaneous acceleration data converted in the coordinate system according to the preset time step to obtain the instantaneous speed change;

[0122] Integrating the instantaneous speed change to obtain the instantaneous displacement change;

[0123] Generating the short-time theoretical displacement vector according to the instantaneous displacement change.

[0124] In some embodiments, the instantaneous acceleration data can be first converted from the device coordinate system to the geographic coordinate system according to the gyroscope data. For example, the attitude information of the device in the three-dimensional space, such as the pitch angle, roll angle and yaw angle, can be obtained by using the built-in gyroscope data of the target pet tracking device. Through these attitude information, a rotation matrix can be constructed to convert the original instantaneous acceleration data measured in the device's own coordinate system to the geographic coordinate system (such as the station-centered coordinate system) which is relatively fixed to the earth's surface. The purpose of this conversion is to ensure that the subsequent displacement calculation is carried out in a unified and stable reference system, thereby eliminating the errors introduced by the change of device attitude.

[0125] Then, according to a preset time step, the instantaneous acceleration data converted by the coordinate system is integrated to obtain an instantaneous speed change, and the instantaneous speed change is integrated to obtain an instantaneous displacement change. Numerical integration methods can be used to achieve this. For example, Euler integration, trapezoidal integration or Simpson integration algorithms can be used. The preset time step refers to the time interval of data sampling, for example, 100 samples per second, and the time step is 0.01 seconds. By integrating the acceleration once, the speed change in each time step can be estimated, and by integrating the speed change twice, the displacement change in the corresponding time period can be estimated. According to the instantaneous displacement change, a short-time theoretical displacement vector is generated. This vector not only contains the size information of the displacement, but also contains the direction information of the displacement, and can accurately reflect the theoretical moving track and trend of the pet in a short time.

[0126] Through the above technical solutions, the embodiment can accurately calculate a short-time theoretical displacement vector based on the inertial motion data of the pet itself. This provides a reliable and independent reference benchmark for subsequent identification of target change information caused by environmental interference. Compared with displacement judgment relying only on GPS positioning data, the embodiment can effectively distinguish between real pet movement and position or direction abnormalities caused by signal interference, significantly improving the recognition accuracy and robustness of position jump information and direction mutation information, thereby providing more accurate data support for subsequent signal interference area identification and pet loss warning.

[0127] In some embodiments, in step S205, determining the consistency deviation threshold can include, but is not limited to, the following steps:

[0128] Step S301, obtaining the acceleration change rate of the pet;

[0129] Step S302, calculating the motion state change smoothness according to the acceleration change rate;

[0130] Step S303, determining the threshold adjustment amount according to the motion state change smoothness;

[0131] Step S304, determining the consistency deviation threshold according to the threshold adjustment amount and the environmental noise.

[0132] In some embodiments, the determination manner of the consistency deviation threshold value fails to fully consider the actual motion state of the pet and the dynamic changes of the environment, which may result in insufficient recognition accuracy of the environmental disturbance, thereby affecting the accuracy of the pet disconnection warning. For example, when the pet is in a state of intense motion, its displacement change may be large, and if a fixed threshold value is used, normal motion may be misjudged as a disturbance; conversely, when the environmental noise is large, if the threshold value is set too low, the real signal anomaly may not be effectively identified. Therefore, the acceleration change rate of the pet can be obtained first, which is the rate of change of the instantaneous acceleration of the pet with respect to time, and which can reflect the smoothness or intensity of the pet's motion. For example, when the pet suddenly accelerates or decelerates, the acceleration change rate will significantly increase; when the pet moves at a constant speed, the acceleration change rate tends to zero. This data can be obtained by twice differentiating or differentiating the instantaneous acceleration data collected by the inertial sensor built-in the target pet tracking device.

