Animal behavior management system based on intelligent dog training device

Abstract

The invention relates to the technical field of information retrieval, in particular to an animal behavior management system based on an intelligent dog training device, and the animal behavior management system based on the intelligent dog training device comprises a behavior data collection module, a high-dimensional offset calculation module, a behavior boundary recognition module, an abnormal influence degree analysis module and a behavior index updating module. According to the method, the comprehensiveness and accuracy of behavior monitoring are improved by integrating environmental factors and multi-dimensional behavior data, abnormal behaviors are accurately identified and predicted by utilizing time sequence analysis and comprehensive processing of motion data, the accuracy of behavior pattern identification is enhanced, and the accuracy of behavior pattern identification is improved by analyzing the change of motion rhythms and the transfer of behavior paths. The abnormal behaviors are stably recognized, the coping complexity of the system is improved, the behavior index is dynamically updated, the matching degree is optimized, the system can adapt to behavior evolution, and the reasonability and effect of an intervention strategy are improved.

Description

Technical Field

[0001] The invention relates to the technical field of information retrieval, and in particular to an animal behavior management system based on an intelligent dog training device. Background Art

[0002] The field of information retrieval technology includes data storage, organization, query and retrieval methods, and is widely used in application scenarios such as database management, search engines, and intelligent recommendation systems. The core content of this technical field involves building data index structures, designing query optimization strategies, and improving data retrieval efficiency. It usually uses inverted indexes, Boolean retrieval, vector space models, etc. to improve data matching and retrieval speed. In addition, this field also covers technologies such as natural language processing, semantic analysis, and machine learning to improve the relevance and intelligence of information, making it easier for users to quickly obtain the required information from large amounts of data.

[0003] Among them, the animal behavior management system based on the intelligent dog trainer refers to the use of information retrieval technology to collect, analyze, store and query animal behavior data to achieve animal behavior management. The system collects animal activity data through animal behavior sensor equipment, analyzes behavior patterns based on feature extraction methods, and stores the data in a structured database. Subsequently, a behavior information index is constructed through data indexing technology to achieve retrieval queries based on keywords, time series, and behavior patterns, and support multi-dimensional data analysis. The system also uses classification methods to classify behavior data in order to quickly retrieve specific behavior records, combine pattern matching methods to identify abnormal behaviors, and implement automatic behavior intervention strategies based on rule settings.

[0004] Existing technologies are limited to single behavioral features in data collection and fail to fully consider environmental factors, resulting in behavioral identification being susceptible to interference in complex environments, reducing the accuracy of behavioral data. In terms of abnormal behavior analysis, they mainly rely on fixed pattern matching and lack analysis of behavioral transfer trends, making it difficult to effectively distinguish short-term anomalies from long-term anomalies, resulting in unstable abnormal behavior judgments. In terms of behavioral data storage and indexing, they only rely on static retrieval methods, fail to establish dynamic behavioral associations, and find it difficult to track changes in behavioral patterns, resulting in insufficient adaptability in long-term behavioral management. In terms of abnormal behavior impact analysis, the evaluation method is too static and lacks quantitative analysis of the scope of abnormal behavior impact, resulting in a lack of specificity in abnormal intervention strategies, which may lead to excessive or insufficient intervention and reduce the effectiveness of intervention measures. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an animal behavior management system based on an intelligent dog training device.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: an animal behavior management system based on an intelligent dog training device comprises: The behavior data collection module obtains dog behavior data, calls the dog training device sensor data, records the movement trajectory and environmental factors, and screens the environmental behavior characteristic data.

[0007] The high-dimensional offset calculation module extracts time series features based on environmental behavior feature data, calculates trajectory position difference and offset rate, records the current behavior path, determines the offset change rate, and combines motion acceleration and direction changes to screen the behavior transition rate to generate offset dynamic prediction results.

[0008] The behavior boundary recognition module calls the offset dynamic prediction results, records the continuity disruption of abnormal behaviors, divides time periods, calculates movement frequency, sprint proportion, and stationary time, analyzes changes in movement rhythm, identifies behavior path deviations, and generates behavior fuzzy intervals.

[0009] The abnormal impact analysis module evaluates the impact of abnormal behavior on the normal pattern based on the behavior fuzzy interval, calculates the decrease in the probability of behavior transfer, behavior delay and break rate, and generates the impact ratio of abnormal behavior.

[0010] The behavior index update module calls the abnormal behavior impact ratio, analyzes the behavior interval changes, calculates the new behavior matching degree, compares the probability of behavior recurrence, filters the index update data, and generates the intervention strategy after the behavior update.

[0011] As a further solution of the present invention, the environmental behavior characteristic data includes motion characteristic data, trajectory data, and environmental characteristic data; the offset dynamic prediction result includes trajectory change characteristics, offset rate characteristics, and abnormal behavior offset values; the behavior fuzzy interval includes behavior continuity characteristics, time partition characteristics, and behavior path change characteristics; the abnormal behavior impact proportion includes behavior transfer characteristics, abnormal behavior impact characteristics, and behavior fracture characteristics; the behavior post-update intervention strategy includes behavior interval characteristics, new behavior matching characteristics, and behavior index update characteristics.

[0012] As a further solution of the present invention, the behavior data collection module includes: The motion data extraction submodule calls the built-in sensor data of the intelligent dog trainer to obtain the dog's motion acceleration, behavior sequence, activity frequency, calculate the behavior duration, count the rhythm change rate and the stability of the adjacent behavior sequence, analyze the dog's motion trajectory points, record the number of sprints and the proportion of static time, and generate motion feature data; The behavior pattern analysis submodule identifies the dog's continuous behavior pattern based on the motion feature data, analyzes the behavior conversion trend, extracts the probability of behavior occurrence, and calculates the correlation between the dog's differentiated behaviors using the formula: ; The stability coefficient of the dog's behavior sequence is calculated and combined with the behavior conversion trend to obtain the behavior pattern characteristics; in, Represents behavioral pattern characteristics, Represents the duration of a single behavior, represents the behavior transition interval, is the total number of behaviors, is the time attenuation coefficient, represents the total duration of the behavior sequence, represents the sum of the products of the duration of all behaviors and the interval between behavioral transitions; The environmental characteristic screening submodule calculates the impact of environmental characteristic changes on the behavioral pattern based on the behavioral pattern characteristics and combines the noise level, light intensity, and ground vibration conditions of the space where the dog is located, screens highly correlated environmental variables, and obtains environmental behavioral characteristic data.

