Wild animal intelligent monitoring equipment data transmission method based on unmanned aerial vehicle cooperation

Abstract

The invention discloses a wild animal intelligent monitoring equipment data transmission method based on unmanned aerial vehicle cooperation, and relates to the field of wild animal intelligent monitoring, and the method comprises the steps: carrying out the data collection and local storage through intelligent monitoring equipment; routing inspection path planning and preparation are carried out on the unmanned aerial vehicle; data interaction transmission is performed through the unmanned aerial vehicle and the intelligent monitoring equipment; and the data acquired by the unmanned aerial vehicle is transmitted back and analyzed at the cloud. The problems of low data transmission efficiency, high cost, poor safety and the like caused by the fact that a traditional intelligent monitoring device depends on manual card taking are solved, and an efficient closed loop of collection, transmission and analysis of wild animal monitoring data is achieved.

Description

Technical Field

[0001] This invention relates to the field of intelligent wildlife monitoring, and specifically to a data transmission method for intelligent wildlife monitoring equipment based on drone collaboration. Background Technology

[0002] In wildlife monitoring scenarios such as national nature reserves, intelligent monitoring equipment is the core data acquisition device, but existing technical solutions have the following significant drawbacks: The data transmission cycle is long and passive: intelligent monitoring equipment can only store image data locally, and manual personnel need to go deep into the protected area regularly (usually once a month) to retrieve the cards. The data cycle from collection to uploading to the analysis platform can take several weeks, making it impossible to capture key information such as the activity trajectory of rare species and abnormal behavior (such as injury or poaching interference) in a timely manner, thus missing the window for ecological protection decision-making. High labor costs and prominent safety risks: The protected areas are mostly complex terrains such as mountains, dense forests, and wetlands. Manual card retrieval requires a lot of manpower and resources. Staff also face safety hazards such as getting lost, being attacked by wild animals, and severe weather (heavy rain, heavy snow). Especially at remote monitoring points, the difficulty and risk of card retrieval increase exponentially. Low data management efficiency: The data stored by intelligent monitoring devices contains a large amount of invalid content (such as falsely triggered images with only background shaking or blank screens). After manual card retrieval, valid data must be screened one by one, which further prolongs the analysis cycle and makes it easy to miss high-value images due to human operation errors. Lack of equipment synergy: There is no linkage mechanism between intelligent monitoring equipment and external data transmission carriers. It is impossible to use mobile devices to overcome the limitation of "physical card removal", resulting in the break of the monitoring link. It is impossible to form a real-time and automated data transmission system, which is difficult to adapt to the dynamic needs of modern ecological monitoring. Summary of the Invention

[0003] To address the aforementioned shortcomings in existing technologies, this invention provides a data transmission method for intelligent wildlife monitoring equipment based on drone collaboration. This method solves the problems of inefficient data transmission, high cost, and poor security caused by the reliance on manual card retrieval in traditional intelligent monitoring equipment, and achieves an efficient closed loop of "collection-transmission-analysis" for wildlife monitoring data.

[0004] To achieve the aforementioned objectives, the present invention employs the following technical solution: a data transmission method for intelligent wildlife monitoring equipment based on unmanned aerial vehicle (UAV) collaboration, comprising the following steps: S1: Data collection and local storage are performed using intelligent monitoring equipment; S2: Plan and prepare inspection routes for drones; S3: Data exchange and transmission between drones and intelligent monitoring equipment; S4: Transmit the data acquired by the drone back to the cloud for analysis.

[0005] Further, S1 includes: S11: The intelligent monitoring equipment is deployed in the target monitoring area of ​​the protected area. The equipment adopts a dual triggering mechanism of thermal sensing and motion. When the animal's body temperature is detected to be >25℃ or the image pixel change rate is >5%, it will automatically start shooting and generate a 1080P resolution color or infrared image. It will also automatically switch to infrared mode at night. S12: The intelligent monitoring device has a built-in industrial-grade storage unit of at least 128GB. It names the captured images in the format of "capture timestamp-device number-serial number". At the same time, it records the ambient temperature and light intensity at the time of capture through integrated temperature and light sensors and stores them together with the images. S13: The intelligent monitoring device supports data tagging function, automatically classifies continuously captured images of the same target, and sets up a storage protection mechanism. When the remaining power is <10%, it prioritizes the retention of stored data and stops capturing new images.