[0133] Then, according to the acceleration change rate, the motion state change smoothness is calculated, wherein the motion state change smoothness refers to the smoothness of the pet's motion trajectory. For example, when the acceleration change rate of the pet is small, it indicates that the change of its motion state is relatively gentle, and at this time, the motion state change smoothness is high; conversely, when the acceleration change rate is large, it indicates that the change of its motion state is relatively intense, and at this time, the motion state change smoothness is low. This smoothness can be calculated according to the inverse of the acceleration change rate, or through a pre-set mapping function, the purpose of which is to quantify the motion characteristics of the pet.

[0134] Then, according to the acceleration change rate, the motion state change smoothness is calculated, wherein the motion state change smoothness refers to the smoothness of the pet's motion trajectory. For example, when the acceleration change rate of the pet is small, it indicates that the change of its motion state is relatively gentle, and at this time, the motion state change smoothness is high; conversely, when the acceleration change rate is large, it indicates that the change of its motion state is relatively intense, and at this time, the motion state change smoothness is low. This smoothness can be calculated according to the inverse of the acceleration change rate, or through a pre-set mapping function, the purpose of which is to quantify the motion characteristics of the pet.

[0135] Finally, the consistency deviation threshold is determined according to the threshold adjustment amount and the environmental noise. The environmental noise refers to external interference factors affecting the accuracy of positioning data, such as strength fluctuations of GPS signals, multipath effects, building obstructions, etc. When determining the consistency deviation threshold, the environmental noise can be taken into account. For example, when the environmental noise is large, even if the pet is moving normally, its positioning data may have large fluctuations, and the consistency deviation threshold needs to be appropriately increased to tolerate larger normal deviations; on the contrary, when the environmental noise is small, a stricter threshold can be used. The consistency deviation threshold can be calculated by combining the basic threshold, the threshold adjustment amount, and the environmental noise, for example, by using weighted averaging or table lookup to determine.

[0136] To more clearly illustrate the technical solutions, specific examples are used for explanation below. Assume that the target pet tracking device is tracking a pet dog. In one case, the pet dog is walking slowly or resting on the grass. At this time, its acceleration change rate is low, and the calculated motion state change smoothness is high. The system will determine a smaller threshold adjustment amount according to the higher smoothness, and finally determine a relatively strict consistency deviation threshold in combination with the current lower environmental noise (e.g. in an open area). In this way, even slight positioning data jumps or direction mutations are more likely to be identified as environmental interference, so that timely warnings can be issued.

[0137] In another case, the pet dog is playing and chasing another dog, and its motion state is characterized by frequent acceleration, deceleration, and sharp turns. At this time, its acceleration change rate will significantly increase, and the calculated motion state change smoothness will be low. The system will determine a larger threshold adjustment amount according to the lower smoothness, and finally determine a relatively loose consistency deviation threshold in combination with the possible environmental noise (e.g. at the edge of a forest). This dynamic adjustment can effectively distinguish between displacement deviations caused by the pet's own intense motion and real environmental interference, avoiding misjudgment of the pet's normal play behavior as a loss of contact risk, thereby reducing unnecessary warnings and improving user experience.

[0138] In addition, if the pet dog enters an area with poor signal (e.g. the entrance to a basement), even if the pet is moving smoothly, the environmental noise will significantly increase. The system will adjust the consistency deviation threshold accordingly based on the increased environmental noise, so that it can tolerate larger positioning data fluctuations and avoid frequent false alarms due to signal quality degradation. Only when the deviation exceeds a reasonable range will a warning be issued.

[0139] By the technical solution, the consistency deviation threshold can be dynamically adjusted according to the real-time motion state of the pet and the environmental noise level. This significantly improves the recognition accuracy and robustness of the target change information caused by environmental interference, effectively avoiding false positives or false negatives caused by fixed thresholds. Specifically, when the pet is normally active, false positives can be effectively reduced, avoiding unnecessary disconnection warnings; when the environmental signal quality is poor, it can also more accurately determine whether it is a real disturbance, thereby ensuring the timeliness and accuracy of the pet disconnection warning, and improving the reliability of the entire pet GPS positioning and tracking fence alarm system.