[0013] As a further solution of the present invention, the high-dimensional offset calculation module includes: The trajectory position calculation submodule extracts the time series behavior characteristics based on the environmental behavior characteristic data, obtains the motion trajectory coordinates of adjacent time points, calculates the trajectory position change value within each time step, sets the time interval, calculates the trajectory vector change amount, and obtains the trajectory position difference value by combining the trajectory data; The deviation rate analysis submodule calculates the deviation rate between adjacent behavior feature vectors based on the trajectory position difference, using the formula: ; Calculate and obtain the behavior deviation rate, compare the behavior trajectory data with the current behavior path data, and obtain the behavior deviation change rate; in, represents the behavior deviation rate, Representative The trajectory coordinate value at each time point, Representative The target trajectory coordinate value at each time point, is the total number of trajectory points within the time step, is the trajectory deviation adjustment factor, Represents the deviation trend of the behavior trajectory, is the behavior vector change rate correction; The abnormal behavior screening submodule calculates the abnormal behavior offset value based on the change rate of the behavior offset, and screens the data with a larger behavior conversion rate in combination with the dog's movement acceleration and direction change to obtain the offset dynamic prediction result.

[0014] As a further solution of the present invention, the behavior boundary recognition module includes: The behavior continuity calculation submodule extracts the behavior sequence before and after the abnormal behavior based on the offset dynamic prediction result, calculates the time interval between continuous behaviors, counts the behavior destruction rate, divides the behavior time period, and obtains the behavior continuity destruction value; The motion rhythm analysis submodule calculates the motion frequency, sprint proportion, and static duration for each time period based on the behavior continuity destruction value, and uses a multi-scale behavior distribution analysis method to construct a behavior change matrix within the time period, and analyzes the motion state stability of the time partition using the formula: ; Obtain the amplitude of the movement rhythm change by calculation, compare the calculation results of adjacent time periods, construct the behavior transfer trend vector, establish the time dynamic transfer matrix, and obtain the behavior transfer change coefficient; in, Represents the amplitude of the movement rhythm change, Representative The movement frequency value at each time point, Representative The duration of stillness at a time point, is the total number of behavior records within the time step, is the behavioral volatility weight, Representative The rate of change of behavior at a time point, represents the behavior change cycle parameter, is the behavior change smoothing factor, Represents the degree of fluctuation of the behavior range, is the motion adjustment factor, Represents the change parameter of sprint behavior proportion, A dispersion correction parameter representing the behavior rhythm within a time period; The behavior path comparison submodule compares the normal behavior path with the abnormal behavior path based on the behavior transfer variation coefficient, calculates the path fitting index, screens the similar path intervals, and obtains the behavior fuzzy intervals.

[0015] As a further solution of the present invention, the abnormal impact analysis module includes: The behavior sequence stability assessment submodule extracts the behavior sequence before and after the abnormal behavior based on the behavior fuzzy interval, compares the behavior stability before and after the abnormality, calculates the distribution probability of the behavior category within the time period, calculates the behavior frequency fluctuation, determines the disturbance degree of the behavior sequence, calculates the behavior stability reduction ratio according to the disturbance measurement standard, and obtains the behavior sequence stability offset value; The behavior transfer impact calculation submodule calculates the decrease in the behavior transfer probability within the abnormal behavior period based on the behavior sequence stability offset value, calculates the conversion rate from normal behavior to abnormal behavior, constructs a time segmented behavior impact model, and calculates the dynamic impact of abnormal behavior on the normal behavior sequence, using the formula: ; The behavior transfer breakage rate is obtained by calculation, the continuous interference effect of abnormal behavior on normal behavior in the behavior pattern is analyzed, and the overall change range of the behavior pattern is calculated.

[0016] in, represents the behavioral transfer break rate, Representative The normal behavior transition probability at a time point, represents the transition probability when abnormal behavior occurs, is the total number of behavior records within the time step, represents the duration of normal behavior, represents the duration of abnormal behavior; The abnormal behavior impact assessment submodule calculates the impact ratio of abnormal behavior on normal behavior patterns based on the behavior transfer fracture rate, extracts the impact of abnormal behavior on normal behavior, analyzes the expansion range of abnormal behavior on the time axis, analyzes the impact degree of abnormal behavior on behavior stability, establishes a measurement index of the impact of abnormality on behavior patterns, and obtains the impact ratio of abnormal behavior.

[0017] As a further solution of the present invention, the behavior index updating module includes: The behavior interval analysis submodule calls the abnormal behavior impact ratio, obtains the duration sequence of the dog's behavior, calculates the behavior interval time based on the behavior data, calculates the behavior interval change rate, analyzes the correlation between the abnormal behavior recurrence probability and the normal behavior interval, extracts the behavior interval change trend, and uses the formula: ; Calculate the behavior interval change ratio , obtain the behavior interval change trend; in, represents the behavior interval change ratio, For the The time between behaviors, is the interval time of the previous behavior, is the total number of times the behavior occurred; The behavior matching evaluation submodule constructs a behavior continuity weight matrix based on the behavior interval change trend, calculates the matching degree of abnormal behavior in the overall behavior pattern, evaluates the matching degree, compares the probability of recurrence of normal behavior, screens candidate behavior data that meets the behavior pattern, and obtains the behavior matching evaluation result; The behavior index screening submodule calls the behavior matching evaluation result, screens the data nodes of the behavior index, calculates the behavior impact range after the index update, compares the behavior change proportion before and after the abnormal behavior update, screens the behavior index update plan, and obtains the intervention strategy after the behavior update.

