Laser bird repelling method based on multi-modal sensing and dynamic path planning

Abstract

The invention provides a laser bird repelling method based on multi-modal sensing and dynamic path planning. According to the laser bird repelling method based on multi-modal sensing and dynamic path planning, the adaptability and long-term repelling efficiency of a wind power plant bird repelling system can be effectively improved, accurate and safe repelling of different birds is achieved through multi-modal sensing and dynamic path planning, and stable operation is kept under the complex environment condition.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bird prevention system for wind farm equipment, and particularly relates to a laser bird repelling method based on multi-modal perception and dynamic path planning. BACKGROUND

[0002] As a key node in the protection system of power facilities, wind farms are widely used in the field of renewable energy power generation. The main transformer and power transmission line of the wind farm are long-term exposed to the risk of short-circuit fault caused by bird habitat. In related technologies, various bird repelling methods are constructed through the collaborative work of bird prevention thorn, sound wave bird repeller and laser bird repeller. Specifically, this technical system covers the whole process from physical protection, acoustic interference to optical repulsion, including key links such as bird identification, path control and energy supply management. With the continuous expansion of the scale of wind farms and the increasing demand for intelligent operation and maintenance, the existing bird repelling system needs to be systematically optimized in terms of adaptability, safety and energy supply stability.

[0003] However, in the existing bird repelling method, fixed frequency sound waves or single laser irradiation strategy are directly used, and the deep integration of multi-modal perception and dynamic path planning is not realized, which may lead to rapid adaptation of birds, reduced repelling efficiency, or ecological and personal safety risks caused by laser misfire, thereby affecting the long-term effectiveness and operation reliability of the system. In addition, the traditional energy supply method is prone to power supply interruption under continuous rain or light wind conditions, causing bird repelling blind area and further increasing the equipment failure rate. Although the existing technology has improved in some local aspects, it still lacks intelligent identification of bird behavior characteristics and multi-technology collaborative response mechanism, and it is difficult to meet the continuous protection needs in complex environments. SUMMARY

[0004] The present application aims to at least partially solve one of the technical problems in the related art.

[0005] To this end, a first object of the present application is to propose a laser bird repelling method based on multi-modal perception and dynamic path planning.

[0006] A second object of the present application is to propose a laser bird repelling device based on multi-modal perception and dynamic path planning.

[0007] A third object of the present application is to propose an electronic device.

[0008] A fourth object of the present application is to propose a computer readable storage medium.

[0009] A fifth object of the present application is to propose a computer program product.

[0010] To achieve the above purpose, the first object of the present application is to provide a laser bird repelling method based on multi-modal perception and dynamic path planning, comprising:

[0011] S1, through the cooperative acquisition of the species, position and head posture information of birds by the distributed visual recognition unit and radar auxiliary unit, and in combination with the wind speed data collected by the environment perception unit, comprehensive perception data of bird behavior characteristics and environmental parameters is generated;

[0012] S2, based on the bird species and head posture information, an adaptive trajectory library is called to generate a corresponding dynamic laser irradiation path, the path forms a ring-shaped light spot with a radius of 30-50 cm with the bird landing point as the center, and dynamically avoids the eye area of the bird;

[0013] S3, according to the wind speed data, a laser offset compensation algorithm driven by wind speed feedback is used to calculate the laser offset compensation angle, and the compensation angle is superimposed into the dynamic laser irradiation path to realize the anti-disturbance optimization of laser irradiation;

[0014] S4, the pan-tilt of the laser emitting device is controlled to rotate horizontally by 360° and vertically by 360° according to the dynamic laser irradiation path, and laser meeting the eye safety standard is emitted to complete the bird repelling operation.

[0015] Optionally, the cooperative acquisition of the species, position and head posture information of birds by the distributed visual recognition unit and radar auxiliary unit, and in combination with the wind speed data collected by the environment perception unit, the generation of comprehensive perception data of bird behavior characteristics and environmental parameters, further comprises:

[0016] The visual recognition unit adopts a lightweight YOLO V8 model and is deployed on the top of the main transformer and near the insulator string to identify the bird species, position and head orientation angle θ in real time, and the recognition confidence is greater than 90%;

[0017] The radar auxiliary unit is deployed at the key tower of the booster station and the outgoing line, and uses a millimeter wave radar to detect the three-dimensional coordinates (X, Y, Z) of birds within a range of 200 m and is used to compensate for the blind area of visual recognition.

[0018] Optionally, based on the bird species and head posture information, an adaptive trajectory library is called to generate a corresponding dynamic laser irradiation path, the path forms a ring-shaped light spot with a radius of 30-50 cm with the bird landing point as the center, and dynamically avoids the eye area of the bird, further comprising:

[0019] The adaptive trajectory library contains a fast zigzag path for large-sized birds and a spiral diffusion path for small-sized birds;

[0020] The radius of the annular light spot is preferably 40cm, and the laser irradiation path is dynamically scanned along the horizontal and vertical planes with the bird's landing point as the center to ensure coverage of the repelling area.

[0021] Optionally, the step of calculating the laser offset compensation angle based on the wind speed data using a wind speed feedback-driven laser offset compensation algorithm, and superimposing the compensation angle onto the dynamic laser irradiation path to achieve anti-disturbance optimization of laser irradiation, further includes:

[0022] The laser offset compensation algorithm is expressed as Δθ=k·V wind +b, where Δθ is the laser offset compensation angle, V wind Let be the wind speed, and k and b be parameters determined through equipment calibration experiments. The calibration process includes testing the laser offset under different wind speed conditions and fitting the optimal parameter combination.