[0140] In some embodiments, in step S303, determining the threshold adjustment amount according to the motion state change smoothness can include but is not limited to the following steps:

[0141] Obtaining a first vertical direction acceleration change feature of the target pet tracking device within a preset time window;

[0142] Obtaining a horizontal direction acceleration change feature of the target pet tracking device within a preset time window;

[0143] According to the first vertical direction acceleration change feature and the horizontal direction acceleration change feature, identifying the complex motion mode of the pet;

[0144] According to the complex motion mode and the motion state change smoothness, determining the threshold adjustment amount.

[0145] In some embodiments, since only the motion state change smoothness may not fully reflect the real motion state of the pet in a complex environment, for example, when the pet performs nonlinear motion such as jumping and rolling, the acceleration change feature may be complex, and if not distinguished, the determined threshold adjustment amount may not be accurate, thereby affecting the accuracy of the consistency deviation threshold and increasing the risk of false positives or false negatives. Therefore, a first vertical direction acceleration change feature of the target pet tracking device within a preset time window can be obtained, wherein the first vertical direction acceleration change feature refers to the mode or trend of the acceleration of the target pet tracking device in the vertical direction with respect to time, for example, it can include statistical quantities such as peak value, valley value, average value, variance, and change frequency of the vertical direction acceleration. The purpose is to capture the motion characteristics of the pet in the vertical direction, such as jumping, climbing, or descending, etc.

[0146] Then, a horizontal direction acceleration change feature of the target pet tracking device within a preset time window is obtained, wherein the horizontal direction acceleration change feature refers to the mode or trend of the acceleration of the target pet tracking device in the horizontal direction with respect to time, for example, it can include instantaneous size, direction change rate, average speed, etc. of the horizontal direction acceleration. The purpose is to capture the motion characteristics of the pet in the plane, such as running, turning, or stopping, etc.

[0147] According to the first vertical direction acceleration change feature and the horizontal direction acceleration change feature, a complex motion mode of the pet is identified. For example, by analyzing the acceleration change features of the pet in the vertical and horizontal directions, it can be determined whether the pet is currently in a non-linear or violent motion state such as jumping, rolling, sudden stopping, sharp turning, etc. For example, when the vertical direction acceleration shows significant periodic peaks, and the horizontal direction acceleration changes little, it may indicate that the pet is jumping; when the horizontal direction acceleration shows sharp and rapid direction changes, it may indicate that the pet is turning sharply. The identification of these complex motion modes aims to more accurately reflect the real motion state of the pet at a certain moment.

[0148] Finally, according to the complex motion mode and the motion state change smoothness, the threshold adjustment amount is determined. For example, when the pet is in a stable motion state (high motion state change smoothness), even if there is slight acceleration change, it may not be necessary to make a large threshold adjustment; while when the pet is in a complex motion mode (such as jumping or rolling), even if the motion state change smoothness is low, the threshold adjustment amount may need to be appropriately enlarged or reduced according to the characteristics of the complex motion mode, in order to avoid misjudging the positioning data fluctuations caused by normal motion as environmental interference, or missing real interference.

[0149] In order to more clearly illustrate the technical solutions, specific examples are used for explanation below. Assume that a pet dog is playing in a park, during which it frequently performs actions such as jumping and rolling. The target pet tracking device continuously collects its acceleration data. First, the system obtains the first vertical direction acceleration change feature and the horizontal direction acceleration change feature of the pet dog within a preset time window (for example, the last 5 seconds). For example, the vertical direction acceleration may show periodic high-amplitude fluctuations, while the horizontal direction acceleration may show rapid and irregular direction changes. Then, according to these acceleration change features, the system identifies that the pet dog is currently in a complex motion mode of “jumping” and “rolling”. At the same time, the system also calculates the motion state change smoothness of the pet. Since the pet dog is performing violent motion, its motion state change smoothness may be relatively low. According to the complex motion mode of “jumping” and “rolling” that has been identified, the threshold adjustment amount can be intelligently adjusted. For example, the system may appropriately relax the consistency deviation threshold when identifying these complex motion modes, so that it can tolerate the data fluctuations caused by the normal violent motion of the pet. Thus, even if the positioning data of the pet dog fluctuates greatly when it is jumping and rolling, the system will not misjudge it as signal interference, thereby avoiding the generation of false alarms and improving the accuracy of the alarm and user experience.