[0018] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by integrating environmental factors and multi-dimensional behavioral data, the comprehensiveness and accuracy of behavior monitoring are improved, and time series analysis and comprehensive processing of motion data are used to accurately identify and predict abnormal behaviors, thereby enhancing the accuracy of behavior pattern recognition. By analyzing changes in movement rhythm and transfer of behavioral paths, abnormal behaviors can be stably identified, and the system's response complexity is improved. The behavior index is dynamically updated and the matching degree is optimized, so that the system can adapt to the evolution of behavior, thereby improving the rationality and effectiveness of intervention strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a system flow chart of the present invention; Figure 2 It is a flow chart of the behavior data collection module of the present invention; Figure 3 It is a flow chart of the high-dimensional offset calculation module of the present invention; Figure 4 It is a flow chart of the behavior boundary identification module of the present invention; Figure 5 This is a flow chart of the abnormal impact analysis module of the present invention; Figure 6 This is a flow chart of the behavior index update module of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0021] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0022] See also Figure 1 , the animal behavior management system based on the intelligent dog training device includes: The behavior data collection module obtains the behavior data stream of the dog wearing the smart dog training device, collects the movement acceleration, behavior sequence, activity frequency, behavior duration, rhythm change rate, and stability of the sequence of adjacent behaviors, calls the built-in sensor data of the smart dog training device, records the dog's movement trajectory points, sprint times, and static time percentage, and combines the noise level, light intensity, and ground vibration conditions of the space where the dog is located to screen the environmental behavior characteristic data; The high-dimensional offset calculation module extracts time series behavior characteristics based on environmental behavior feature data, calculates the trajectory position difference of adjacent time points, analyzes the offset rate between adjacent behavior feature vectors, and determines the behavior offset change rate by comparing the behavior trajectory data with the current behavior path data. According to the behavior offset rate, the abnormal behavior offset value is calculated, and the behavior change rate is screened in combination with the dog's movement acceleration and direction changes to generate the offset dynamic prediction results. The behavior boundary recognition module calls the offset dynamic prediction results, records the behavior continuity disruption before and after the abnormal behavior, divides the time period and calculates the movement frequency, sprint ratio, and static duration of each period. According to the behavior change trend of time partition, the movement rhythm change amplitude of adjacent time periods is analyzed, the change coefficient of the behavior transfer path is identified, and the difference between the normal behavior path and the abnormal behavior path is compared to generate the behavior fuzzy interval. The abnormal impact analysis module calls the behavior fuzzy interval, analyzes the stability of the behavior sequence before and after the abnormal behavior according to the transfer path of the dog's behavior pattern, calculates the decrease in the probability of behavior transfer, records the delay of the behavior, evaluates the impact of abnormal behavior on the regular behavior pattern, calculates the behavior transfer fracture rate according to the impact ratio, and generates the abnormal behavior impact ratio; The behavior index update module calls the abnormal behavior impact ratio, calculates the changing pattern of the behavior interval time based on the duration of the dog's behavior and the probability of abnormal behavior recurrence, analyzes the matching degree of the new behavior according to the behavior continuity weight matrix, compares the probability of repeated occurrence of the dog's behavior, screens the behavior index update data nodes, and generates the intervention strategy after the behavior update.

[0023] Environmental behavior feature data include motion feature data, trajectory data, and environmental feature data. The dynamic prediction results of offset include trajectory change characteristics, offset rate characteristics, and abnormal behavior offset values. The behavior fuzzy interval includes behavior continuity characteristics, time partition characteristics, and behavior path change characteristics. The abnormal behavior impact proportion includes behavior transfer characteristics, abnormal behavior impact characteristics, and behavior fracture characteristics. The intervention strategy after behavior update includes behavior interval characteristics, new behavior matching characteristics, and behavior index update characteristics.

[0024] See also Figure 2 , the behavior data collection module includes: The motion data extraction submodule calls the built-in sensor data of the intelligent dog trainer to obtain the dog's motion acceleration, behavior sequence, activity frequency, calculate the behavior duration, count the rhythm change rate and the stability of the adjacent behavior sequence, analyze the dog's motion trajectory points, record the number of sprints and the proportion of static time, and generate motion feature data; First, it is necessary to obtain real-time data from the accelerometer, gyroscope and geomagnetic sensor. These data can record the dog's spatial position, movement direction and acceleration changes at different time points. For example, assuming that a dog moves at an acceleration of 0.5m / s2, its movement trajectory can be obtained by combining time axis analysis. Next, the extraction of behavioral sequences depends on the data division of continuous time windows. When the time window is set to 1s, the motion feature vector per second can be composed of three-axis acceleration values ​​(Ax, Ay, Az) and angular velocity values ​​(Gx, Gy, Gz). The data filtering method is then used to remove outliers. For example, data points with noise levels exceeding 2 times the standard deviation are considered abnormal and removed. The activity frequency is calculated by Fourier transform analysis of the dog's gait cycle. Taking a 10s data window as an example, it is calculated that the main step frequency is distributed in the range of 1.5Hz to 3.5Hz, so the dog's current activity frequency can be defined as 2.5Hz. Subsequently, the behavior duration is obtained by calculating the time span of each independent behavior. For example, if the dog runs continuously for 5 seconds, the duration of the behavior is The time is 5s, and the rhythm change rate is calculated by the ratio of speed change to time. If the dog's speed increases from 0.5m / s to 1.5m / s within 2 seconds, the rhythm change rate is (1.5-0.5) / 2=0.5m / s2. The stability of the sequence of adjacent behaviors can be calculated by statistically analyzing the probability matrix of each behavior conversion. For example, if the dog changes from walking to running in 50% of cases, the conversion probability is 0.5. Based on these parameters, the dog's movement trajectory points can be analyzed, and the geographical coordinates can be collected through the GPS module, and combined with the timestamp to obtain To a continuous trajectory, for example, the dog moves along (X1, Y1) → (X2, Y2) → (X3, Y3) within 10 seconds, and finally forms a motion trajectory. The recording of the number of sprints depends on the speed peak detection. For example, if 3 peaks with a speed exceeding 2m / s are detected within 10 seconds, the number of sprints is 3 times. The proportion of static time is calculated by counting the proportion of time when the speed is close to 0m / s. For example, the dog is static for 10 seconds within 30 seconds, and the static time accounts for 10 / 30=33.3%, and finally the motion feature data is obtained.