[0006] Further, S2 includes: S21: The cloud management platform pre-imports the GPS location information of all intelligent monitoring devices in the protected area, and combines the drone parameters with the terrain data of the protected area to generate the optimal inspection route; S22: The drone is equipped with a dedicated communication module, at least 128GB of high-speed storage module and GPS navigation module. Before flight, the optimal inspection route, target intelligent monitoring equipment list and encryption key are loaded through the cloud management platform.

[0007] Furthermore, the generation of the optimal inspection path includes: S211: Construct a 1:1 digital twin scenario by integrating the topography and ecological cycle of the protected area, simulate the theoretically feasible path library of drones under physical constraints and ecologically sensitive areas, and mark no-fly zones and ecologically high-value areas; S212: The intelligent monitoring equipment transmits wind speed and species trace data in real time through micro weather stations and edge ecology identification, dynamically corrects the digital twin scene, and the drone scans the environment with binocular vision and millimeter-wave radar when it is running, and compares the digital twin model to correct obstacles and energy consumption data. S213: An improved Q-learning algorithm is adopted to quantify inspection efficiency and ecological value into a reward function, train the path decision model, and after each inspection is completed, the data is sent back to optimize the path decision model in reverse, and output the main path and backup path.

[0008] Further, S3 includes: S31: The UAV flies to the target intelligent monitoring equipment according to the planned path, and sends a data transmission request signal to the intelligent monitoring equipment through a dedicated communication module. After receiving the signal, the intelligent monitoring equipment verifies the identity of the UAV through a preset encryption key. After successful verification, a two-way communication connection is established. S32: The intelligent monitoring device transmits a data index table to the drone, which contains metadata such as the file name, shooting time, file size, ambient temperature and light intensity of the image to be transmitted. After receiving the data, the drone verifies the integrity of the index table using the MD5 checksum algorithm. If any data is missing, it requests retransmission. S33: Based on the index table, the UAV and intelligent monitoring equipment use a chunked transmission protocol to transmit image data; S34: After the data transmission of a single intelligent monitoring device is completed, the intelligent monitoring device marks the transmitted image in its local storage and sends a transmission completion signal to the drone. The drone records the next inspection time of the intelligent monitoring device and then flies to the next target intelligent monitoring device.

[0009] Further, S4 includes: S41: After the drone completes the data reception of all target intelligent monitoring equipment, it returns to the base station in the protected area according to the planned path. During the return process, if it passes through an area with basic network coverage, it uploads high-value images to the cloud platform in advance. S42: After the drone returns to the base station, it transmits all the stored image data to the base station server via a wired high-speed interface. The base station server performs a second verification on the data to eliminate transmission errors. S43: The base station server uploads complete data to the cloud management platform via a wired network or a wireless communication module adapted for the field. The cloud management platform classifies and processes the image data, establishes a data catalog according to "intelligent monitoring device number - shooting date", and optimizes species identification by combining image shooting timestamps and environmental data to generate a wildlife monitoring report.

[0010] Furthermore, the optimization of species identification by combining image capture timestamps and environmental data includes: S431: The shooting time is parsed into a time period label and the temperature is mapped into a temperature zone label using a multi-dimensional feature fusion model. Image features and environmental features are weighted and fused through an attention mechanism. The multi-dimensional feature fusion model adopts a dual-branch structure of CNN image recognition and environmental label encoding. S432: An improved Softmax classification is adopted, and environmental matching degree is introduced as a temperature coefficient to filter out species that do not match the time period and temperature at all. The output results include an explanation of environmental matching degree. S433: Construct a species library with environmental attributes based on historical observation data of the protected area, and use transfer learning to continuously fine-tune the multi-dimensional feature fusion model and dynamically update the time period and temperature association rules.

[0011] A data transmission system for intelligent wildlife monitoring equipment based on drone collaboration, used to implement the above method, includes: The intelligent monitoring equipment data acquisition module is used for data acquisition and local storage; The drone collaborative transmission module is used for drone inspection path planning and preparation. The base station data relay module is used for data exchange and transmission between the drone and the intelligent monitoring equipment; The cloud management platform module is used to transmit and analyze the data acquired by the drone in the cloud.