[0150] By the technical solution, the complex motion mode of the pet is recognized, normal acceleration fluctuation caused by intense motion of the pet itself and abnormal positioning data fluctuation caused by environmental interference can be distinguished. This significantly improves the accuracy of target change information recognition, effectively reduces false positives caused by normal complex motion of the pet, and ensures that a warning can be generated in time and accurately when a real signal interference occurs. Therefore, the embodiment has higher robustness and reliability in actual application, and is especially suitable for the case that the pet moves in a variable and complex outdoor environment.

[0151] In some embodiments, in step S303, determining the threshold adjustment amount according to the motion state change smoothness can include but is not limited to the following steps:

[0152] In step S401, first inertial sensing data of a target pet tracking device and second positioning data and second inertial sensing data of a surrounding pet tracking device are acquired.

[0153] In step S402, relative parameters between the target pet tracking device and the surrounding pet tracking device are calculated according to the first positioning data, the first inertial sensing data, the second positioning data and the second inertial sensing data, the relative parameters including relative distance, relative speed and direction relationship.

[0154] In step S403, an interactive behavior mode of the pet is recognized according to the relative parameters.

[0155] In step S404, a threshold adjustment amount is determined according to the interactive behavior mode and the motion state change smoothness.

[0156] In some embodiments, in the scene where multiple pets coexist or complex interactive behaviors exist between pets, only determining the threshold adjustment amount according to the motion state change smoothness of a single pet can not fully reflect the actual motion complexity, thereby affecting the accuracy of the consistency bias threshold, and can cause misjudgment or missed judgment of environmental interference. Therefore, the data of the surrounding pet tracking device is introduced to recognize the interactive behavior mode of the pet, so as to more accurately adjust the threshold. First, the first inertial sensing data of the target pet tracking device and the second positioning data and the second inertial sensing data of the surrounding pet tracking device are acquired. The inertial sensing data refers to data collected by an inertial sensor (such as an accelerometer or a gyroscope) built in the target pet tracking device, which is used to describe the motion state of the target pet itself, such as instantaneous acceleration and angular velocity.

[0157] Then, according to the first positioning data, the first inertial sensing data, the second positioning data and the second inertial sensing data, relative parameters between the target pet tracking device and the surrounding pet tracking devices are calculated to quantify the spatial and motion relationship between the target pet and the surrounding pets. The relative parameters include relative distance, relative speed and direction relationship. The relative distance can be calculated by the Euclidean distance or the geographic distance between the first positioning data and the second positioning data. The relative speed can be obtained by differentiating the change rate of the respective positioning data over time, or by differentiating the instantaneous speed obtained by integrating the inertial sensing data. The direction relationship can be determined by comparing the motion directions or attitude information of the two devices in the geographic coordinate system, for example, by calculating the included angle of the motion direction vectors of the two devices. The calculation of these relative parameters aims to accurately depict the mutual position, speed and orientation relationship between the pets.

[0158] According to the relative parameters, the interaction behavior patterns of the pets are identified. For example, when the relative distance between the two pet tracking devices remains within a certain range, the relative speed is small, and the direction relationship shows that the two are oriented consistently or move closer to each other, it can be identified as a "follow" or "move side by side" pattern. If the relative distance changes rapidly, the relative speed is large, and the direction relationship shows that the two frequently change orientation or chase each other, it can be identified as a "play" or "tussle" pattern. By pattern recognition of these relative parameters, it can be distinguished whether the pets are moving independently or in some interactive state.