[0025] The behavior pattern analysis submodule identifies the dog's continuous behavior pattern based on motion feature data, analyzes the behavior conversion trend, extracts the probability of behavior occurrence, and calculates the correlation between the dog's differentiated behaviors using the formula: ; The stability coefficient of the dog's behavior sequence is calculated and combined with the behavior conversion trend to obtain the behavior pattern characteristics; in, Represents behavioral pattern characteristics, Represents the duration of a single behavior, represents the behavior transition interval, is the total number of behaviors, is the time attenuation coefficient, represents the total duration of the behavior sequence, represents the sum of the products of the duration of all behaviors and the interval between behavioral transitions; First, the motion data is classified by setting behavior labels (such as stillness, walking, running, and jumping). For example, if the acceleration is less than 0.2m / s2 and the speed is less than 0.3m / s, it is judged as a still state. If the acceleration is between 0.3m / s2-1.0m / s2 and the speed is between 0.3m / s-1.5m / s, it is judged as a walking state. The conversion probabilities between different behaviors are statistically calculated by constructing a Markov state transition matrix. For example, the probability of a dog changing from walking to running is 0.6, and the probability of a dog changing from running to stillness is 0.3. The behavior conversion trend can be analyzed using the behavior sequence. Column data calculation, for example, in 100 behavior samples, the proportion of dogs switching from walking to running is 60%, then the conversion probability is 0.6. In addition, the probability of behavior occurrence can be calculated through global behavior statistics. For example, in 500 seconds of data, the dog is only in a walking state for 200 seconds, then the probability of walking behavior is 200 / 500=0.4. The degree of correlation between different behaviors can be calculated using the Pearson correlation coefficient. For example, in 30 groups of data samples, the correlation coefficient between walking and running behavior time is 0.75, indicating that the two are strongly correlated. The formula is: ; in, is the duration of the behavior. If a dog completes a behavior within 5 seconds, , is the behavior transition interval. For example, if a behavior occurs 2s after another behavior, then , is the time attenuation coefficient. If , is the total duration of the behavior sequence. Assuming that the duration of this behavior sequence is 100s, then , substituting these data into the calculation, we get: ; Finally, the behavioral pattern characteristics were obtained, and the results showed that the current dog behavior pattern had low stability and high variability during the behavior conversion process.

[0026] The environmental feature screening submodule calculates the impact of environmental feature changes on behavioral patterns based on behavioral pattern characteristics, combined with the noise level, light intensity, and ground vibration conditions of the dog's space, screens highly correlated environmental variables, and obtains environmental behavioral feature data; First, the noise level is detected by the sound sensor, and the measurement unit is dB. For example, the environmental noise recorded in different time periods is 50dB, 65dB and 80dB respectively, of which 65dB is the background noise. When this value is exceeded, it is judged as a high-noise environment. The light intensity is measured by the light sensor, with lx (lux) as the unit. For example, the measured values ​​in the early morning, noon and evening are 200lx, 15000lx and 500lx respectively, and noon is a high-light environment. The ground vibration is collected by the acceleration sensor. For example, the recorded vibration acceleration values ​​are 0.01m / s2, 0.05m / s2, 0.2m / s2. If the vibration acceleration is higher than 0.1m / s2, it is considered that there is a large vibration on the ground. After obtaining these data, the correlation between each environmental factor and the behavioral pattern characteristics is calculated. For example, the correlation coefficient between noise level and behavioral stability is -0.6, indicating that a high noise environment will reduce the stability of dog behavior. The correlation coefficient of light intensity is 0.4, indicating that a high light environment may increase the activity frequency of dogs. The correlation coefficient between ground vibration and behavioral conversion is -0.7, indicating that larger ground vibrations will lead to unstable behavioral patterns. By screening highly correlated environmental variables, environmental behavioral characteristic data are obtained.

[0027] See also Figure 3 , the high-dimensional offset calculation module includes: The trajectory position calculation submodule extracts the time series behavior characteristics based on the environmental behavior characteristic data, obtains the motion trajectory coordinates of adjacent time points, calculates the trajectory position change value within each time step, sets the time interval, calculates the trajectory vector change, and combines the trajectory data to obtain the trajectory position difference; When extracting time series behavior features, we first need to obtain the dog's movement trajectory data at different time points, including three-dimensional coordinate information ( ), and perform time series processing, for example, at a certain moment The collected coordinate data is (1.2m, 3.5m, 0.8m). The data collected at the time is (1.4m, 3.8m, 0.9m), then the trajectory position change between the two time points can be calculated as , , , which is used to calculate the trajectory offset at the set time interval When the velocity is 1s, the velocity of the trajectory deviation vector can be expressed as ,Combining the multi-time point data, a complete trajectory sequence can be formed. By counting the ,position differences of the dogs' ,trajectory positions at different time steps ,can be obtained.

