Directional driving method and system based on intelligent identification, medium and program product

Abstract

The invention discloses a directional driving method and system based on intelligent recognition, a medium and a program product, and the method comprises the steps: recognizing animal types in a target region; extracting three-dimensional coordinates of an upper boundary point, a lower boundary point, a left boundary point and a right boundary point of the animal from the depth data; calculating a longitudinal orientation angle and a transverse orientation angle; based on the pointing angle, ultrasonic waves with the wave band and the frequency changing from small to large are adopted for first-time driving; if the first-time driving is not successful, performing second-time driving by adopting a preset laser color of which the intensity is changed from weak to strong and which corresponds to the animal type based on the pointing angle; recording an ultrasonic parameter or a laser parameter when the animal leaves the target area, and storing the ultrasonic parameter or the laser parameter in a preset animal repelling parameter library, wherein the repelling parameter corresponds to the animal type; and under the condition that the driving parameters corresponding to the animal type are found in the preset animal driving parameter library, driving the animal according to the driving parameters. The animal driving method and device are used for improving animal driving timeliness, and then the driving effect is improved.

Description

Technical Field

[0001] This application belongs to the field of identification-oriented repellent, and in particular relates to a method, system, medium and program product based on intelligent identification-oriented repellent. Background Technology

[0002] In the field of animal deterrence, due to the behavioral habits of animals and interference factors in different scenarios, the deterrence effect is often difficult to achieve the expected goal. Especially in places such as airports and farmlands, the unexpected appearance of birds and other animals can easily cause major accidents or serious economic losses. Traditional deterrence methods can only be passive and are difficult to effectively guarantee the deterrence effect. At the same time, they can also easily cause unnecessary disturbance to the animals that need protection.

[0003] In this technology, cameras can capture real-time images of a site, and computer vision technology can be used to identify animals appearing in the images. Subsequently, sound and light-based deterrent devices can be used to drive away the identified animals. This technology achieves automatic identification and deterrence of animals in the site, offering better real-time performance and automation compared to traditional manual methods.

[0004] However, existing image recognition-based intelligent herding methods have varying effects on different animal species and environmental conditions. Especially when animals are in complex and changeable locations, the lack of targeted herding strategies can easily lead to problems such as untimely or unsatisfactory herding results. Summary of the Invention

[0005] This application provides a method, system, medium, and program product based on intelligent identification and directional animal repelling, which can improve the timeliness of animal repelling and thus enhance the repelling effect.

[0006] In the first aspect, this application provides a method for intelligent identification and directional driving away, which receives video data and depth data collected by a monitoring camera, wherein the depth data includes three-dimensional point cloud information of animals in the target area; Identify animal types within the target area based on video and depth data; Once it is determined that the animal type belongs to the preset type of animal that needs to be driven away, the three-dimensional coordinates of the animal's upper boundary point, lower boundary point, left boundary point, and right boundary point are extracted from the depth data. The longitudinal orientation angle is calculated based on the three-dimensional coordinates of the upper and lower boundary points, and the lateral orientation angle is calculated based on the three-dimensional coordinates of the left and right boundary points. The longitudinal and lateral orientation angles are used to determine the pointing angle of the driving signal. If no animal type-matching de-driving parameter is found in the preset animal de-driving parameter library, the first de-driving action is performed using ultrasound with varying wavebands and frequencies from small to large, based on the pointing angle. If the first attempt to drive away the animal fails, a second attempt will be made using a preset laser color with varying intensity from weak to strong, based on the pointing angle and corresponding to the animal type. Record the ultrasonic or laser parameters that cause the animal to leave the target area, and store them in the preset animal repelling parameter library corresponding to the animal type. If a driving parameter corresponding to the animal type is found in the preset animal driving parameter library, the animal is driven away according to the driving parameter.

[0007] By employing the aforementioned technical solution, the system can accurately acquire three-dimensional spatial information of animals within the target area by receiving video and depth data from monitoring cameras. Based on this data, animal type identification and boundary point extraction are performed, and precise longitudinal and lateral orientation angles are calculated, ensuring that the repelling signal is accurately directed to the target animal. When no corresponding repelling parameters are available, an initial repelling attempt is made using ultrasound with increasing waveband and frequency. If unsuccessful, a second repelling attempt is made using laser with increasing intensity and customized color. This gradual repelling method reduces the degree of fright to the animals. The system records and stores the parameters from successful repelling attempts in a parameter database, enabling the accumulation and reuse of repelling experience. When encountering the same type of animal, verified and effective repelling parameters can be directly invoked, improving repelling efficiency. This intelligent identification and parameter-adaptive repelling method based on real-time three-dimensional data ensures the effectiveness of repelling while reducing stress responses in animals. Furthermore, continuous optimization of the parameter database enhances the system's intelligence level.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, identifying animal types within a target area based on video data and depth data specifically includes: Extract moving targets from video data and obtain the corresponding depth point cloud data of the moving targets; Construct a 3D feature model of the moving target based on deep point cloud data; The three-dimensional feature model is matched with a pre-set animal feature database to obtain preliminary identification results of animal types; Based on the behavioral characteristics of the moving target and the preliminary identification results, the final animal type is determined.