[0159] Finally, according to the interaction behavior patterns and the motion state change smoothness, the threshold adjustment amount is determined. For example, if the pet is identified to be in a "play" mode, even if its own motion state change smoothness is low (i.e., the motion is intense), the system will correct the threshold adjustment amount according to the interaction behavior pattern, so that it is not too low, thereby avoiding misjudgment of normal play behavior as position jump caused by environmental interference. Conversely, if the pet is in a "still" or "slow moving" mode, but its motion state change smoothness suddenly decreases, it may indicate an abnormality, and the threshold adjustment amount will be adjusted accordingly to increase the sensitivity.

[0160] Through the above technical solutions, the embodiment can more accurately determine the consistency deviation threshold, especially in complex scenarios where multiple pets coexist and interact. By considering the interaction behavior patterns of the pets, it can effectively distinguish between complex motion caused by normal interaction and data anomalies caused by environmental interference, thereby significantly reducing the false alarm rate and improving the accuracy and reliability of the pet lost connection warning. This makes the pet GPS positioning and tracking fence alarm method more robust and adaptable in practical applications, providing more accurate and less disturbed pet safety protection for pet owners.

[0161] In some embodiments, in step S403, identifying the interaction behavior pattern of the pet according to the relative parameters can include but is not limited to the following steps:

[0162] In step S501, first vertical direction acceleration data of the target pet tracking device in a preset time window is acquired.

[0163] In step S502, second vertical direction acceleration data of the surrounding pet tracking device in the preset time window is acquired.

[0164] In step S503, a target motion pattern of the pet in the vertical direction is identified according to the first vertical direction acceleration data and the second vertical direction acceleration data, and the target motion pattern includes a synchronous motion pattern or an asynchronous motion pattern.

[0165] In step S504, an interaction behavior pattern is identified according to the relative parameters and the target motion pattern.

[0166] In some embodiments, since only relying on relative distance, relative speed and direction relationship may be difficult to accurately capture the complex interaction behavior between pets, especially involving the vertical direction, resulting in limited accuracy and robustness of interaction pattern recognition. For example, when two pets are playing or chasing each other, their motion state in the vertical direction may have complex changes in synchronization or asynchronization, and these changes are crucial for accurately determining their interaction behavior pattern. Therefore, first vertical direction acceleration data of the target pet tracking device in a preset time window can be acquired, and second vertical direction acceleration data of the surrounding pet tracking device in the preset time window can be acquired. It can be understood that the vertical direction acceleration data refers to the acceleration value sequence collected by the inertial sensor worn by the target pet tracking device in the vertical direction. These data reflect the instantaneous motion state of the pet in the vertical direction, such as jumping, standing, crouching or falling, etc. The preset time window refers to a period of time for data collection and analysis, and its length can be set according to the actual application scenario and the motion characteristics of the pet, for example, it can be set to several seconds to tens of seconds to ensure that typical interaction behaviors can be captured.

[0167] Then, according to the first vertical direction acceleration data and the second vertical direction acceleration data, a target motion mode of the pets in the vertical direction is identified, aiming to analyze whether the motions of the two pets in the vertical direction are consistent (synchronous motion mode) or independent of each other (asynchronous motion mode). The target motion mode includes the synchronous motion mode or the asynchronous motion mode. The synchronous motion mode refers to that the motion trend or change height of the two pets in the vertical direction has a high correlation, for example, they jump up at the same time or squat down at the same time. The asynchronous motion mode refers to that the motions of the two pets in the vertical direction are relatively independent, for example, one pet jumps up while the other pet remains stationary, or the vertical motions of the two pets are inconsistent in rhythm. Then, according to the relative parameter and the target motion mode, the interaction behavior mode is identified.