[0028] The deviation rate analysis submodule calculates the deviation rate between adjacent behavior feature vectors based on the trajectory position difference, using the formula: ; Calculate and obtain the behavior deviation rate, compare the behavior trajectory data with the current behavior path data, and obtain the behavior deviation change rate; in, represents the behavior deviation rate, Representative The trajectory coordinate value at each time point, Representative The target trajectory coordinate value at each time point, is the total number of trajectory points within the time step, is the trajectory deviation adjustment factor, Represents the deviation trend of the behavior trajectory, is the behavior vector change rate correction; When analyzing the offset rate of adjacent behavior feature vectors, it is necessary to extract the feature vectors of adjacent time points in the behavior sequence, for example, at time point The collected motion feature vector is ( , , ), at time point The collected vector is ( , , ), then the offset of adjacent eigenvectors can be calculated as , , using the formula: ; Among them, , , , , , , , calculated: ; ; ; The behavior deviation rate is obtained by the calculation, and the behavior trajectory data is compared with the current behavior path data according to the rate to obtain the behavior deviation change rate.

[0029] The abnormal behavior screening submodule calculates the abnormal behavior offset value based on the rate of change of the behavior offset, and screens the data with a large behavior conversion rate in combination with the dog's movement acceleration and direction changes to obtain the offset dynamic prediction result.

[0030] When calculating the abnormal behavior deviation value, you first need to set the deviation threshold To determine abnormal behavior, if the threshold is set to , when the calculated If the value is exceeded, it is considered abnormal behavior. The behavior conversion rate is further calculated by combining the dog's movement acceleration and direction change. For example, the dog's current acceleration is , the acceleration at the previous moment is , then the behavior conversion rate can be calculated as , for behavior change rates exceeding The data is judged as high-drift behavior data, and finally the drift dynamic prediction result is obtained.

[0031] See also Figure 4 ,The behavior boundary recognition module includes: The behavior continuity calculation submodule extracts the behavior sequence before and after the abnormal behavior based on the offset dynamic prediction results, calculates the time interval between continuous behaviors, counts the behavior destruction rate, divides the behavior time periods, and obtains the behavior continuity destruction value; First, we need to set a time window to segment the behavior data stream in order to extract the behavior patterns before and after the abnormality. The time window length is set to 5 seconds. Within this interval, the dog's motion parameters are collected, including step frequency, trajectory, speed, direction change, etc. The motion state transition between adjacent time steps is calculated. For example, at time Record that the dog is walking. Enter running mode, If an emergency stop occurs, the calculation and The time interval between the two behaviors was calculated, and the acceleration change trend of the period was recorded. The stability of the behavior conversion was further analyzed. The behavior switching frequency was obtained by counting the number of behavior switching in unit time, and the behavior stability deviation was calculated. For example, within 30 seconds, the dog's gait switching times reached 10 times, and the duration of adjacent behaviors was uneven, indicating that the behavior continuity in this interval was impaired. In addition, the behavior classification label was used to determine whether there was a violent jump in the previous and subsequent behaviors. For example, the dog went directly from a static state to a jumping state, and then to a sprint state. If the continuous behavior category changed greatly, the degree of behavior continuity destruction was high. The behavior change trend was further counted, and the abnormal behavior density in the time period was calculated. If the density exceeded the set threshold, the time period was marked as an abnormal time window, and finally the behavior continuity destruction value was obtained.

[0032] The motion rhythm analysis submodule calculates the motion frequency, sprint proportion, and static duration for each time period based on the behavior continuity destruction value, and uses a multi-scale behavior distribution analysis method to construct a behavior change matrix within the time period, and analyzes the stability of the motion state in the time partition using the formula: ; Obtain the amplitude of the movement rhythm change by calculation, compare the calculation results of adjacent time periods, construct the behavior transfer trend vector, establish the time dynamic transfer matrix, and obtain the behavior transfer change coefficient; in, Represents the amplitude of the movement rhythm change, Representative The movement frequency value at each time point, Representative The duration of stillness at a time point, is the total number of behavior records within the time step, is the behavioral volatility weight, Representative The rate of change of behavior at a time point, represents the behavior change cycle parameter, is the behavior change smoothing factor, Represents the degree of fluctuation of the behavior range, is the motion adjustment factor, Represents the change parameter of sprint behavior proportion, A dispersion correction parameter representing the behavior rhythm within a time period; Calculate the movement frequency, sprint proportion, and static duration for each time period. Use the movement data splitting method to partition the time series. For example, in a 30-second monitoring time, divide it into 6 5-second intervals, and calculate the movement characteristics in each time period separately, setting the time step , calculate the motion frequency in each time step and still time , and then calculate the overall movement rhythm fluctuation degree, using the formula: ; Among them, assuming , , , , , , , , , calculated: ; ; The final operation obtains the amplitude of the movement rhythm change, compares the calculation results of adjacent time periods, calculates the change value of the behavior transfer trend, and establishes the time dynamic transfer matrix to obtain the behavior transfer change coefficient.

[0033] The behavior path comparison submodule compares the normal behavior path with the abnormal behavior path based on the behavior transfer variation coefficient, calculates the path fitting index, screens the similar path intervals, and obtains the behavior fuzzy interval.

[0034] To calculate the trajectory difference of adjacent paths, first extract the dog's trajectory data in different time periods to obtain the path coordinate point sequence. For example, within a 5-second time interval, the dog's movement trajectory points are , , By calculating the spatial displacement difference between adjacent points, the behavior path change trend is obtained, the path conversion matrix is ​​constructed, the normal behavior path is matched with the abnormal behavior path, and the dynamic offset between the paths is calculated. For example, under normal behavior conditions, the dog's path presents a stable curve within 10 seconds, while under abnormal conditions, the path curve has a sharp offset. The degree of path deviation is calculated, and based on the time series analysis method, it is judged whether the path deviation is continuous. If the offset of multiple consecutive time steps exceeds the set threshold, the time interval is determined to be an abnormal path interval, and the path similarity is further calculated. The matching error between the paths is calculated by the dynamic time warping method. If the error value exceeds the set threshold of the normal behavior path, the abnormal behavior path interval is screened out, and finally the behavior fuzzy interval is obtained.