[0023] The wind speed data is collected in real time by the anemometer in the environmental sensing unit and filtered by the edge computing module to ensure the stability and accuracy of the input data.

[0024] Optional, also includes:

[0025] When the AI ​​image recognition unit detects that a person or vehicle has entered the 200-meter warning zone of the bird deterrent device, it triggers the dual-locking mechanism of the laser emitter via 4G / 5G, Bluetooth, or WiFi communication to immediately cut off the laser output and prevent accidental injury.

[0026] To achieve the above objectives, a second aspect of the present invention provides a laser bird deterrent device based on multimodal perception and dynamic path planning, comprising:

[0027] The integrated perception data generation module is used to acquire information on bird species, location, and head posture through the collaborative efforts of distributed visual recognition units and radar-assisted units, and to generate integrated perception data on bird behavior characteristics and environmental parameters by combining wind speed data collected by environmental perception units.

[0028] The dynamic path generation module is used to call the adaptive trajectory library to generate a corresponding dynamic laser illumination path based on the bird species and head posture information. The path forms a ring-shaped light spot with a radius of 30-50cm with the bird's landing point as the center, and dynamically avoids the bird's eye area.

[0029] The laser offset compensation module is used to calculate the laser offset compensation angle based on the wind speed data using a wind speed feedback-driven laser offset compensation algorithm, and to superimpose the compensation angle onto the dynamic laser irradiation path to achieve anti-disturbance optimization of laser irradiation.

[0030] The laser emission control module is used for controlling the horizontal 360-degree and vertical 360-degree rotation of the holder of the laser emission device according to the dynamic laser irradiation path, emitting laser light meeting the human eye safety standard, and completing the bird repelling operation.

[0031] To achieve the above object, the third aspect of the present application provides an electronic device, comprising: a processor, and a memory connected with the processor in communication;

[0032] The memory stores computer execution instructions;

[0033] The processor executes the computer execution instructions stored in the memory to realize the method according to any one of the first aspect.

[0034] To achieve the above object, the fourth aspect of the present application provides a computer readable storage medium, the computer readable storage medium stores computer execution instructions, and the computer execution instructions are used to realize the method according to any one of the first aspect when executed by a processor.

[0035] To achieve the above object, the fifth aspect of the present application provides a computer program product, the computer program is executed by a processor to realize the method according to any one of the first aspect.

[0036] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0037] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the accompanying drawings, wherein:

[0038] Figure 1 A flowchart of a laser bird repelling method based on multi-modal perception and dynamic path planning provided by an embodiment of the present application;

[0039] Figure 2 A structural schematic diagram of a laser bird repelling system based on multi-modal perception and dynamic path planning provided by an embodiment of the present application. DETAILED DESCRIPTION

[0040] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference signs represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application.

[0041] Figure 1A flowchart of a laser bird repelling method based on multi-modal perception and dynamic path planning is provided in the embodiments of the present application. The method is implemented by the laser bird repelling system shown in the drawings. Figure 2

[0042] As shown in the drawings, Figure 1 The method comprises the following steps:

[0043] S1, the species, location and head posture information of birds are obtained by the distributed visual recognition unit and radar auxiliary unit in cooperation, and the wind speed data collected by the environment perception unit is combined to generate comprehensive perception data of bird behavior characteristics and environmental parameters.

[0044] Specifically, this step involves the cooperative application of multi-modal perception technology, aiming to achieve high-precision identification and dynamic behavior modeling of birds in the wind farm area through the joint work of the distributed visual recognition unit, radar auxiliary unit and environment perception unit. In some implementations, the visual recognition unit uses a lightweight YOLO V8 model and is deployed on the top of the main transformer, insulator string and other key positions, has real-time recognition capability, and the recognition confidence is higher than 90%, which can accurately obtain the species, two-dimensional coordinates (U, V) and head posture angle θ (represented by pitch angle and yaw angle) of birds. The radar auxiliary unit uses a millimeter wave radar system and is deployed at the key tower of the booster station and the external transmission line, which has a detection range of 200 meters and can provide three-dimensional coordinates (X, Y, Z) and flight trajectory information, effectively compensating for the perception blind area of the visual unit in night, rain and fog or shielding environment.

[0045] Further, the environment perception unit integrates an anemometer, an irradiance meter and a rain and fog detector to collect environmental parameters such as wind speed, light intensity and humidity in real time. In actual application, this step is suitable for key equipment areas such as wind farm main transformers, transmission line insulators and other areas susceptible to bird disturbance. Through multi-source data fusion, the system can construct a comprehensive perception data set of bird behavior characteristics and environmental parameters, providing accurate input for subsequent dynamic path planning and laser bird repelling strategies. Its technical value lies in significantly improving the accuracy and real-time performance of bird identification, while enhancing the environmental adaptability, laying a data foundation for safe, efficient and intelligent bird repelling control.

[0046] Further, S1 comprises:

[0047] S11, the visual recognition unit uses a lightweight YOLO V8 model and is deployed on the top of the main transformer and near the insulator string, which can identify the species, location and head orientation angle θ of birds in real time, and the recognition confidence is greater than 90%.