[0012] The beneficial effects of this invention are: (1) Data transmission efficiency has been greatly improved: the data acquisition cycle has been shortened from "once a month" to "once a week" when manually collecting cards, key high-value data can be uploaded 24 hours in advance, and the analysis delay has been reduced by 80%, meeting the needs of dynamic monitoring and emergency response of rare species; (2) Reduced labor costs and safety risks: No staff need to enter complex terrain to retrieve cards. The annual labor cost of a single national-level protected area can be reduced by more than 60%, and the safety hazards of people getting lost or encountering wild animals can be completely avoided. (3) More efficient data management: Through local marking by intelligent monitoring equipment and block transmission by drones, the proportion of effective data has increased from 15% in traditional manual screening to 90%, reducing the occupation of storage and transmission resources by invalid data; (4) Strong environmental adaptability: The intelligent monitoring equipment supports solar power supply and wide temperature operation, the drone path adapts to terrain and meteorological conditions, and the dedicated communication module ensures signal stability in complex environments. It can operate stably for a long time in protected areas with inconvenient network and harsh environment. (5) Excellent scalability: It can be adapted to different types of wildlife monitoring needs (such as large mammals and birds), and can also be extended to the data transmission scenario of "drone + other intelligent monitoring equipment" to realize the unified management of multi-dimensional monitoring data of the protected area. Attached Figure Description

[0013] Figure 1 This is a flowchart of a data transmission method for a drone-based intelligent wildlife monitoring device.

[0014] Figure 2 This is a timing diagram of data transmission between the drone and the intelligent monitoring equipment. Detailed Implementation

[0015] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0016] Example 1, as Figure 1 and Figure 2As shown, a data transmission method for a drone-based intelligent wildlife monitoring device includes the following steps: S1: Data collection and local storage using intelligent monitoring equipment, specifically including: S11: The intelligent monitoring equipment is deployed in the target monitoring area of ​​the protected area (such as the habitat of rare species and migration corridors). The equipment adopts a dual trigger mechanism of thermal sensing and motion. When the animal's body temperature is detected to be >25℃ (thermal trigger) or the image pixel change rate is >5% (motion trigger), it will automatically start shooting and generate a 1080P resolution color or infrared image. It will also automatically switch to infrared mode at night to ensure clear species characteristics. S12: The intelligent monitoring device has a built-in industrial-grade storage unit of at least 128GB. It names the captured images in the format of "capture timestamp-device number-serial number" (e.g., "202409051420_D08_001.jpg"). At the same time, it records the ambient temperature and light intensity at the time of capture through the integrated temperature sensor (accuracy ±0.5℃) and light sensor (range 0-10000lux), and stores them together with the images to provide environmental dimension reference for subsequent species identification. S13: The intelligent monitoring device supports data tagging function, which automatically classifies continuously captured images of the same target (such as a group of wild animals) to avoid data fragmentation, and sets up a storage protection mechanism. When the remaining power is <10%, it prioritizes the retention of stored data and stops capturing new images.

[0017] S2: Planning and preparing inspection routes for drones, specifically including: S21: The cloud management platform pre-imports the GPS location information of all intelligent monitoring devices in the protected area (obtained through the built-in GPS module of the intelligent monitoring device, with a positioning accuracy of ±5m), and combines the drone parameters (maximum flight time of 60 minutes, flight speed of 15m / s, maximum payload of 1kg) with the terrain data of the protected area (such as areas with a slope >30° being designated as no-fly zones) to generate the optimal inspection route; Generating the optimal inspection path includes: S211: Integrate the topography of the protected area (UAV mapping + satellite remote sensing) and ecological cycles (such as no-fly zones during the breeding season) to construct a 1:1 digital twin scenario, pre-simulate the theoretically feasible path library of UAVs under physical constraints (slope, load) and ecologically sensitive areas, and mark no-fly zones and ecologically high-value areas. S212: The intelligent monitoring equipment transmits wind speed and species trace data in real time through micro weather stations and edge ecology identification, dynamically corrects the digital twin scene, and the drone scans the environment with binocular vision and millimeter-wave radar when it is running, and compares the digital twin model to correct obstacles and energy consumption data. S213: An improved Q-learning algorithm is adopted to quantify inspection efficiency (power consumption, mileage) and ecological value (coverage of high-value areas, detour of sensitive areas) into reward functions, train the path decision model, and after each inspection is completed, the data is sent back to optimize the path decision model in reverse, output the main path (coverage priority equipment + ecological area) and the backup path (to deal with emergencies), and realize dynamic self-optimization.

[0018] S22: The drone is equipped with a dedicated communication module (with long-distance transmission capability, low power consumption, and suitable for signal transmission requirements in complex terrain of protected areas), at least 128GB high-speed storage module (read / write speed 90MB / s) and GPS navigation module (positioning accuracy ±1m). Before flight, the optimal inspection path, target intelligent monitoring equipment list and encryption key (for authentication) are loaded through the cloud management platform to ensure clear mission and secure data transmission.