[0009] By employing the aforementioned technical solution, a multi-dimensional animal recognition system was constructed by extracting moving targets from videos and obtaining their depth point cloud data, combined with 3D feature model matching and behavioral feature analysis. The depth point cloud data provides precise 3D contour information of the target, avoiding recognition errors that may arise from relying solely on 2D images. By constructing a 3D feature model and matching it with a pre-set feature library, the accuracy of animal type recognition was significantly improved. Using behavioral features as an auxiliary criterion further validated the reliability of the preliminary recognition results. This recognition method, combining multi-source data, overcomes potential recognition errors that may occur with a single data source in complex environments, improves the system's ability to recognize animals in different postures and under different lighting conditions, and lays a reliable foundation for subsequent precise herding.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the longitudinal orientation angle is calculated based on the three-dimensional coordinates of the upper and lower boundary points, and the lateral orientation angle is calculated based on the three-dimensional coordinates of the left and right boundary points, specifically including: Construct a first vector group based on the three-dimensional coordinates of the upper and lower boundary points, and construct a second vector group based on the three-dimensional coordinates of the left and right boundary points; Calculate the angle between adjacent vectors in the first vector group to obtain the longitudinal orientation angle; Calculate the angle between adjacent vectors in the second vector group to obtain the lateral orientation angle.

[0011] By employing the aforementioned technical solution, and constructing two sets of vectors based on the upper and lower boundary points and the left and right boundary points respectively, the longitudinal and lateral orientation angles are calculated by the angle between adjacent vectors, thus achieving a precise description of the animal's spatial posture. The vector sets constructed based on three-dimensional coordinates accurately reflect the animal's position and orientation information in space, avoiding the angle calculation errors that may occur when using only two-dimensional projection. By calculating the longitudinal and lateral orientation angles separately, the system can comprehensively grasp the animal's spatial posture, making the direction of the driving signal more precise. This vector-based calculation method has a sound mathematical foundation, the calculation results are stable and reliable, and it is unaffected by changes in animal size and posture, ensuring that the driving signal always accurately points to the target animal.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, after driving the animals according to the driving parameters, the method further includes: Read the historical driving records corresponding to the animal type. The historical driving records include driving parameter values, start time, end time and number of drivings for each driving; Calculate the response time of the animal from receiving the driving signal to starting to move, and the departure time from starting to move to completely leaving the target area during the most recent preset first driving process; When the rate of change of the second consecutive preset number of response time or departure time exceeds the preset threshold, it is determined that the animal has adapted to the current driving parameters; Select at least two different sets of driving parameters from the preset animal driving parameter library, and randomly switch between the two different sets of driving parameters between two adjacent driving events; Acquire electronic map data of the target area, and based on the electronic map data, use a preset algorithm to calculate multiple feasible paths from the animal's current location to the boundary of the target area; Calculate the safety risk value for each feasible path. The safety risk value is determined based on the path length, slope variation, obstacle density, and distance to the restricted area. The feasible path with the lowest safety risk value is selected as the expected driving path, and the pointing angle of the driving signal is adjusted based on the expected driving path.

[0013] By employing the aforementioned technical solution and analyzing the changing trends of animal response and departure times in historical herding records, the adaptability of animals to herding parameters can be promptly identified. Randomly switching different herding parameters during adjacent herding processes effectively prevents animals from developing herding immunity. Multiple feasible paths are calculated using electronic map data, and safety risk values ​​are calculated by comprehensively considering factors such as path length, slope changes, obstacle density, and distance to restricted areas, ultimately selecting the optimal herding path. This intelligent path planning method not only ensures the safety of the herding process but also improves herding efficiency. Dynamically adjusting the pointing angle of the herding signal to align with the expected herding path enhances the controllability of the entire herding process and the achievement of the desired effect.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the security risk value on each feasible path is calculated, specifically including: Extract the distance and elevation difference between any two adjacent points on a feasible path, and calculate the path slope; Count the number and type of obstacles within a preset distance range around the feasible path; Calculate the shortest distance between the feasible path and the boundary of the restricted area; Based on preset weighting coefficients, the path length, slope change, obstacle density, and distance to restricted areas are weighted and summed to obtain the safety risk value for each feasible path.

[0015] By employing the aforementioned technical solution, the path slope is calculated by extracting the distance and height difference between adjacent points on a feasible path. The number and type of obstacles within a preset distance around the path are also statistically analyzed. Simultaneously, the shortest distance to the boundary of the restricted area is calculated. These factors are then weighted and summed according to preset weighting coefficients to obtain a safety risk value, enabling the system to comprehensively assess the safety of each feasible path. The calculation of the path slope avoids unexpected risks to animals due to drastic terrain undulations during herding. The statistics on obstacles prevent animals from colliding with or being blocked by obstacles during escape. The calculation of the distance to the restricted area ensures that animals do not accidentally enter dangerous or restricted areas. Through this multi-dimensional safety risk assessment method, the system can select the optimal herding path, ensuring both herding efficiency and minimizing the risk of injury to animals during the herding process, thus improving the safety of animal herding.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after adjusting the pointing angle of the driving signal based on the expected driving path, the method further includes: During the process of driving away animals, the movement trajectory of the animals is detected in real time, and the real-time location and movement speed of the animals are obtained. Calculate the deviation between the animal's real-time location and the expected driving path; When the deviation exceeds a preset distance threshold, the feasible path is recalculated based on the animal's real-time location; Select the feasible path with the lowest recalculated security risk value as the updated expected deportation path. Adjust the direction of the driving signal to align it with the direction of the updated expected driving path.