[0048] ​Specifically, in some implementations, the visual recognition unit employs a lightweight YOLO V8 model, which is deployed on the top of the main transformer and near the insulator string, to identify the bird species, position, and head orientation angle θ in real time with an identification confidence greater than 90%. This step is based on computer vision and edge computing technology, aiming to achieve high-precision, low-latency detection and behavior analysis of birds, providing key inputs for subsequent laser bird repelling path planning.

[0049] In terms of technical implementation, the YOLO V8 model in the system is subjected to model pruning and quantization processing, with the model volume compressed to less than 30% of the original model using INT8 quantization, and the inference speed increased to more than 20FPS, meeting the real-time processing needs of edge devices under resource constraints. The camera uses an industrial-grade high-definition infrared lens with a resolution of 1920x1080 and a frame rate of 30FPS, supporting bird identification in low-light environments. The model is deployed on an edge computing module (such as the NVIDIA Jetson series of embedded platforms), reducing data transmission delay and improving response speed through local inference.

[0050] In terms of parameter indicators, the identification confidence threshold is set to 90%, meaning that the bird detection results output by the model must satisfy IoU (intersection over union) ≥ 0.5 and confidence ≥ 0.9 to be adopted by the system. The calculation of the head orientation angle θ is based on bird head key point detection, obtaining the azimuth angle relative to the laser emitting device through geometric vector analysis, with an accuracy of within ±3°. The identification range covers the bird activity area from 10m to 50m within the camera field of view, meeting the monitoring needs of the main transformer and insulator string.

[0051] In terms of application scenarios, this visual recognition unit is deployed on the top of the main transformer and near the insulator string in the wind farm, effectively monitoring the bird landing behavior on the equipment, especially suitable for identifying and tracking common birds such as magpies, crows, and sparrows. In the night or low-light conditions, the infrared imaging function can ensure continuous identification and avoid missed detection due to insufficient light.

[0052] In terms of technical effects, this step realizes real-time, high-precision identification and posture estimation of birds, providing reliable bird behavior data for the laser bird repelling device, ensuring that the laser irradiation path avoids the bird eye area, meeting the IEC 60825 human eye safety standard. At the same time, through lightweight model design, the system power consumption and hardware cost are reduced, and the deployment flexibility and running stability of the overall system are improved.

[0053] S12, the radar auxiliary unit is deployed at the key tower of the booster station and the outgoing line, using a millimeter wave radar to detect the three-dimensional coordinates (X, Y, Z) of birds within a range of 200m, and to compensate for the blind area of visual recognition.

[0054] Specifically, the radar auxiliary unit is deployed at the booster station and the key tower of the outgoing line, and uses millimeter wave radar technology to detect the three-dimensional coordinates (X, Y, Z) of birds within a range of 200 m in real time. Its core role is to compensate for the blind area problem of the visual recognition unit in complex environments, and to improve the spatial perception accuracy and coverage capability of the bird repelling system. In some implementations, the radar unit can use a 77GHz frequency band FMCW (Frequency Modulated Continuous Wave) millimeter wave radar, which has a detection distance of more than 200 m, an angle resolution better than 1°, a ranging accuracy of ±0.1 m, and a speed accuracy of ±0.1 m / s, meeting the high-precision bird trajectory tracking requirements in complex terrain and sheltered environments of wind farms. The radar transmits continuous frequency modulation signals and receives echoes, and uses time-frequency analysis technology to extract the motion characteristics of birds, including flight height, speed vector, and trajectory change trend, thereby constructing a dynamic model of birds in three-dimensional space.

[0055] Further, the radar auxiliary unit and the visual recognition unit form a multi-modal perception fusion mechanism. The visual recognition unit classifies and recognizes the posture of birds based on the YOLO V8 model, but is limited by obstructions, changes in lighting, and the installation angle of the device, and has certain recognition blind spots. The radar unit provides continuous spatial coordinate information through non-visual means, especially in areas such as night, rain, fog, or behind the device where vision is difficult to cover, which can effectively supplement the bird position data. At the data fusion level, the system can use Kalman filtering algorithm to align the visual and radar data in time and space and predict the trajectory, improving the robustness and real-time performance of bird positioning.

[0056] This step is deployed in high-risk areas such as the top of the booster station transformer and the key tower of the outgoing line in actual application. The radar unit is usually installed on a gimbal or fixed bracket, with a pitch adjustment capability of ±15° to adapt to bird activity at different altitudes. Its technical value lies in significantly improving the spatial perception capability of the bird repelling system, providing accurate bird position information for subsequent laser path planning, thereby achieving safe and efficient repelling effect, avoiding laser misfiring at bird eyes or injuring personnel, and ensuring the safe operation of wind farm equipment and personnel.

[0057] S2, based on the bird species and head posture information, calling an adaptive trajectory library to generate a corresponding dynamic laser irradiation path, the path forming a ring-shaped light spot with a radius of 30-50 cm with the bird landing point as the center, and dynamically avoiding the bird eye area.

[0058] Specifically, this step involves dynamic laser path planning based on bird species identification and head posture estimation. Its core is to call an adaptive trajectory library to generate a ring-shaped light spot path with a radius of 30-50 cm (preferably 40 cm) with the bird landing point as the center, and dynamically avoid the bird eye area through posture information, thereby achieving efficient and safe bird repelling effect.