[0035] See also Figure 5 , the abnormal impact analysis module includes: The behavior sequence stability assessment submodule extracts the behavior sequence before and after the abnormal behavior based on the behavior fuzzy interval, compares the behavior stability before and after the abnormality, calculates the distribution probability of the behavior category within the time period, calculates the behavior frequency fluctuation, determines the disturbance degree of the behavior sequence, calculates the behavior stability reduction ratio according to the disturbance measurement standard, and obtains the behavior sequence stability offset value; First, build a time window Divide the behavior data stream, store the data before and after the abnormal behavior separately, perform time series analysis, and set the time step The time window is 5 seconds. In each time window, the distribution of dog behavior categories is counted, including static, walking, running, jumping and other states, and the frequency of occurrence of each category in the time window is calculated. For example, the number of behavior category conversions in a 5-second window is 8 times, of which the static state accounts for 20%, the walking state accounts for 50%, and the running state accounts for 30%. The stability index of behavior switching is calculated. If the deviation of the behavior category distribution in the previous and next windows exceeds 20%, it is determined that the stability of the behavior sequence has decreased. In addition, the duration of abnormal behavior is calculated. and the average duration of normal behavior For comparison, if The delay of abnormal behavior is recorded, and the impact of abnormal behavior on the stability of subsequent behavior is counted, and finally the stability offset value of the behavior sequence is obtained.

[0036] The behavior transfer impact calculation submodule calculates the decrease in the probability of behavior transfer within the abnormal behavior period based on the behavior sequence stability offset value, calculates the conversion rate from normal behavior to abnormal behavior, constructs a time-segmented behavior impact model, and calculates the dynamic impact of abnormal behavior on the normal behavior sequence using the formula: ; The behavior transfer breakage rate is obtained by calculation, the continuous interference effect of abnormal behavior on normal behavior in the behavior pattern is analyzed, and the overall change range of the behavior pattern is calculated.

[0037] in, represents the behavioral transfer break rate, Representative The normal behavior transition probability at a time point, represents the transition probability when abnormal behavior occurs, is the total number of behavior records within the time step, represents the duration of normal behavior, represents the duration of abnormal behavior; Suppose we monitor a dog in a 10-second time window, of which the first 5 seconds are normal behavior and the last 5 seconds are abnormal behavior. The number of normal behavior sampling points is , the number of abnormal behavior sampling points is .

[0038] Calculate the change in behavior transition probability Set the transition probability of normal behavior at 5 time points to: ; The transition probability of abnormal behavior at 5 time points is set as: ; Calculate the change in transition probability at each time point: ; ; ; ; ; Calculate the average change in behavior transition probability: ; Calculating behavior duration changes Set the duration of normal behavior (Second): ; Set the duration of abnormal behavior (Second): ; Calculate the change in duration at each time point: ; ; ; ; ; Calculate the mean of the change in behavior duration: ; Calculation of behavioral transfer break rate ; Finally, the behavioral transfer break rate was calculated , this value indicates that the abnormal behavior has undergone significant changes in transition probability and duration compared to the normal behavior. If it exceeds the set threshold (such as 2.0), it indicates that the abnormal behavior has a significant disturbance effect on the normal behavior pattern.

[0039] The abnormal behavior impact assessment submodule calculates the impact ratio of abnormal behavior on normal behavior patterns based on the behavior transfer fracture rate, extracts the impact of abnormal behavior on normal behavior, analyzes the extension range of abnormal behavior on the time axis, analyzes the impact of abnormal behavior on behavior stability, establishes a measurement index for the impact of abnormal behavior on behavior patterns, and obtains the impact ratio of abnormal behavior; Based on the behavior transfer fracture rate, calculate the impact ratio of abnormal behavior on normal behavior pattern, extract the impact of abnormal behavior on normal behavior, first calculate the expansion range of abnormal behavior, set the time step And record the propagation trend of abnormal behavior in multiple time steps. If the duration of abnormal behavior exceeds twice the average duration of normal behavior, calculate the impact weight of abnormal behavior on subsequent behavior, evaluate the impact ratio of abnormal behavior in the overall behavior pattern, establish abnormal impact measurement indicators of behavior pattern, and compare the path similarity of normal behavior and abnormal behavior. If the abnormal behavior path deviation exceeds the set threshold, record the impact weight of abnormal behavior, and finally obtain the impact ratio of abnormal behavior.

[0040] See also Figure 6 , the behavior index update module includes: The behavior interval analysis submodule calls the abnormal behavior impact ratio, obtains the duration sequence of the dog's behavior, calculates the behavior interval time based on the behavior data, calculates the behavior interval change rate, analyzes the correlation between the probability of abnormal behavior recurrence and the normal behavior interval, and extracts the behavior interval change trend using the formula: ; Calculate the behavior interval change ratio , obtain the behavior interval change trend; in, represents the behavior interval change ratio, For the The time between behaviors, is the interval time of the previous behavior, is the total number of times the behavior occurred; First, the system calls the abnormal behavior impact ratio and extracts the duration sequence of dog behaviors from historical data, including the time interval between each behavior. The data source can be the behavior sensor worn by the dog or manual observation records. These data may include: movement behavior (such as running, walking, standing), resting behavior (such as lying down, stillness), abnormal behavior (such as impatience, excessive exercise); To ensure the integrity of the data, the preprocessing stage includes: removing outliers: if some interval time deviates significantly from the overall data (such as greater than 3 times the standard deviation of the mean), the data point is removed; time normalization: due to the different activity levels of different dogs, the time interval needs to be converted into relative values ​​so that the data of different individuals can be compared and analyzed; data sorting: sorting by timestamp to ensure the consistency of the time series when calculating the trend of behavior interval changes.

[0041] Calculate the time between behaviors: For a certain dog, the behavior time interval data can be expressed as: ; in, Representative The time interval between behaviors, represents the time interval of the previous behavior, is the total number of behaviors.