[0017] By employing the aforementioned technical solution, the system acquires the animal's real-time location and speed by detecting its movement trajectory in real time, and calculates the deviation distance from the expected herding path. When the deviation exceeds a threshold, the system replans the path, allowing it to dynamically adapt to the animal's actual movement. When an animal deviates from the expected path due to fright or environmental disturbance, the system immediately recalculates the lowest-risk feasible path based on its current location and adjusts the direction of the herding signal accordingly. This real-time adjustment mechanism promptly corrects the animal's movement direction, preventing it from continuously deviating from the ideal path and increasing the difficulty of herding or creating additional risks. Through continuous monitoring and dynamic adjustment during the herding process, the system improves the accuracy and controllability of herding, reducing the probability of herding failure.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, feasible paths are recalculated based on the animal's real-time location, specifically including: Based on electronic map data within a preset range around the animal's real-time location, a preset algorithm is used to calculate multiple new feasible paths from the animal's real-time location to the boundary of the target area; Calculate the safety risk value for the new feasible path; The new feasible paths are sorted according to their security risk values; The new feasible path with the lowest safety risk value is selected as the updated expected deportation path.

[0019] By employing the aforementioned technical solution, new feasible paths are calculated based on electronic map data within a preset range around the animal's real-time location. These paths are then evaluated for safety risk and ranked. The path with the lowest risk value is selected as the updated expected driving path, enabling the system to quickly find the optimal alternative after the animal deviates from its original path. This path update mechanism considers the actual terrain features of the animal's current location, ensuring that the newly calculated path conforms to the current situation while maximizing safety. Because the system still adheres to the principle of minimizing safety risk when replanning paths, the safety and controllability of the driving process can be maintained even when temporary path adjustments are necessary. This precise dynamic path optimization method improves the system's ability to adapt to complex environmental changes and enhances the robustness and reliability of the driving process.

[0020] Secondly, embodiments of this application provide an intelligent identification-based directional driving system, which includes: one or more processors and a memory; the memory is coupled to one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the system to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer program product that, when run on a system, causes the system to execute the method described in any possible implementation of the first aspect.

[0023] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. This application provides an intelligent identification-based directional determination method. By receiving video and depth data from a monitoring camera, it can accurately acquire the three-dimensional spatial information of animals within a target area. Based on this data, animal type identification and boundary point extraction are performed, and precise longitudinal and lateral orientation angles are calculated, enabling the determination signal to accurately target the animal. When no corresponding determination parameters are available, an initial determination attempt is made using ultrasound with increasing waveband and frequency. If unsuccessful, a second determination attempt is made using laser with increasing intensity and customized color. This gradual determination method reduces the degree of fright to the animal. The system records the parameters of successful determinations and stores them in a parameter database, enabling the accumulation and reuse of determination experience. When encountering the same type of animal, the verified effective determination parameters can be directly called, improving determination efficiency. This intelligent identification and parameter adaptive determination method based on real-time three-dimensional data ensures the effectiveness of determination while reducing stress responses to animals. Furthermore, continuous optimization of the parameter database enhances the system's intelligence level.

[0024] 2. This application provides an intelligent identification-based directional herding method. By analyzing the changing trends of animal response and departure times in historical herding records, it promptly identifies animal adaptations to herding parameters. By randomly switching different herding parameters during adjacent herding processes, it effectively prevents animals from developing herding immunity. Multiple feasible paths are calculated using electronic map data, and a safety risk value is calculated by comprehensively considering factors such as path length, slope changes, obstacle density, and distance to restricted areas, selecting the optimal herding path. This intelligent path planning method not only ensures the safety of the herding process but also improves herding efficiency. By dynamically adjusting the pointing angle of the herding signal to align with the expected herding path, it enhances the controllability of the entire herding process and the achievement of the expected results.

[0025] 3. This application provides an intelligent identification-based directional herding method. It obtains the real-time position and speed of the animal by detecting its movement trajectory in real time, and calculates the deviation distance from the expected herding path. When the deviation distance exceeds a threshold, the path is replanned, allowing the system to dynamically adapt to the animal's actual movement. When the animal deviates from the expected path due to fright or environmental disturbance, the system immediately recalculates the feasible path with the lowest safety risk based on the current position and adjusts the pointing angle of the herding signal accordingly. This real-time adjustment mechanism can promptly correct the animal's movement direction, preventing the animal from continuously deviating from the ideal path, thus avoiding increased herding difficulty or additional risks. Through continuous monitoring and dynamic adjustment during the herding process, the system improves the accuracy and controllability of herding, and reduces the probability of herding failure. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating a method for intelligent identification and directional driving away in an embodiment of this application.

[0027] Figure 2 This is another flowchart illustrating a method for intelligent identification and directional driving away in an embodiment of this application.

[0028] Figure 3 This is a schematic diagram of the physical device structure of an intelligent identification and directional driving system provided in an embodiment of this application. Detailed Implementation

[0029] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0030] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0031] The following example is used in conjunction with Figure 1 The present application describes a method for targeted expulsion based on intelligent identification in an embodiment of the present application: Please see Figure 1 This is a flowchart illustrating a method for intelligent identification and directional driving away in an embodiment of this application.

[0032] S101. Receive video data and depth data collected by the surveillance camera, and identify the animal types within the target area based on the video data and depth data; The system receives video data and depth data collected by surveillance cameras. The depth data includes 3D point cloud information of animals within the target area. Based on the video data and depth data, the system identifies the animal types within the target area. Specifically, this includes: extracting moving targets from the video data and obtaining the corresponding depth point cloud data; constructing a 3D feature model of the moving target based on the depth point cloud data; matching the 3D feature model with a preset animal feature database to obtain preliminary identification results of the animal type; and determining the final animal type based on the behavioral characteristics of the moving target and the preliminary identification results.

[0033] In this step, the system first receives video and depth data from the surveillance camera. The video data is real-time visual information of the target area, while the depth data contains three-dimensional information such as the distance and contour of objects within the target area. By fusing and analyzing the video and depth data, the system can identify animals within the target area. The animal identification in this step is not limited to a specific algorithm; for example, it can use deep learning-based object detection and classification algorithms, or traditional feature extraction and matching methods.