[0168] Through the above technical solution, the embodiment can more accurately and more specifically identify the interaction behavior mode between the pets. By introducing the vertical direction acceleration data and identifying the target motion mode in the vertical direction, complex interactions that are difficult to identify only by the horizontal relative parameter can be effectively distinguished, such as synchronous jumping, asynchronous chasing or mutual playing, etc. This enables the system to have a deeper understanding of the behavior state of the pets, so as to provide more accurate input for subsequent threshold adjustment amount determination, and further improve the accuracy and timeliness of the pet loss warning, effectively avoid false positives or false negatives, and enhance the reliability of pet safety protection.

[0169] In some embodiments, in step S503, according to the first vertical direction acceleration data and the second vertical direction acceleration data, the target motion mode of the pets in the vertical direction is identified, which can include but is not limited to the following steps:

[0170] Time series alignment is performed on the first vertical direction acceleration data and the second vertical direction acceleration data.

[0171] The vertical direction motion correlation between the first vertical direction acceleration data and the second vertical direction acceleration data after time series alignment is calculated.

[0172] If the vertical direction motion correlation is greater than a preset correlation threshold, it is determined that the target motion mode is the synchronous motion mode.

[0173] If the vertical direction motion correlation is less than the preset correlation threshold, it is determined that the target motion mode is the asynchronous motion mode.

[0174] In some embodiments, the first and second vertical direction acceleration data can be time series aligned first to eliminate minor time deviations caused by device startup time, data transmission delay, or sampling frequency difference, etc., to ensure the comparability of data points of the two pet tracking devices on the time axis. For example, dynamic time warping (DTW) algorithm or cross-correlation based method can be used for alignment to find the best time correspondence.

[0175] Then the vertical direction motion correlation between the time series aligned first and second vertical direction acceleration data is calculated. For example, statistical analysis can be performed to quantify the similarity of the motion patterns of the first and second vertical direction acceleration data in the vertical direction. The vertical direction motion correlation can be calculated using Pearson correlation coefficient, Spearman rank correlation coefficient, or cross-correlation function, etc. Higher correlation value indicates that the motion trends of the two pets in the vertical direction are highly consistent, such as jumping, climbing or squatting at the same time; while lower correlation value indicates that their vertical motions are relatively independent or inconsistent.

[0176] If the vertical direction motion correlation is greater than a preset correlation threshold, the target motion pattern is determined as a synchronous motion pattern; if the vertical direction motion correlation is less than the preset correlation threshold, the target motion pattern is determined as an asynchronous motion pattern. The preset correlation threshold is a configurable parameter for distinguishing between synchronous and asynchronous motion patterns. The threshold can be empirically set according to the actual application scenario, pet species, motion characteristics, and experimental data, or optimized through machine learning method. When the calculated vertical direction motion correlation is greater than the preset correlation threshold, it indicates that the motion of the two pets in the vertical direction has significant synchronicity, and thus is determined as a synchronous motion pattern. Conversely, when the vertical direction motion correlation is less than or equal to the preset correlation threshold, it is considered that their vertical motions lack significant synchronicity, and thus is determined as an asynchronous motion pattern.

[0177] Through the above technical solutions, the present embodiment can accurately distinguish whether the pets are moving synchronously or asynchronously by introducing vertical direction acceleration data and performing time series alignment and correlation analysis. This significantly improves the accuracy and robustness of recognizing pet interaction behavior patterns, for example, more accurately determining whether the pets are playing together (synchronous jumping), chasing each other (asynchronous motion), or just independently active in the same area. This refined motion pattern recognition capability further enhances the intelligent level of the pet lost connection warning system, making the warning judgment more accurate, effectively avoiding false positives or false negatives, and thus better protecting the safety of pets.