[0059] Further, S2 comprises:

[0060] S21, the adaptive trajectory library contains a fast zigzag path for large birds and a spiral diffusion path for small birds.

[0061] Specifically, in the dynamic laser path planning algorithm of the present application, the construction of the adaptive trajectory library is the core link to achieve efficient and safe bird repelling. According to the size difference of birds, the trajectory library presets two typical irradiation paths: a fast zigzag path for large birds (such as white heron, grey heron, etc.), and a spiral diffusion path for small birds (such as sparrow, magpie, etc.). In some implementations, the fast zigzag path scans the bird activity area with a high frequency and small amplitude zigzag trajectory of the laser beam, with a scanning frequency usually between 10-20Hz, and a path offset angle controlled within ±15°, ensuring that the laser covers multiple landing points of the bird in a short time, thereby enhancing its repelling effect. This path is suitable for large birds with relatively slow movements, which stimulates their visual system through rapid changes in the light spot and breaks their adaptability to fixed light sources.

[0062] For small birds, the spiral diffusion path adopts a low-frequency and large-range scanning strategy, usually taking the bird landing point as the center to generate a circular scanning trajectory with a radius of 30-50cm, and the laser beam spreads outward along the spiral path at a frequency of 0.5-1.5Hz, covering its possible activity range. This path design takes into account the agility and short-distance flight characteristics of small birds, and improves their escape probability through continuous but non-point light spot disturbance.

[0063] Further, the calling of the adaptive trajectory library depends on the bird classification information provided by the multi-modal perception module, combined with the recognition result (confidence > 90%) of the YOLO V8 model, the system can accurately match the repelling strategy. This step is particularly suitable for key equipment areas such as wind farm main transformers and insulator strings in actual application, significantly improving the bird repelling efficiency through dynamic path planning, while reducing the risk of injury, and achieving intelligent response to different bird behavior characteristics.

[0064] S22, the radius of the ring-shaped light spot is preferably 40cm, and the laser irradiation path takes the bird landing point as the center and dynamically scans along the horizontal and vertical planes to ensure coverage of the repelling area.

[0065] Specifically, in some implementations, the radius of the annular light spot is preferably 40 cm, and the laser irradiation path is dynamically scanned with the bird landing point as the center, along the horizontal plane and the vertical plane, and the technical implementation principle is based on the optimized matching of bird behavior characteristics and laser repelling effect. This step realizes efficient and safe coverage of the bird landing area through the laser gimbal control system combined with bird posture recognition and path planning algorithm. In the specific operation mode, the system first identifies the bird landing point coordinates (x, y, z) through the YOLO V8 model, and determines the accurate position of the bird in space combined with the three-dimensional positioning data of the millimeter wave radar. Subsequently, the path planning algorithm generates an annular scanning trajectory with a radius of 40 cm with the landing point as the center. The radius is verified by experiments and can effectively cover the activity radius of the bird, while avoiding direct laser irradiation of the eye area. During scanning in the horizontal plane and the vertical plane, the gimbal driver controls the laser emitter to continuously rotate at a step angle of 0.5°-2°, and the scanning frequency is adjustable, ranging from 1 to 5 Hz, to adapt to the reaction speed of different birds. Further, this step introduces a wind speed feedback mechanism, which offsets the compensation through the formula Δθ=k·V wind +b, to ensure the stability and accuracy of the laser path in a wind disturbance environment. This step meets the radiation safety requirements for human eye safety laser equipment in IEC 60825-1:2020, and the laser wavelength is preferably 1550 nm, which has good atmospheric penetration and lower biological tissue damage risk. In actual application scenarios, this step is widely applicable to bird repelling protection of key equipment such as wind farm main transformers and transmission line insulators, especially in areas such as tower platforms and equipment supports where birds frequently inhabit, which can effectively improve the repelling efficiency and reduce the risk of injury. Through this dynamic scanning strategy, the system can achieve continuous bird repelling in complex environmental conditions while ensuring precise control of the laser irradiation path and human eye safety, which is an important technical support for the invention in improving bird repelling intelligence and safety.

[0066] S3, according to the wind speed data, using a laser offset compensation algorithm driven by wind speed feedback, calculating a laser offset compensation angle, and superimposing the compensation angle into the dynamic laser irradiation path to realize anti-disturbance optimization of laser irradiation.

[0067] Specifically, this step involves a laser offset compensation algorithm based on wind speed feedback, which dynamically adjusts the offset angle of the laser irradiation path through real-time wind speed data to improve the anti-disturbance ability and irradiation accuracy of the laser bird repelling system. In some implementations, the algorithm uses a linear model Δθ=k·V wind +b, where Δθ is the laser offset compensation angle, V windwhere V is the wind speed, k and b are parameters determined through device calibration experiments, and the calibration process includes testing the laser offset under different wind speed conditions and fitting to obtain the optimal parameter combination. This model is based on aerodynamic principles and considers the influence of wind speed on the structural stability of the laser emitting device and the propagation path of the light beam, thereby achieving dynamic correction of the laser irradiation point under wind disturbance.