[0034] In one specific implementation, the system can employ the following techniques to achieve animal identification: First, a moving target detection algorithm is used to extract moving targets from video data, obtaining their corresponding depth point cloud data. Then, a three-dimensional feature model of the target is constructed based on the depth point cloud data. Next, the three-dimensional feature model is matched against a pre-set animal feature database to obtain preliminary animal type identification results. Finally, the system can further combine the moving target's behavioral characteristics, such as movement speed and trajectory, to optimize and determine the final animal type identification result.

[0035] S102. If the animal type is determined to be a preset animal type that needs to be driven away, then extract the three-dimensional coordinates of the animal's upper boundary point, lower boundary point, left boundary point and right boundary point from the depth data. This step involves extracting the 3D coordinates of the boundary points of animals that need to be herded, providing data support for subsequent calculations of the herding angle. Specifically, the system first determines whether the identified animal type belongs to a preset category that needs to be herded. If so, it extracts the 3D coordinates of the animal's top, bottom, left, and right boundary points from the depth data. It's important to note that the origin and axis orientation of the 3D coordinate system used for the depth data should be consistent with the camera's position and orientation. The system can set a certain threshold based on the animal's size to determine the boundary points, eliminating interference from protruding parts (such as ears and tails).

[0036] S103. Calculate the longitudinal orientation angle based on the three-dimensional coordinates of the upper and lower boundary points, and calculate the lateral orientation angle based on the three-dimensional coordinates of the left and right boundary points. The system calculates the longitudinal orientation angle based on the three-dimensional coordinates of the upper and lower boundary points, and the lateral orientation angle based on the three-dimensional coordinates of the left and right boundary points. The longitudinal and lateral orientation angles are used to determine the pointing angle of the driving signal. Specifically, a first vector group is constructed based on the three-dimensional coordinates of the upper and lower boundary points, and a second vector group is constructed based on the three-dimensional coordinates of the left and right boundary points. The angle between adjacent vectors in the first vector group is calculated to obtain the longitudinal orientation angle; the angle between adjacent vectors in the second vector group is calculated to obtain the lateral orientation angle.

[0037] After obtaining the three-dimensional coordinates of the animal's boundary points, the system needs to further calculate the pointing angle of the driving device, including both longitudinal and lateral dimensions. The longitudinal orientation angle is calculated based on the coordinates of the animal's upper and lower boundary points. Specifically, two vectors can be constructed: one pointing from the upper boundary point to the center point, and the other pointing from the center point to the lower boundary point. The angle between these two vectors yields the longitudinal angle. Similarly, the lateral orientation angle can be calculated by the angle between two vectors constructed from the coordinates of the left and right boundary points. The system then controls the pitch and rotation angles of the driving device based on the calculated angles, ensuring that the sound waves or beams are accurately directed at the animal's location.

[0038] In practical applications, due to the movement characteristics of animals, the coordinates of their boundary points may change over time, leading to inaccurate calculated angles. To address this issue, the system can employ a smoothing filtering algorithm to track and update the boundary point coordinates, reducing angle fluctuations caused by changes in animal posture. Furthermore, the system can predict the animal's position over a future period based on its movement trends, adjusting the driving angle in advance for more proactive and real-time driving control.

[0039] S104. If it is determined that no driving parameter corresponding to the animal type is found in the preset animal driving parameter library, the first driving is carried out using ultrasound with both the waveband and frequency changing from small to large based on the pointing angle. For animal types not yet included in the repulsion parameter database, the system will attempt to repel them using parameter combinations that gradually increase in size. The first method is ultrasonic repulsion. Based on the pointing angle calculated in step S103, the system controls the ultrasonic transmitter, sequentially changing the ultrasonic parameters from low to high frequency and from narrow to wide bands, observing the animal's reaction. The advantages of ultrasonic repulsion are low energy consumption, low equipment cost, and no potential harm to humans or animals. However, its disadvantages include a short repulsion distance and the possibility of animal adaptation.

[0040] S105. If the first attempt to drive away the animal fails, a second attempt will be made using a preset laser color with varying intensity from weak to strong, based on the pointing angle and corresponding to the animal type. When ultrasonic repelling fails, the system initiates a secondary repelling process, using a laser to drive the animal away. Compared to ultrasound, laser repelling offers advantages in terms of longer range and stronger directionality, but it is relatively more energy-intensive and costly. The system pre-sets laser colors corresponding to different animals (e.g., green for birds, red for mammals), and during repelling, combines laser color and intensity enhancement from low to high until the animal is driven out of the target area. Similarly, the laser emission angle is determined based on the previously calculated animal boundary points.

[0041] Considering the potential harm of lasers to animals' eyes, the system should monitor the animals' activity in real time during the herding process. If any abnormalities occur, such as prolonged immobility, laser irradiation should be immediately stopped, and alternative herding methods should be used, or management personnel should be notified to handle the situation on-site. Furthermore, the system can dynamically adjust the laser intensity based on the ambient light intensity at the herding site, minimizing harm to the animals while ensuring effective herding.

[0042] S106. Record the ultrasonic or laser parameters that cause the animal to leave the target area and store them in the preset animal driving parameter library corresponding to the animal type. During the herding process, once an animal is detected leaving the target area, it indicates that the currently used herding parameters are effective for that animal type. At this point, the system should record these parameters and write them into a preset herding parameter library, establishing a mapping relationship with the corresponding animal type. The accumulation and retention of these effective parameters can be used for rapid herding when the animal reappears, and also provides a reference for herding the same animal in other areas and situations.