[0042] Calculate the change ratio of each behavior interval: ; For example, suppose a dog has the following behavior intervals: Second, Second, Second, Second Then calculate the change ratio of each time interval: ; Calculate the behavior interval change ratio: Using the given formula: ; Calculate the average behavior interval change ratio. For example, the calculation process for the above data is: ; ; Calculated Indicates the degree of fluctuation in the time interval of the dog's behavior. If this value is large, it means that the behavior interval fluctuates greatly, and the dog may be in an unstable behavior pattern, such as being stimulated by external factors or having anxious behavior. If the value is small, it means that the dog's behavior interval is relatively stable and conforms to the normal behavior pattern.

[0043] The behavior matching evaluation submodule builds a behavior continuity weight matrix based on the behavior interval change trend, calculates the matching degree of abnormal behavior in the overall behavior pattern, evaluates the matching degree, compares the probability of recurrence of normal behavior, screens candidate behavior data that meets the behavior pattern, and obtains the behavior matching evaluation results; The behavior matching evaluation submodule begins by analyzing the dog behavior data collected by the smart dog trainer, including the trend of its behavior interval changes, to establish a behavior continuity weight matrix. This matrix is ​​constructed based on the frequency and duration of each behavior in order to accurately calculate the weight of each behavior. For example, if a dog often performs sitting behavior during training, but rarely shows jumping behavior, the weight matrix will reflect this behavior pattern, and the weight of sitting behavior is higher than jumping. Then, the system uses this matrix to calculate the match between abnormal behavior and normal behavior patterns, and evaluates the degree of match by comparing the statistical probability difference between abnormal behavior and common behavior. Finally, the system will screen out candidate behavior data that matches the normal behavior pattern, including but not limited to those behaviors that occur occasionally but are still within the normal range, thereby generating a behavior matching evaluation result. For example, for a dog that suddenly begins to frequently try to escape, the system may mark this behavior as high risk and requires further intervention.

[0044] The behavior index screening submodule calls the behavior matching evaluation results, screens the data nodes of the behavior index, calculates the behavior impact range after the index update, compares the behavior change proportion before and after the abnormal behavior update, screens the behavior index update plan, and obtains the intervention strategy after the behavior update.

[0045] The behavior matching evaluation results are called to screen the behavior data nodes. This process includes calculating the impact range of each data node to determine which behavior nodes need to be updated. For example, if a dog's escape behavior is marked as abnormal, the relevant behavior nodes will be identified as needing to be updated to modify the behavior pattern and training methods. After that, the system will calculate the impact range of the behavior after the index update and evaluate the proportion of behavior changes by comparing the data before and after the abnormal behavior update. This calculation helps determine the most effective behavior intervention strategy, such as adding specific training tasks or adjusting training intensity. Finally, the screened behavior index update plan will be used to optimize the training program and obtain intervention strategies after the behavior is updated. This may include fine-tuning the training method or re-evaluating the dog's behavior pattern to ensure that the training effect is consistent with expectations.

[0046] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. The animal behavior management system based on the intelligent dog training device is characterized by: The system comprises: The behavior data collection module obtains dog behavior data, calls the dog training device sensor data, records the movement trajectory and environmental factors, and screens the environmental behavior characteristic data; The high-dimensional offset calculation module extracts time series features based on the environmental behavior feature data, calculates the trajectory position difference and the offset rate, records the current behavior path, determines the offset change rate, and combines the motion acceleration and direction change to screen the behavior transition rate to generate an offset dynamic prediction result; The behavior boundary recognition module calls the dynamic prediction result of the deviation, records the continuity destruction of abnormal behavior, divides the time period, calculates the movement frequency, sprint proportion, and static time, analyzes the change of movement rhythm, identifies the behavior path deviation, and generates the behavior fuzzy interval; The abnormal impact analysis module evaluates the impact of abnormal behavior on the normal mode based on the behavior fuzzy interval, calculates the decrease in the probability of behavior transfer, the behavior delay and the break rate, and generates the abnormal behavior impact ratio; The behavior index update module calls the abnormal behavior impact ratio, analyzes the behavior interval change, calculates the new behavior matching degree, compares the behavior recurrence probability, screens the index update data, and generates the behavior post-update intervention strategy.

2. The animal behavior management system based on the intelligent dog training device according to claim 1 is characterized in that: The environmental behavior feature data includes motion feature data, trajectory data, and environmental feature data; the offset dynamic prediction result includes trajectory change feature, offset rate feature, and abnormal behavior offset value; the behavior fuzzy interval includes behavior continuity feature, time partition feature, and behavior path change feature; the abnormal behavior impact proportion includes behavior transfer feature, abnormal behavior impact feature, and behavior fracture feature; the behavior post-update intervention strategy includes behavior interval feature, new behavior matching feature, and behavior index update feature.

3. The animal behavior management system based on the intelligent dog training device according to claim 1, characterized in that: The behavior data collection module includes: The motion data extraction submodule calls the built-in sensor data of the intelligent dog trainer to obtain the dog's motion acceleration, behavior sequence, activity frequency, calculate the behavior duration, count the rhythm change rate and the stability of the adjacent behavior sequence, analyze the dog's motion trajectory points, record the number of sprints and the proportion of static time, and generate motion feature data; The behavior pattern analysis submodule identifies the dog's continuous behavior pattern based on the motion feature data, analyzes the behavior conversion trend, extracts the probability of behavior occurrence, and calculates the correlation between the dog's differentiated behaviors using the formula: ； The stability coefficient of the dog's behavior sequence is calculated and combined with the behavior conversion trend to obtain the behavior pattern characteristics; in, Represents behavioral pattern characteristics, Represents the duration of a single behavior, represents the behavior transition interval, is the total number of behaviors, is the time attenuation coefficient, represents the total duration of the behavior sequence, represents the sum of the products of the duration of all behaviors and the interval between behavioral transitions; The environmental characteristic screening submodule calculates the impact of environmental characteristic changes on the behavioral pattern based on the behavioral pattern characteristics and combines the noise level, light intensity, and ground vibration conditions of the space where the dog is located, screens highly correlated environmental variables, and obtains environmental behavioral characteristic data.