[0178] The beneficial effects of implementing the embodiments of the present application include that the embodiments of the present application first receive the first positioning data sent by the target pet tracking device, identify target change information caused by environmental interference according to the first positioning data, then identify a signal interference area according to the frequency and duration of the target change information, identify a signal loss area according to the electronic fence area and the signal quality information sent by the plurality of target pet tracking devices, if the pet is in a poor signal area, obtain the moving direction of the pet, if the moving direction points to the electronic fence boundary or the poor signal area, generate a pet disconnection warning, so that the pet GPS positioning tracking and alarm triggering can be realized in combination with the signal interference area and the signal loss area, and the positioning tracking accuracy and the alarm timeliness are improved.

[0179] As shown in Figure 2 The embodiments of the present application also provide a pet GPS positioning tracking fence alarm system, which comprises:

[0180] The data receiving module 601 is used for receiving first positioning data sent by a target pet tracking device.

[0181] The change information identifying module 602 is used for identifying target change information caused by environmental interference according to the first positioning data, and the target change information comprises position jump information and direction mutation information.

[0182] The signal interference identifying module 603 is used for identifying a signal interference area according to the frequency and duration of the target change information.

[0183] The signal loss identifying module 604 is used for identifying a signal loss area according to the electronic fence area and the signal quality information sent by the plurality of target pet tracking devices, and the signal quality information comprises the number of connectable satellites, signal strength or positioning data disturbance value.

[0184] The direction obtaining module 605 is used for obtaining the moving direction of the pet if the pet is in a poor signal area, and the moving direction is collected by a geomagnetic sensor built in the target pet tracking device, and the poor signal area comprises the signal interference area or the signal loss area.

[0185] The warning generating module 606 is used for generating a pet disconnection warning if the moving direction points to the electronic fence boundary or the poor signal area.

[0186] The contents in the method embodiments are all applicable to the system embodiments, the system embodiments specifically implement the same functions as the method embodiments, and achieve the same beneficial effects as the method embodiments.

[0187] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

Claims

1. A pet GPS positioning tracking fence alarm method, characterized in that, The method comprises the following steps: receiving first positioning data sent by a target pet tracking device; identifying target change information caused by environmental interference according to the first positioning data, the target change information comprising position jump information and direction mutation information; identifying a signal interference area according to the frequency and duration of the target change information; identifying a signal loss area according to the electronic fence area and signal quality information sent by a plurality of target pet tracking devices, the signal quality information comprising the number of connectable satellites, signal strength or positioning data disturbance value; if the pet is in a poor signal area, obtaining the moving direction of the pet, the moving direction being collected by a geomagnetic sensor built in the target pet tracking device, the poor signal area comprising the signal interference area or the signal loss area; if the moving direction points to the electronic fence boundary or the poor signal area, generating a pet disconnection warning.

2. The method of claim 1, wherein, The method of identifying a signal loss area according to the electronic fence area and signal quality information sent by a plurality of target pet tracking devices comprises: dividing the electronic fence area to obtain a plurality of grid areas; performing signal judgment on each grid area within a preset detection duration; if each signal quality information in the grid area is signal quality degradation, determining the grid area as the signal loss area.

3. The method of claim 1, wherein, The method of identifying target change information caused by environmental interference according to the first positioning data comprises: obtaining instantaneous acceleration data of the pet, the instantaneous acceleration data being collected by an inertial sensor built in the target pet tracking device; calculating short-time theoretical displacement vector according to the instantaneous acceleration data; calculating real-time displacement vector according to the first positioning data; comparing the short-time theoretical displacement vector and the real-time displacement vector to obtain consistency deviation, the consistency deviation comprising direction deviation or size deviation; determining a consistency deviation threshold; if the consistency deviation is greater than the consistency deviation threshold, taking the position change data in the real-time displacement vector as the position jump information, and taking the direction change data of the short-time theoretical displacement vector and the real-time displacement vector as the direction mutation information.