[0043] In addition to recording driving parameters, the parameter library can also correlate other data during the driving process, such as driving duration and animal escape trajectory. This will help evaluate and optimize the performance of driving strategies and devices. Considering that animals may exhibit seasonal and regional behavioral differences, the parameter library should also support the organization and management of driving data according to multiple dimensions such as time and location.

[0044] S107. If it is determined that a driving parameter corresponding to the animal type is found in the preset animal driving parameter library, drive the animal according to the driving parameter.

[0045] When the system retrieves the appropriate deflection parameters for the identified animal type from the parameter library, it can directly call these parameter settings to control the deflection equipment to complete the deflection task quickly and efficiently. Compared to the gradual attempts in steps S104 and S105, directly using the optimal parameters will significantly shorten the deflection time and reduce the consumption of system resources. Furthermore, since the parameters used have been validated in practice, the deflection success rate will also be significantly improved.

[0046] Even when using preset parameters, the system should continuously monitor the animal's state and reactions during the herding process. If any adverse effects occur, such as the animal exhibiting aggression or stress, the system should immediately stop the current herding attempt, switch to another approach, or issue an alarm to prompt manual intervention, ensuring the safety of the herding process. Furthermore, since animal behavior may change over time, the system should periodically verify and update its parameter database, eliminating outdated herding parameters and adding new, optimized combinations to ensure the herding plan remains up-to-date and adapts to changes in animal behavior.

[0047] In the above embodiments, by receiving video and depth data collected by a monitoring camera, the three-dimensional spatial information of animals within the target area can be accurately obtained. Based on this data, animal type identification and boundary point extraction are performed, and precise longitudinal and lateral orientation angles are calculated, enabling the repelling signal to accurately target the animal. When no corresponding repelling parameters are available, an initial repelling attempt is made using ultrasound with increasing band and frequency. If unsuccessful, a second repelling attempt is made using laser with increasing intensity and customized color. This gradual repelling method reduces the degree of fright to the animals. The system records the parameters of successful repellings and stores them in a parameter database, enabling the accumulation and reuse of repelling experience. When encountering the same type of animal, the verified and effective repelling parameters can be directly called, improving repelling efficiency. This intelligent identification and parameter-adaptive repelling method based on real-time three-dimensional data ensures the effectiveness of repelling while reducing stress responses in animals. Furthermore, continuous optimization of the parameter database enhances the system's intelligence level.

[0048] The first embodiment introduced a basic animal identification and herding process, including how to acquire animal information, calculate the herding angle, and establish and use a herding parameter database. However, in practical applications, animals may adapt to fixed herding methods, and the choice of herding path has a significant impact on the herding effect and safety. The following section combines... Figure 2 Another method for targeted expulsion based on intelligent identification in the embodiments of this application is described below: Please see Figure 2 This is another flowchart illustrating a method for intelligent identification and directional driving away in an embodiment of this application.

[0049] S201. Read the historical driving records corresponding to the animal type, and calculate the response time of the animal from receiving the driving signal to starting to move, and the departure time from starting to move to completely leaving the target area during the most recent preset first driving process. The system reads the historical driving records corresponding to the animal type. The historical driving records include driving parameter values, the start time, end time and number of driving times for each driving. It also calculates the response time from receiving the driving signal to starting to move, and the departure time from starting to move to completely leaving the target area during the most recent preset first driving process.

[0050] This step aims to assess the animal's response speed and driving efficiency by analyzing recent driving history data. Specifically, the system first reads historical driving records associated with the currently identified animal type. These records should include key information such as parameter settings used for each driving action, start and end times, and number of driving attempts. After obtaining this data, the system will focus on calculating the animal's response time and departure time in the most recent few driving attempts (a preset number, such as the last 5). Response time refers to the time interval from when the animal first receives the driving signal to when it exhibits substantial avoidance behavior (such as starting to move). Departure time refers to the time taken from when the animal begins to move until it completely leaves the target driving area. These two indicators can quantitatively reflect the animal's sensitivity to driving signals and the overall effectiveness of the driving.

[0051] In one specific implementation, the system can calculate the response time and departure time as follows: First, for each driving record, extract the start time T1 of the driving signal, the time T2 when the animal begins to move, and the time T3 when the animal leaves the area. Then, the response time can be obtained by subtracting T1 from T2, and the departure time is T3-T2. Considering that animal activity may have a certain degree of randomness, the system can take the average of the response time and departure time of the most recent driving records to reduce the influence of random factors. It should be noted that the time points T1, T2, and T3 can be obtained by analyzing the monitoring video stream or the trajectory data of the animal's location. Meanwhile, the boundary of the target area should also be pre-defined based on actual needs and environmental characteristics.

[0052] In practice, animals may linger near the herding area for extended periods, making it difficult to accurately determine the end time of the herding process. To address this issue, the system can set a herding timeout threshold. If an animal has been away from the area for more than this threshold (e.g., 30 minutes), even if the animal has not completely moved away, the current time will be recorded as the end point of the herding process, forcing the next round of herding to begin. Additionally, the system can weight the departure time based on the animal's movement speed and distance from the boundary.

[0053] S202. When the rate of change of the second consecutive preset number of response time or departure time exceeds the preset threshold, it is determined that the animal has adapted to the current driving parameters. This step determines whether the animal has become adapted to or desensitized to the current herding strategy. The system tracks and calculates the rate of change of the animal's response time or departure time during the most recent several herding attempts (a preset number, such as the last three). The rate of change measures the trend of these two indicators as the number of herding attempts increases. It can be calculated by linearly fitting the response time (or departure time) of the most recent N herding attempts, and the absolute value of the slope is the rate of change. When the rate of change exceeds a preset threshold (e.g., 20%), it can be inferred that the animal has begun to adapt to the current herding pattern, and the herding effect will gradually decrease over time.