[0068] Specifically, in the dynamic laser path planning algorithm of the present application, the parameters k and b are key compensation coefficients determined through device calibration experiments, which are used to model the linear relationship between laser offset and wind speed. The core technical principle of this step is to collect actual offset angle data of the laser emitting device under different wind speed conditions, establish a mathematical model between wind speed and laser offset angle, and thereby correct the laser irradiation path in real time during actual operation, improving the environmental adaptability and safety of the bird repelling system.

[0069] In specific implementation, the calibration process is usually carried out in a laboratory or a simulated environment of a wind farm. First, a test platform with controllable wind speed is set up, using an industrial-grade wind speed generating device (such as an axial flow fan or a wind tunnel system), setting multiple wind speed gradients (such as 0 m / s, 3 m / s, 6 m / s, 9 m / s, 12 m / s), and under each wind speed condition, recording the actual offset angle Δθ of the laser emitting device under the action of wind through a high-precision angle sensor (accuracy ±0.1°). At the same time, a laser positioning system (such as a laser interferometer or a high-resolution CCD camera) is used to measure the landing point of the laser spot, ensuring the accuracy of the offset data.

[0070] In some implementations, the calibration data is linearly fitted by the least squares method to obtain the optimal parameter combination k and b, so that the fitting error of the formula Δθ = k·V wind +b is minimized. For example, in a calibration experiment with a wind speed range of 0-12 m / s, k ≈ 0.27° / (m / s) and b ≈ 0.5° can be obtained, so that when the wind speed is 8 m / s, the laser offset compensation angle is calculated as Δθ = 0.27×8 + 0.5 = 3.2°. This parameter combination can be stored in the system control unit for real-time calling by the dynamic path planning algorithm.

[0071] This step has significant technical value in actual application, especially in complex meteorological conditions of a wind farm. Through the wind speed feedback mechanism, the system can dynamically correct the laser irradiation path to avoid spot offset caused by wind disturbance, thereby ensuring that the laser always acts on the landing area of birds rather than sensitive parts, improving the repelling efficiency and reducing the risk of injury. In addition, this calibration method complies with the industry standard for laser safety control (such as IEC 60825), providing reliable data support for a laser bird repelling system with high precision and high safety.

[0072] In addition, wind speed data is collected in real time by a wind speed meter in the environmental perception unit and filtered by the edge computing module to ensure the stability and accuracy of the input data. This step is a key link to realize dynamic compensation of the laser bird repelling path. The technical implementation of this step is based on a distributed sensing architecture and an edge computing cooperative processing mechanism, aiming to improve the collection accuracy of environmental parameters, thereby providing reliable input basis for laser path planning.

[0073] In some implementations, the environmental perception unit is deployed on the laser bird repelling device support near the key equipment of the wind farm, such as the main transformer and the power transmission tower, and integrates a high-precision anemometer. The sampling frequency is usually set to 10Hz to 20Hz to meet the real-time requirement. The anemometer adopts a three-cup or ultrasonic structure, with a measurement range of 0-30m / s, an accuracy error of ≤±2%, and support for IP67 protection level to ensure stable operation under complex weather conditions such as high wind speed, rain and fog. The collected raw wind speed data is transmitted to the edge computing module through RS-485 or CAN bus. The module usually carries an embedded processor such as NXP i.MX 8 series or Rockchip RK3568, runs a lightweight Linux system, and supports real-time data processing and multi-sensor fusion.

[0074] Further, the edge computing module performs digital filtering on the wind speed data to eliminate transient disturbances and sensor noise. Optionally, sliding average filtering or Kalman filtering algorithm is used, where the sliding average window length is set to 5-10 sampling points, and Kalman filtering is dynamically adjusted according to the system noise and observation noise covariance matrix to ensure the smoothness and stability of the output wind speed value. The filtered wind speed data is used as a key input parameter in the dynamic path planning algorithm to calculate the laser offset compensation angle.

[0075] This step is particularly suitable for wind farms in coastal or highland areas where wind speed changes dramatically. Through real-time wind speed feedback, the system can dynamically correct the laser emission angle to avoid beam deviation caused by wind disturbance, thereby improving the accuracy and safety of the bird repelling path. In addition, this mechanism effectively reduces the risk of misfiring caused by environmental factors, enhancing the robustness and adaptability of the laser bird repelling system. It is a basic guarantee for the cooperative work of the human eye safety double-locking mechanism and the dynamic path planning algorithm.

[0076] S4, the gimbal of the laser emission device rotates horizontally 360° and vertically 360° according to the dynamic laser irradiation path, emits laser that meets the human eye safety standard, and completes the bird repelling operation.

[0077] Specifically, this step involves controlling the gimbal of the laser emitting device to rotate 360° horizontally and 360° vertically according to the dynamic laser irradiation path to achieve safe and efficient bird repelling. This operation is based on the real-time position, posture, and environmental parameters of the birds provided by the multi-modal perception module, combined with a dynamic path planning algorithm, to ensure that the laser irradiation path has both wide coverage and meets the requirements of safety and repelling efficiency.

[0078] In terms of technical implementation, the gimbal uses a dual-axis servo motor driving structure to control the rotation in the horizontal (azimuth angle) and vertical (elevation angle) directions, supporting full-angle scanning. The laser emitting device is installed at the end of the gimbal, and the angle information is fed back through a high-precision encoder to realize closed-loop control. The gimbal control system receives instructions from the edge computing module, including the three-dimensional coordinates of the target bird, the head posture angle θ, and the environmental compensation angle Δθ. For example, when the wind speed is 8 m / s, Δθ can be calculated as 3.2° to correct the offset of the laser under wind disturbance.