4. The method of claim 3, wherein, The method of calculating short-time theoretical displacement vector according to the instantaneous acceleration data comprises: converting the instantaneous acceleration data from device coordinate system to geographic coordinate system according to gyroscope data; integrating the converted instantaneous acceleration data according to a preset time step to obtain instantaneous speed change; integrating the instantaneous speed change to obtain instantaneous displacement change; generating the short-time theoretical displacement vector according to the instantaneous displacement change.

5. The method of claim 3, wherein, The method of determining a consistency deviation threshold comprises: obtaining acceleration change rate of the pet; calculating motion state change smoothness according to the acceleration change rate; determining threshold adjustment amount according to the motion state change smoothness; determining the consistency deviation threshold according to the threshold adjustment amount and environmental noise.

6. The method of claim 5, wherein, The method of determining threshold adjustment amount according to the motion state change smoothness comprises: obtaining a first vertical direction acceleration change feature of the target pet tracking device within a preset time window; obtaining a horizontal direction acceleration change feature of the target pet tracking device within a preset time window; identifying a complex motion mode of the pet according to the first vertical direction acceleration change feature and the horizontal direction acceleration change feature; determining the threshold adjustment amount according to the complex motion mode and the motion state change smoothness.

7. The method of claim 5, wherein, The determination of the threshold adjustment amount according to the motion state change smoothness comprises: obtaining first inertial sensing data of the target pet tracking device, and second positioning data and second inertial sensing data of a surrounding pet tracking device; calculating relative parameters between the target pet tracking device and the surrounding pet tracking device according to the first positioning data, the first inertial sensing data, the second positioning data and the second inertial sensing data, the relative parameters comprising relative distance, relative speed and direction relationship; identifying an interactive behavior mode of the pet according to the relative parameters; determining the threshold adjustment amount according to the interactive behavior mode and the motion state change smoothness.

8. The method of claim 7, wherein, The identification of the interactive behavior mode of the pet according to the relative parameters comprises: obtaining first vertical direction acceleration data of the target pet tracking device within a preset time window; obtaining second vertical direction acceleration data of the surrounding pet tracking device within a preset time window; identifying a target motion mode of the pet in the vertical direction according to the first vertical direction acceleration data and the second vertical direction acceleration data, the target motion mode comprising a synchronous motion mode or an asynchronous motion mode; identifying the interactive behavior mode according to the relative parameters and the target motion mode.

9. The method of claim 8, wherein, The identification of the target motion mode of the pet in the vertical direction according to the first vertical direction acceleration data and the second vertical direction acceleration data comprises: time series alignment of the first vertical direction acceleration data and the second vertical direction acceleration data; calculating vertical direction motion correlation between the first vertical direction acceleration data and the second vertical direction acceleration data after time series alignment; if the vertical direction motion correlation is greater than a preset correlation threshold, determining that the target motion mode is a synchronous motion mode; if the vertical direction motion correlation is less than a preset correlation threshold, determining that the target motion mode is an asynchronous motion mode.

10. A pet GPS positioning tracking fence alarm system, characterized in that, comprises: a data receiving module configured to receive first positioning data sent by a target pet tracking device; a change information identifying module configured to identify target change information caused by environmental interference according to the first positioning data, the target change information comprising position jump information and direction mutation information; a signal interference identifying module configured to identify a signal interference area according to a frequency and a duration of the target change information; a signal loss identifying module configured to identify a signal loss area according to an electronic fence area and signal quality information sent by a plurality of target pet tracking devices, the signal quality information comprising a number of connectable satellites, signal strength or positioning data disturbance value; An orientation acquisition module is configured to acquire a moving orientation of the pet if the pet is in a poor signal area, wherein the moving orientation is collected by a geomagnetic sensor built in the target pet tracking device, and the poor signal area includes the signal interference area or the signal loss area; An early warning generation module is configured to generate a pet disconnection early warning if the moving orientation points to the boundary of the electronic fence or the poor signal area.

Images