[0054] It's important to note that response and departure times for a single chase may fluctuate due to environmental disturbances and individual animal differences. Therefore, when calculating the rate of change, the system can consider data from multiple consecutive chases to smooth out the impact of occasional disturbances. Furthermore, exceeding a threshold is only one preliminary indicator of animal adaptation. The system can also combine other factors, such as the animal's activity trajectory and the success rate of the chase, to comprehensively assess the animal's adaptation. For example, if a high rate of change is accompanied by habitual behaviors such as wandering or returning during the chase, the likelihood of adaptation increases further.

[0055] S203. Select at least two different sets of driving parameters from the preset animal driving parameter library, and randomly switch between the two different sets of driving parameters between two adjacent driving events. This step is a specific strategy for addressing animal adaptation. The basic idea is to randomly switch between multiple sets of preset drive-off parameters, making it difficult for the animal to quickly form a stable conditioned reflex, thereby reducing the risk of adaptation. First, the system filters two or more sets of drive-off parameters from historical drive-off data that are effective for the current animal type. These parameters have been proven in previous drive-offs to induce avoidance behavior. Then, during the actual drive-off process, the system uses a random switching method, alternating between different sets of parameters between two adjacent drive-offs. This switching can be completely random, or it can follow a certain probability distribution (such as a uniform distribution, normal distribution, etc.), giving different parameter sets a chance to be selected. The specific form of the probability distribution can be determined based on factors such as the previous drive-off effects and frequency of use of each parameter set.

[0056] For example, if the parameter library contains three sets of historical driving parameters (denoted as A, B, and C), the system can design the following random switching strategy: Each time a parameter needs to be switched, a random number r uniformly distributed between 0 and 1 is generated. If r < 1 / 3, parameter set A is selected; if 1 / 3 ≤ r < 2 / 3, parameter set B is selected; if r ≥ 2 / 3, parameter set C is selected. In this way, the three sets of parameters have equal probabilities of being selected, increasing the randomness of the driving process. The system can also dynamically adjust the selection probability of each parameter set based on the latest driving feedback. For example, if using parameter set A significantly prolongs the animal's response time, the system can appropriately reduce the weight of A in the random switching to reduce the waste of time and energy caused by ineffective driving.

[0057] S204. Obtain electronic map data of the target area, and based on the electronic map data, use a preset algorithm to calculate multiple feasible paths from the animal's current location to the boundary of the target area; This step incorporates electronic map data to support the optimization of animal herding paths. The system first acquires an electronic map of the target area, which should include key information such as terrain, buildings, and roads, presented in a machine-readable digital format. Then, using the animal's current location as the starting point and the boundary of the target area as the ending point, the system runs a pre-defined path planning algorithm to generate several feasible herding paths from the starting point to the ending point. Feasibility here primarily considers factors such as path connectivity, length, and the complexity of the terrain along the route. The algorithm for calculating the path can employ traditional graph search methods, such as Dijkstra's algorithm or A* algorithm. Alternatively, heuristic intelligent optimization algorithms, such as ant colony optimization or genetic algorithms, can be used. The choice of algorithm depends on the size of the electronic map, the real-time requirements of path planning, and the constraints of computing resources.

[0058] For example, if the map of the target area can be represented as a directed weighted graph G(V, E), where V are the nodes in the graph and E are the edges connecting the nodes, the system can use Dijkstra's algorithm, taking the node corresponding to the animal's current position as the source node, and run a single-source shortest path algorithm to find the shortest path between the source node and all nodes on the target boundary. These paths can serve as alternative solutions for driving the animal away. Considering the complexity of actual terrain, the edges between nodes should not be simple straight lines, but should correspond to roads, paths, etc. in reality. The weight of the edge can be calculated based on parameters such as road length, width, and curvature to reflect the difficulty of traversing the road segment.

[0059] S205. Calculate the safety risk value on each feasible path, select the feasible path with the lowest safety risk value as the expected driving path, and adjust the pointing angle of the driving signal based on the expected driving path.

[0060] The system calculates the safety risk value for each feasible path, specifically including: extracting the distance and height difference between any two adjacent points on the feasible path, calculating the path slope; counting the number and type of obstacles within a preset distance range around the feasible path; calculating the shortest distance between the feasible path and the boundary of the restricted area; and, based on preset weighting coefficients, weighting and summing the path length, slope change, obstacle density, and distance to the restricted area to obtain the safety risk value for each feasible path. The feasible path with the lowest safety risk value is selected as the expected deportation path, and the pointing angle of the deportation signal is adjusted based on the expected deportation path.

[0061] The system needs to select the optimal route from multiple feasible paths to drive away animals, using a safety risk value as the evaluation metric. A lower safety risk value indicates a safer path for the animals, potentially leading to a better driving effect. The key to this step is defining and calculating the safety risk value. The system can comprehensively consider factors such as path length, slope variation, obstacle distribution along the route, and distance from restricted areas, obtaining the safety risk value for each path through a weighted summation, and selecting the path with the lowest risk as the driving strategy. After calculation, the pointing angle of the driving signal needs to be adjusted according to the selected path to guide the animals in the desired direction.