[0019] S3: Data exchange and transmission between drones and intelligent monitoring equipment, specifically including: S31: The UAV flies to the target intelligent monitoring equipment according to the planned path, and sends a data transmission request signal to the intelligent monitoring equipment through a dedicated communication module. After receiving the signal, the intelligent monitoring equipment verifies the identity of the UAV through a preset encryption key. After successful verification, a two-way communication connection is established. The information for the dedicated communication module is as follows: Data is exchanged with intelligent monitoring devices via an industrial-grade WiFi 7 intelligent collaborative communication module (EC - WiFi 7 - Pro - 01). The key parameters and design of the module are as follows: 1. Physical layer characteristics: It operates primarily in the 6GHz band (compatible with multiple bands), with a 160MHz channel supporting an effective transmission distance of 30m (attenuation <15% in complex terrain), and an actual user rate of 1.2 - 1.5Gbps, meeting the requirement of transmitting 1080P images at 0.8 - 1.2 seconds per frame.

[0020] 2. Protocols and Security: Customized low-latency and high-reliability protocol, supporting TWT wake-up (standby power consumption <0.3W); adopts WPA3+ device PSK authentication, with a 100% illegal access identification rate, ensuring the security of ecosystem data.

[0021] Environment adaptation: IP67 protection, wide operating temperature range of -40℃ to 85℃, dimensions of 30mm×20mm×3mm, weight of <5g, integration and adaptation cycle of <2 weeks, balancing harsh outdoor environments with lightweight integration.

[0022] The module supports 6GHz fast handshake between drones and devices (latency <20ms), and dynamically switches to 2.4GHz to protect critical data when the battery is low. The device uses sub-GHz pre-wake-up, and the power consumption per interaction is <0.1mAh. It achieves a 98% transmission success rate and a 30% improvement in device battery life in complex scenarios, solving the problems of "poor compatibility, insecurity, and high power consumption" of traditional communication modules.

[0023] S32: The intelligent monitoring device transmits a data index table to the drone, which contains metadata such as the file name, shooting time, file size, ambient temperature and light intensity of the image to be transmitted. After receiving the data, the drone verifies the integrity of the index table using the MD5 checksum algorithm. If any data is missing, it requests retransmission. S33: Based on an index table, the drone and intelligent monitoring equipment use a chunked transmission protocol to transmit image data. Specifically: A single image is divided into 1MB data blocks. After each block is transmitted, the drone sends a "successful reception" confirmation signal to the intelligent monitoring device. If no confirmation signal is received for three consecutive times, the intelligent monitoring device automatically triggers retransmission to ensure that no data is lost. Support for priority transmission of high-value data: The intelligent monitoring equipment prioritizes the transmission of suspected high-value images (such as images containing large mammals or rare birds) based on image pixel characteristics (such as whether the target's size and outline match the preset characteristics of rare species), ensuring that key data is transferred first.

[0024] S34: After the data transmission of a single intelligent monitoring device is completed, the intelligent monitoring device marks the transmitted image in its local storage (marked as "synchronized") and sends a transmission completion signal to the drone. The drone records the next inspection time of the intelligent monitoring device (default interval of 7 days, which can be adjusted via the cloud) and then flies to the next target intelligent monitoring device.

[0025] S4: Transmit the data acquired by the drone back to the cloud for analysis, specifically including: S41: After receiving data from all target intelligent monitoring devices, the drone returns to the protected area base station (usually located at the entrance of the protected area, with mains power supply and basic network conditions) according to the planned path. During the return process, if it passes through an area with basic network coverage, it uploads high-value images to the cloud platform in advance (such as images of suspected rare species or abnormal behavior) to achieve "rapid reporting" of key information. S42: After the drone returns to the base station, it transmits all the stored image data to the base station server via a wired high-speed interface (the transmission rate meets the needs of rapid transmission of a large number of images). The base station server performs a second verification on the data (comparing the data volume with the records in the index table to eliminate transmission errors and ensure data integrity. S43: The base station server uploads complete data to the cloud management platform via a wired network or a wireless communication module adapted for the field. The cloud management platform classifies and processes the image data, establishes a data catalog according to "intelligent monitoring device number - shooting date", and optimizes species identification by combining image shooting timestamps and environmental data to generate a wildlife monitoring report.