[0079] In terms of parameter indicators, the laser emitting device uses a 1550 nm wavelength laser, which meets the Class 1 eye safety standard specified in IEC 60825-1:2011, ensuring strong atmospheric penetration while not causing direct harm to birds or humans. The gimbal rotation speed is adjustable, with a horizontal rotation speed range of 0.1°-360° / s and a vertical rotation speed of 0.05°-180° / s to adapt to the reaction speed and flight trajectory of different birds. The laser spot diameter is maintained between 30-50 cm in the target area, preferably 40 cm, to cover the bird's landing point and avoid excessive concentrated irradiation.

[0080] In application scenarios, this step is suitable for bird protection of key equipment such as wind farm main transformers and transmission line insulator strings. When birds stay or prepare to land near the equipment, the system identifies their species and posture through the YOLO V8 model and generates a dynamic irradiation path combined with millimeter wave radar data. The gimbal continuously rotates according to the path instructions, and the laser forms a dynamic spot in the target area, stimulating the bird's visual nerves and prompting it to actively move away from the equipment area.

[0081] The technical effect of this step is to achieve dynamic coverage of the laser irradiation path through the omnidirectional rotating gimbal, combined with the eye-avoidance strategy and environmental compensation algorithm, significantly improving the repelling efficiency and reducing the risk of injury. At the same time, its closed-loop control mechanism ensures the accuracy and stability of laser irradiation, providing all-weather and intelligent bird protection capabilities for wind farms.

[0082] In addition, in the embodiment of the present application, when the AI image recognition unit detects that personnel or vehicles enter the 200-meter warning zone of the bird repelling device, it triggers the double-locking mechanism of the laser emitting device through 4G / 5G, Bluetooth, or WiFi communication methods, immediately cutting off the laser output to prevent injury.

[0083] Specifically, this step involves a human eye safety double-locking mechanism in the laser bird repelling device, which is based on the cooperation of the AI image recognition unit and the communication module to achieve real-time response to personnel or vehicle entering the 200-meter warning zone and immediate shutdown of laser output, thereby effectively avoiding the risk of accidental injury. In some implementations, the AI image recognition unit uses a lightweight target detection model based on YOLO V8, deployed on an edge computing node, with real-time recognition capability, recognition confidence higher than 90%, and recognition delay controlled within 50ms. The recognized targets include but are not limited to moving objects such as humans and vehicles, and the detection range covers the area within the 200-meter radius of the bird repelling device. Through the use of wide-angle cameras and multi-angle deployment strategies, visual blind spots are eliminated.

[0084] When the recognition unit detects personnel or vehicles entering the warning zone, the system immediately starts the communication trigger mechanism. Optionally, the communication methods include 4G / 5G wireless communication, Bluetooth, or WiFi, among which 4G / 5G communication is suitable for stable connection in long-distance and complex terrain, supporting low latency (<100ms) and high reliability (error rate <10^-6); Bluetooth and WiFi are suitable for short-distance and low-power consumption scenarios, with faster response speed (<50ms). Further, a bidirectional data link is established between the communication module and the laser emission device to ensure the immediate issuance of instructions and the synchronization of state feedback.

[0085] After receiving the locking instruction, the laser emission device immediately executes the double-locking mechanism: on the one hand, it cuts off the power supply circuit of the laser through the control circuit; on the other hand, it locks the laser emission direction through the pan-tilt control module to prevent temporary misfiring due to mechanical inertia. This mechanism meets the requirements of IEC 60825-1:2014 for human eye safety laser equipment, ensuring that the laser output can be completely stopped within 100ms in emergency situations.

[0086] This step plays a key role in the overall system safety, especially in areas with frequent personnel activities such as wind farms, effectively preventing laser injuries to inspection personnel or nearby residents, improving the safety level and deployability of the system, and providing important technical support for intelligent, safe, and efficient bird repelling.

[0087] The laser bird repelling method based on multi-modal perception and dynamic path planning of the embodiments of the present application, when detecting personnel or vehicles entering the 200-meter warning zone of the bird repelling device, the system triggers the double-locking mechanism of the laser emission device in real time through multi-channel communication, further improves the human eye safety protection level, effectively avoids the risk of laser misirradiation, and significantly enhances the safety and reliability of the system in complex human-machine interaction environments while ensuring the efficiency of bird repelling.

[0088] To achieve the above-embodiment, the application further provides a laser bird repelling device based on multi-modal perception and dynamic path planning.

[0089] The comprehensive perception data generation module is configured to acquire the species, position and head posture information of the birds through the distributed visual recognition unit and radar auxiliary unit, and generate comprehensive perception data of the bird behavior characteristics and environmental parameters in combination with the wind speed data collected by the environmental perception unit.

[0090] The dynamic path generation module is configured to call an adaptive trajectory library based on the bird species and head posture information, and generate a corresponding dynamic laser irradiation path, wherein the path forms a ring-shaped light spot with a radius of 30-50 cm with the bird landing point as the center, and dynamically avoids the bird eye area.

[0091] The laser offset compensation module is configured to calculate a laser offset compensation angle by using a wind speed feedback driven laser offset compensation algorithm according to the wind speed data, and superimpose the compensation angle into the dynamic laser irradiation path to realize the anti-disturbance optimization of laser irradiation.