[0062] In practical implementation, the system can adopt the following technical solution: First, for each feasible path, extract the three-dimensional coordinates of any two adjacent points, calculate the spatial distance and height difference between the two points, and obtain the local path slope. Then, with the path as the center, set a certain buffer zone, count the number of different types of obstacles (such as trees, rocks, etc.) in the zone, and assess the path's traversal difficulty. Next, calculate the shortest distance from each point on the path to the boundary of the restricted area, generating a distance field between the path and the restricted area. Finally, assign weight coefficients to the four indicators of path length, slope change, obstacle density, and distance to the restricted area, and sum them up to obtain a comprehensive safety risk value. The weight coefficients can be adjusted according to the habits of different animals and the terrain characteristics of the driving area. After calculating the risk values ​​of all paths, select the path with the smallest value as the driving route, and adjust the driving angle accordingly to make the animal move along that path.

[0063] In the above embodiments, by analyzing the changing trends of animal response and departure times in historical herding records, the adaptability of animals to herding parameters can be detected in a timely manner. By randomly switching different herding parameters during adjacent herding processes, herding immunity in animals is effectively prevented. Multiple feasible paths are calculated using electronic map data, and safety risk values ​​are calculated by comprehensively considering factors such as path length, slope changes, obstacle density, and distance to restricted areas, selecting the optimal herding path. This intelligent path planning method not only ensures the safety of the herding process but also improves herding efficiency. By dynamically adjusting the pointing angle of the herding signal to align it with the expected herding path, the controllability of the entire herding process and the degree to which the expected results are achieved are enhanced.

[0064] Furthermore, in another embodiment, after adjusting the pointing angle of the driving signal based on the expected driving path, the method further includes: detecting the animal's movement trajectory in real time during the driving process, and obtaining the animal's real-time position and movement speed. Calculate the deviation between the animal's real-time location and the expected driving path; When the deviation exceeds a preset distance threshold, the feasible path is recalculated based on the animal's real-time location. Specifically, this includes: calculating multiple new feasible paths from the animal's real-time location to the target area boundary using a preset algorithm based on electronic map data within a preset range around the animal's real-time location; calculating the safety risk value of the new feasible paths; sorting the new feasible paths according to the safety risk value; selecting the new feasible path with the lowest safety risk value as the updated expected driving path; and selecting the recalculated feasible path with the lowest safety risk value as the updated expected driving path. Adjust the direction of the driving signal to align it with the direction of the updated expected driving path.

[0065] Building upon the above embodiments, the system can further incorporate a real-time path update mechanism to adapt to random changes in animal movement trajectories. Specifically, during the herding process, the system detects the animal's position coordinates and instantaneous velocity vector in real time, calculating its lateral deviation distance from the expected herding path. Once this distance exceeds a preset threshold, it is considered that the animal has significantly deviated from the original route, at which point a path replanning process needs to be triggered.

[0066] The core of path replanning is to find new feasible paths based on the animal's current location and a comprehensive analysis of the surrounding terrain features. The system first acquires high-precision electronic map data within a preset radius, centered on the animal's real-time location. Then, using intelligent algorithms such as heuristic search and random trees, it calculates several alternative paths from the current location to the target area boundary. Next, a safety risk assessment is performed on each alternative path, similar to the previous steps, comprehensively considering the path's geometric parameters and surrounding environmental information to obtain a risk coefficient. After assessing all paths, they are sorted according to their risk coefficients from lowest to highest, and the path with the lowest risk is selected as the new expected driving path. Finally, the system adjusts the pointing angles of driving signals such as ultrasound and lasers to align with the direction vector of the updated path, guiding the animal back to a safe track.

[0067] The rationale behind the above real-time path update scheme lies in its ability to fully utilize real-time feedback on animal location and, combined with dynamic changes in the surrounding environment, continuously adapt and adjust the herding strategy, thus maintaining a high success rate even under complex conditions. However, this scheme also has significant limitations: frequent path replanning and herding angle adjustments may exacerbate animal stress responses, leading to more uncontrollable behavior. Therefore, balancing path optimization with animal behavioral stability, and setting reasonable update trigger mechanisms and frequencies, are key issues that need to be considered in system implementation.

[0068] One possible improvement is to adopt a prediction-based path update strategy. Instead of immediately initiating replanning after detecting an animal deviating from the planned route, this strategy uses machine learning algorithms to analyze the animal's movement trends over a previous period, predicting its location distribution in the near future, and then making local adjustments to the path based on the prediction. This can smooth out frequent updates caused by random animal wandering, making the herding process smoother. Furthermore, the system can build behavioral preference models for different animal species and individuals, learning and mastering their movement habits to provide prior knowledge for subsequent herding decisions, further reducing blind attempts and improving the accuracy of herding.

[0069] In the above embodiments, the system obtains the real-time position and speed of the animal by detecting its movement trajectory in real time, and calculates the deviation distance from the expected driving path. When the deviation distance exceeds a threshold, the system replans the path, enabling it to dynamically adapt to the animal's actual movement. When the animal deviates from the expected path due to fright or environmental disturbance, the system immediately recalculates the feasible path with the lowest safety risk based on the current position and adjusts the pointing angle of the driving signal accordingly. This real-time adjustment mechanism can promptly correct the animal's movement direction, preventing the animal from continuously deviating from the ideal path, thus avoiding increased driving difficulty or additional risks. Through continuous monitoring and dynamic adjustment during the driving process, the system improves the accuracy and controllability of driving and reduces the probability of driving failure.

[0070] The system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of a physical device structure based on an intelligent identification and directional driving system provided in an embodiment of this application.