4. The animal behavior management system based on the intelligent dog training device according to claim 1, characterized in that: The high-dimensional offset calculation module includes: The trajectory position calculation submodule extracts the time series behavior characteristics based on the environmental behavior characteristic data, obtains the motion trajectory coordinates of adjacent time points, calculates the trajectory position change value within each time step, sets the time interval, calculates the trajectory vector change amount, and obtains the trajectory position difference value by combining the trajectory data; The deviation rate analysis submodule calculates the deviation rate between adjacent behavior feature vectors based on the trajectory position difference, using the formula: ； Calculate and obtain the behavior deviation rate, compare the behavior trajectory data with the current behavior path data, and obtain the behavior deviation change rate; in, represents the behavior deviation rate, Representative The trajectory coordinate value at each time point, Representative The target trajectory coordinate value at each time point, is the total number of trajectory points within the time step, is the trajectory deviation adjustment factor, Represents the deviation trend of the behavior trajectory, is the behavior vector change rate correction; The abnormal behavior screening submodule calculates the abnormal behavior offset value based on the change rate of the behavior offset, and screens the data with a larger behavior conversion rate in combination with the dog's movement acceleration and direction change to obtain the offset dynamic prediction result.

5. The animal behavior management system based on the intelligent dog training device according to claim 1, characterized in that: The behavior boundary recognition module includes: The behavior continuity calculation submodule extracts the behavior sequence before and after the abnormal behavior based on the offset dynamic prediction result, calculates the time interval between continuous behaviors, counts the behavior destruction rate, divides the behavior time period, and obtains the behavior continuity destruction value; The motion rhythm analysis submodule calculates the motion frequency, sprint proportion, and static duration for each time period based on the behavior continuity destruction value, and uses a multi-scale behavior distribution analysis method to construct a behavior change matrix within the time period, and analyzes the motion state stability of the time partition using the formula: ； Obtain the amplitude of the movement rhythm change by calculation, compare the calculation results of adjacent time periods, construct the behavior transfer trend vector, establish the time dynamic transfer matrix, and obtain the behavior transfer change coefficient; in, Represents the amplitude of the movement rhythm change, Representative The movement frequency value at each time point, Representative The duration of stillness at a time point, is the total number of behavior records within the time step, is the behavioral volatility weight, Representative The rate of change of behavior at a time point, represents the behavior change cycle parameter, is the behavior change smoothing factor, Represents the degree of fluctuation of the behavior range, is the motion adjustment factor, Represents the change parameter of sprint behavior proportion, A dispersion correction parameter representing the behavior rhythm within a time period; The behavior path comparison submodule compares the normal behavior path with the abnormal behavior path based on the behavior transfer variation coefficient, calculates the path fitting index, screens the similar path intervals, and obtains the behavior fuzzy intervals.

6. The animal behavior management system based on the intelligent dog training device according to claim 1, characterized in that: The abnormal impact analysis module includes: The behavior sequence stability assessment submodule extracts the behavior sequence before and after the abnormal behavior based on the behavior fuzzy interval, compares the behavior stability before and after the abnormality, calculates the distribution probability of the behavior category within the time period, calculates the behavior frequency fluctuation, determines the disturbance degree of the behavior sequence, calculates the behavior stability reduction ratio according to the disturbance measurement standard, and obtains the behavior sequence stability offset value; The behavior transfer impact calculation submodule calculates the decrease in the behavior transfer probability within the abnormal behavior period based on the behavior sequence stability offset value, calculates the conversion rate from normal behavior to abnormal behavior, constructs a time segmented behavior impact model, and calculates the dynamic impact of abnormal behavior on the normal behavior sequence, using the formula: ； Calculate the behavior transfer break rate, analyze the continuous interference of abnormal behavior on normal behavior in the behavior pattern, and calculate the overall change range of the behavior pattern; in, represents the behavioral transfer break rate, Representative The normal behavior transition probability at a time point, represents the transition probability when abnormal behavior occurs, is the total number of behavior records within the time step, represents the duration of normal behavior, represents the duration of abnormal behavior; The abnormal behavior impact assessment submodule calculates the impact ratio of abnormal behavior on normal behavior patterns based on the behavior transfer fracture rate, extracts the impact of abnormal behavior on normal behavior, analyzes the expansion range of abnormal behavior on the time axis, analyzes the impact degree of abnormal behavior on behavior stability, establishes a measurement index of the impact of abnormality on behavior patterns, and obtains the impact ratio of abnormal behavior.

7. The animal behavior management system based on the intelligent dog training device according to claim 1, characterized in that: The behavior index updating module includes: The behavior interval analysis submodule calls the abnormal behavior impact ratio, obtains the duration sequence of the dog's behavior, calculates the behavior interval time based on the behavior data, calculates the behavior interval change rate, analyzes the correlation between the abnormal behavior recurrence probability and the normal behavior interval, extracts the behavior interval change trend, and uses the formula: ； Calculate the behavior interval change ratio , obtain the behavior interval change trend; in, represents the behavior interval change ratio, For the The time between behaviors, is the interval time of the previous behavior, is the total number of times the behavior occurred; The behavior matching evaluation submodule constructs a behavior continuity weight matrix based on the behavior interval change trend, calculates the matching degree of abnormal behavior in the overall behavior pattern, evaluates the matching degree, compares the probability of recurrence of normal behavior, screens candidate behavior data that meets the behavior pattern, and obtains the behavior matching evaluation result; The behavior index screening submodule calls the behavior matching evaluation result, screens the data nodes of the behavior index, calculates the behavior impact range after the index update, compares the behavior change proportion before and after the abnormal behavior update, screens the behavior index update plan, and obtains the intervention strategy after the behavior update.

Images