[0092] The laser emission control module is configured to control the gimbal of the laser emission device to rotate horizontally by 360° and vertically by 360° according to the dynamic laser irradiation path, emit laser light meeting the human eye safety standard, and complete the bird repelling operation.

[0093] As to the device in the above-embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be described in detail here.

[0094] To achieve the above-embodiment, the application further provides an electronic device, which comprises a processor and a memory connected with the processor in communication; the memory stores computer execution instructions; and the processor executes the computer execution instructions stored in the memory to realize the method provided in the above-embodiment.

[0095] To achieve the above-embodiment, the application further provides a computer readable storage medium, which stores computer execution instructions; and the computer execution instructions are executed by the processor to realize the method provided in the above-embodiment.

[0096] To achieve the above-embodiment, the application further provides a computer program product, which comprises a computer program; and the computer program is executed by the processor to realize the method provided in the above-embodiment.

[0097] The collection, storage, use, processing, transmission, provision and disclosure of the user personal information involved in the application comply with the relevant laws and regulations, and do not violate public order and good customs.

[0098] It is important to note that user's personal information shall be collected for legitimate and reasonable uses of the service and not shared or sold outside of those legitimate uses. Further, such collection / sharing shall occur after the user's informed consent has been given which includes, but is not limited to, notifying the user, reading the user agreement / user notification, and signing an agreement / authorization that includes authorization of relevant user information, prior to the user's use of the functionality. Further, any necessary steps shall be taken to safeguard and secure access to such personal information data and ensure that others with access to the personal information data adhere to their privacy policies and procedures.

[0099] The present application contemplates providing an implementation in which the user can elect to opt in or opt out of allowing the use or access of personal information data. That is, the present disclosure contemplates providing the user with control to determine whether, or the extent to which, mined personal information data can be surrendered to a service or application. In one implementation, the user can be provided with options in response to which the user can opt in or opt out of allowing the use or access of personal information data. In another implementation, the user can be provided with options in response to which the user can select to allow or not allow the use or access of personal information data. In yet another implementation, the user can be provided with options in response to which the user can elect to enable, disable, or limit the use or access of personal information data.

[0100] In the foregoing various embodiments described, reference has been made to descriptive terms such as "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" etc. which are intended to indicate that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. The appearances of such phrases in various places in the specification are not necessarily intended to refer to the same embodiment or example. Further, when such phrases as "in one embodiment" or "in some embodiments" are used, there is no intention to suggest that a feature, structure, material, or characteristic defined in one embodiment is recommended to be included in all embodiments of the application. In addition, whether or not a specific feature, structure, material or characteristic is described in conjunction with an embodiment or example, it is intended to be possible to include, in some way, such feature, structure, material or characteristic in at least one embodiment or example of the application. Furthermore, the different embodiments and examples described in this specification can be combined with each other, where appropriate, in order to produce further embodiments and examples, and the features, structures, materials or characteristics of the different embodiments and examples can be combined with each other, where appropriate, in order to produce further embodiments and examples.

[0101] In addition, the terms "first", "second", etc. are used herein only to describe various features, and do not imply a relative importance or a specific order of precedence. Thus, a feature defined with a "first", "second", etc. can implicitly or explicitly include at least one of the feature. In the description of the application, the meaning of "a plurality" is at least two, for example, two, three, etc., unless otherwise specifically defined.

[0102] Any processes or methods described in the flow charts or otherwise described herein represent embodiments which can be employed, and in which the steps or actions can be performed in the order shown, in a different order than illustrated, or concurrently. Accordingly, the scope of the present application is not intended to be limited to the particular embodiments described in the specification. In addition, any individual elements of the specific implementation or use of those elements can also not be intended to be a required characteristic. Rather, claims can cover all grouts implementing the features or combinations of features presented in the specification or any combination of consequences or features whether or not the combination is explicitly presented in the specification.

[0103] The logic and / or steps represented in flow diagrams or otherwise described herein, for example, can be considered as a sequence of instructions to implement logic functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, processor- containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. In the context of this specification, a "computer-readable medium" can be any means that can contain, store, communicate, propagate or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a machine-readable storage device (e.g., magnetic, optical or other) a machine-readable storage diskette (e.g., floppy disk, optical disk, CD- ROM, etc.), a machine- readable storage medium (e.g., ROM, EPROM, etc.), or a machine-readable signal (e.g., electrical, optical, acoustical or other). Particular examples (a non-exhaustive list) of the computer- readable medium include the following: a connection (e.g., electrical) that is wired, wireless, optical, acoustical or other that has the one or more wires; a portable computer diskette (magnetic, optical or other); a RAM (random access memory); a ROM (read only memory); an EPROM (erasable programmable ROM); an EEPROM (electrically erasable programmable ROM); and a portable compact disc read only memory (CD-ROM). Note that the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example via the optical scanner of a device or device or via acoustical scanning, then electronically captured, interpreted or processed in a suitable manner if necessary, and then stored in a computer memory.