[0071] It should be noted that, Figure 3 The structure of the system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0072] like Figure 3 As shown, the system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on a program stored in Read-Only Memory (ROM) 302 or a program loaded from storage portion 308 into Random Access Memory (RAM) 303, such as executing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0073] The following components are connected to I / O interface 305: input section 306 including a camera, infrared sensor, etc.; output section 307 including a liquid crystal display (LCD) and speakers, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0074] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the various functions defined in the present invention.

[0075] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0076] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0077] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or it may exist independently and not assembled into the system. The storage medium carries one or more computer programs that, when executed by a processor of a system, cause the system to implement the methods provided in the above embodiments.

[0078] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0079] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0080] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0081] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for targeted repelling based on intelligent identification, characterized in that, include: Receive video data and depth data collected by a surveillance camera, wherein the depth data includes three-dimensional point cloud information of animals within the target area; The animal types within the target area are identified based on the video data and the depth data; If the animal type is determined to be a preset animal type that needs to be driven away, then the three-dimensional coordinates of the animal's upper boundary point, lower boundary point, left boundary point and right boundary point are extracted from the depth data; The longitudinal orientation angle is calculated based on the three-dimensional coordinates of the upper boundary point and the lower boundary point, and the lateral orientation angle is calculated based on the three-dimensional coordinates of the left boundary point and the right boundary point. The longitudinal orientation angle and the lateral orientation angle are used to determine the pointing angle of the driving signal. If it is determined that no driving parameter corresponding to the animal type is found in the preset animal driving parameter library, the first driving is carried out using ultrasound with both the waveband and frequency changing from small to large based on the pointing angle. If the first attempt to drive away the animal fails, a second attempt to drive it away will be made using a preset laser color with varying intensity from weak to strong, corresponding to the type of animal, based on the pointing angle. Record the ultrasonic or laser parameters that cause the animal to leave the target area, and store them in the preset animal repelling parameter library corresponding to the animal type. If a driving parameter corresponding to the animal type is found in the preset animal driving parameter library, the animal is driven away according to the driving parameter.

2. The method according to claim 1, characterized in that, The step of identifying the animal type within the target area based on the video data and the depth data specifically includes: Extract moving targets from the video data and obtain the depth point cloud data corresponding to the moving targets; A three-dimensional feature model of the moving target is constructed based on the depth point cloud data; The three-dimensional feature model is matched with a preset animal feature database to obtain a preliminary identification result of the animal type; Based on the behavioral characteristics of the moving target and the preliminary identification results, the final animal type is determined.

3. The method according to claim 1, characterized in that, The calculation of the longitudinal orientation angle based on the three-dimensional coordinates of the upper and lower boundary points, and the calculation of the lateral orientation angle based on the three-dimensional coordinates of the left and right boundary points, specifically include: A first vector group is constructed based on the three-dimensional coordinates of the upper boundary point and the lower boundary point, and a second vector group is constructed based on the three-dimensional coordinates of the left boundary point and the right boundary point. Calculate the angle between adjacent vectors in the first vector group to obtain the longitudinal orientation angle; Calculate the angle between adjacent vectors in the second vector group to obtain the lateral orientation angle.

4. The method according to claim 1, characterized in that, After driving the animal according to the driving parameters, the method further includes: Read the historical driving records corresponding to the animal type. The historical driving records include driving parameter values, start time, end time, and number of drivings for each driving. Calculate the response time of the animal from receiving the driving signal to starting to move, and the departure time from starting to move to completely leaving the target area during the most recent preset first driving process; When the rate of change of the response time or the departure time exceeds a preset threshold for the second consecutive preset number, it is determined that the animal has adapted to the current driving parameters. Select at least two different sets of driving parameters from the preset animal driving parameter library, and randomly switch between the two different sets of driving parameters between two adjacent driving events; Obtain electronic map data of the target area, and based on the electronic map data, use a preset algorithm to calculate multiple feasible paths from the animal's current location to the boundary of the target area; Calculate the safety risk value for each of the feasible paths, the safety risk value being determined based on path length, slope variation, obstacle density, and distance to restricted areas; The feasible path with the lowest safety risk value is selected as the expected driving path, and the pointing angle of the driving signal is adjusted based on the expected driving path.

5. The method according to claim 4, characterized in that, The calculation of the security risk value for each feasible path specifically includes: Extract the distance and height difference between any two adjacent points on the feasible path, and calculate the path slope; The number and type of obstacles within a preset distance range around the feasible path are counted. Calculate the shortest distance between the feasible path and the boundary of the restricted area; Based on preset weighting coefficients, the path length, slope change, obstacle density, and distance to the restricted area are weighted and summed to obtain the safety risk value for each feasible path.

6. The method according to claim 4, characterized in that, After adjusting the pointing angle of the driving signal based on the expected driving path, the method further includes: The movement trajectory of the animal is detected in real time during the driving process to obtain the animal's real-time location and movement speed; Calculate the deviation distance between the animal's real-time location and the expected driving path; When the deviation distance exceeds a preset distance threshold, the feasible path is recalculated based on the animal's real-time location; Select the feasible path with the lowest recalculated security risk value as the updated expected deportation path. Adjust the pointing angle of the driving signal to align it with the direction of the updated expected driving path.

7. The method according to claim 6, characterized in that, The recalculation of the feasible path based on the animal's real-time location specifically includes: Based on electronic map data within a preset range around the animal's real-time location, the preset algorithm is used to calculate multiple new feasible paths from the animal's real-time location to the boundary of the target area; Calculate the security risk value for the new feasible path; The new feasible paths are sorted according to the security risk values. The new feasible path with the lowest security risk value is selected as the updated expected deportation path.

8. A system for intelligent identification and directional deterrence, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the system, the system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the method as described in any one of claims 1-7.

Images