[0104] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, various steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. As such, in some embodiments, the steps or methods can be implemented in a combination of hardware and software. If implemented in hardware, as in another embodiment, any of the following technologies, or combinations thereof, can be used: a discrete logic circuit having logic gates for implementing logic functions upon data signals, an application specific integrated circuit having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0105] Those skilled in the art can understand that all or part of the steps carried out by the above-mentioned embodiment method can be completed by programs instructing related hardware, and the programs can be stored in a computer-readable storage medium. When the programs are executed, one or a combination of the steps of the method embodiment is included.

[0106] In addition, each function unit in each embodiment of the present application can be integrated in one processing module, or each unit can be physically present separately, or two or more units can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module. When the integrated module is realized in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0107] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

[0108] It should be understood that various forms of flow shown above can be used to reorder, add or delete steps. For example, each step described in the present application can be executed in parallel, sequentially or in different order, as long as the desired results of the technical solutions of the present application can be achieved, which is not limited herein.

[0109] The above detailed description does not constitute a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and replacements can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A laser bird deterrence method based on multimodal perception and dynamic path planning, characterized in that, include: S1, through the collaborative acquisition of bird species, location and head posture information by distributed visual recognition units and radar-assisted units, and combined with wind speed data collected by environmental perception units, generates comprehensive perception data of bird behavior characteristics and environmental parameters. S2, based on the bird species and head posture information, call the adaptive trajectory library to generate a corresponding dynamic laser illumination path. The path forms a ring-shaped light spot with a radius of 30-50cm centered on the bird's landing point, and dynamically avoids the bird's eye area. S3. Based on the wind speed data, the laser offset compensation angle is calculated using a wind speed feedback-driven laser offset compensation algorithm, and the compensation angle is superimposed on the dynamic laser irradiation path to achieve anti-disturbance optimization of laser irradiation. S4, control the gimbal of the laser emitting device to rotate 360° horizontally and 360° vertically according to the dynamic laser irradiation path, emit laser that meets the safety standards for human eyes, and complete the operation of driving away the birds.

2. The method as described in claim 1, characterized in that, The method of acquiring bird species, location, and head posture information through the collaborative acquisition of distributed visual recognition units and radar-assisted units, and combining this with wind speed data collected by environmental perception units to generate comprehensive perception data of bird behavioral characteristics and environmental parameters, also includes: The visual recognition unit uses a lightweight YOLO V8 model and is deployed on the top of the main transformer and near the insulator string. It can identify the species, location and head orientation angle θ of birds in real time, with a recognition confidence level of more than 90%. The radar auxiliary unit is deployed at key towers of the booster station and transmission line. It uses millimeter-wave radar to detect the three-dimensional coordinates (X,Y,Z) of birds within a 200m range and is used to compensate for blind spots in visual recognition.

3. The method as described in claim 1, characterized in that, Based on the bird species and head posture information, an adaptive trajectory library is invoked to generate a corresponding dynamic laser illumination path. This path forms a ring-shaped light spot with a radius of 30-50 cm centered on the bird's landing point, dynamically avoiding the bird's eye area. The method also includes: The adaptive trajectory library includes fast zigzag paths for large birds and spiral diffusion paths for small birds. The radius of the annular light spot is preferably 40cm, and the laser irradiation path is dynamically scanned along the horizontal and vertical planes with the bird's landing point as the center to ensure coverage of the repelling area.

4. The method as described in claim 1, characterized in that, The step of calculating the laser offset compensation angle based on the wind speed data using a wind speed feedback-driven laser offset compensation algorithm, and superimposing the compensation angle onto the dynamic laser irradiation path to achieve anti-disturbance optimization of laser irradiation, further includes: The laser offset compensation algorithm is expressed as Δθ=k·V wind +b, where Δθ is the laser offset compensation angle, V wind Let be the wind speed, and k and b be parameters determined through equipment calibration experiments. The calibration process includes testing the laser offset under different wind speed conditions and fitting the optimal parameter combination. The wind speed data is collected in real time by the anemometer in the environmental sensing unit and filtered by the edge computing module to ensure the stability and accuracy of the input data.

5. The method as described in claim 1, characterized in that, Also includes: When the AI ​​image recognition unit detects that a person or vehicle has entered the 200-meter warning zone of the bird deterrent device, it triggers the double-locking mechanism of the laser emitter via 4G / 5G, Bluetooth, or WiFi communication to immediately cut off the laser output and prevent accidental injury.

6. A laser bird deterrent device based on multimodal perception and dynamic path planning, characterized in that, include: The integrated perception data generation module is used to acquire information on bird species, location, and head posture through the collaborative efforts of distributed visual recognition units and radar-assisted units, and to generate integrated perception data on bird behavior characteristics and environmental parameters by combining wind speed data collected by environmental perception units. The dynamic path generation module is used to call the adaptive trajectory library to generate a corresponding dynamic laser illumination path based on the bird species and head posture information. The path forms a ring-shaped light spot with a radius of 30-50cm with the bird's landing point as the center, and dynamically avoids the bird's eye area. The laser offset compensation module is used to calculate the laser offset compensation angle based on the wind speed data using a wind speed feedback-driven laser offset compensation algorithm, and to superimpose the compensation angle onto the dynamic laser irradiation path to achieve anti-disturbance optimization of laser irradiation. The laser emission control module is used to control the gimbal of the laser emission device to rotate 360° horizontally and 360° vertically according to the dynamic laser irradiation path, emitting lasers that meet human eye safety standards to complete the operation of driving away birds.

7. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-5.

9. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1-5.

Images