[0026] Combining image capture timestamps with environmental data to optimize species identification includes: S431: The shooting time is parsed into time period labels such as "dawn / dusk / noon" using a multi-dimensional feature fusion model, and the temperature is mapped into temperature zone labels such as "warm-loving / cold-resistant". Image features and environmental features are weighted and fused through an attention mechanism. The multi-dimensional feature fusion model adopts a dual-branch structure of CNN image recognition and environmental label encoding. S432: The improved Softmax classification is adopted, and the environmental matching degree is introduced as the temperature coefficient to filter out species that do not match the time and temperature at all (such as snow leopards at noon, which are directly downgraded due to their low historical activity rate). The output results include environmental matching degree descriptions (such as "snow leopard, dawn and dusk + cold resistance matching degree 85%"). S433: Based on historical observation data of the protected area, a species database with environmental attributes is constructed. Transfer learning is used to continuously fine-tune the multi-dimensional feature fusion model, dynamically updating the time period and temperature association rules to adapt to changes in the activity patterns of wild species. This scheme transforms environmental data from "auxiliary information" into "decision-making basis" through algorithms, improving the recognition accuracy of blurred images (such as at night or at long distances) and overcoming the shortcomings of traditional models that "only look at images and are detached from ecological laws."

[0027] Finally, a wildlife monitoring report is generated, which includes information such as species type, frequency of occurrence, and heat map of distribution location, supporting remote viewing and decision-making by reserve managers.

[0028] Example 2: A data transmission system for intelligent wildlife monitoring equipment based on UAV collaboration, used to implement the above method, includes: The intelligent monitoring equipment data acquisition module is used for data acquisition and local storage; Hardware components: The main body of the intelligent monitoring equipment supports dual triggering of thermal sensing (detection range 0-50℃) and motion, has a night vision distance of 30m, and an operating temperature of -30℃ to 70℃, making it suitable for the extreme environment of the protected area. Storage unit: 128GB industrial-grade Micro SD card, supporting cyclic overwrite (deleting the oldest invalid data when storage space is insufficient); Communication and positioning unit: Dedicated communication module adapted for field use (meets the requirements of long-distance, low-power transmission) and GPS positioning module (positioning accuracy ±5m); Environmental sensors: temperature sensor (accuracy ±0.5℃), light sensor (range 0-10000 lux); Functions: Capture wildlife images in real time, record and store environmental data locally, establish secure communication with drones and transmit data; Power supply design: It adopts a 20W polycrystalline silicon solar panel + 10Ah lithium iron phosphate battery pack, supports low power consumption mode (power consumption <0.5W when there is no animal activity), and ensures stable operation for ≥72 hours in continuous rainy weather.

[0029] The drone collaborative transmission module is used for drone inspection path planning and preparation. Hardware components: Drone body: multi-rotor structure, maximum flight time of 60 minutes, payload of 1kg, supports ultrasonic automatic obstacle avoidance (detection distance of 0.5-5m). Communication and storage unit: Dedicated communication module adapted for field use (matching the communication needs of intelligent monitoring equipment to ensure stable link), 128GB high-speed solid-state drive; Navigation and environmental sensing unit: GPS navigation module (positioning accuracy ±1m), small meteorological sensor (real-time acquisition of wind speed and temperature); Function: Receives cloud-based inspection path instructions, flies to the location of the intelligent monitoring device and establishes communication, receives and stores image data, and completes data relay after returning to the base station; Control logic: When the remaining battery power is less than 20%, a low battery return-to-home mechanism is triggered; when wind speed is greater than level 5 or precipitation is detected, the flight is suspended and the aircraft lands at a preset safe point to ensure the safety of equipment and data.

[0030] The base station data relay module is used for data exchange and transmission between the drone and the intelligent monitoring equipment; Hardware components: Server: Industrial-grade host (CPU i5, 8GB memory, 512GB storage), supporting 24-hour uninterrupted operation; Interface unit: wired high-speed interface (for connecting to drones), gigabit Ethernet interface, and wireless communication module adapted for outdoor use (for uploading data to the cloud). Protective structure: IP65 protection rating shell, suitable for outdoor humid and dusty environments; Function: Receive image data transmitted by the drone, perform secondary verification, temporarily store the data and upload it to the cloud platform, and issue the next inspection task instruction to the drone.

[0031] The cloud management platform module is used to transmit and analyze the data acquired by the drone in the cloud.

[0032] Functional modules: Path planning unit: Generates drone inspection paths based on GPS location, terrain, and meteorological data from intelligent monitoring equipment, and supports manual adjustment of no-fly zones and inspection cycles; Data Management Unit: Stores image data according to rules, provides multi-dimensional search functions, and supports data backup and export; Species-assisted identification unit: Combines data such as image capture time and ambient temperature to help determine species type and improve identification accuracy; Equipment monitoring unit: Real-time display of operational data of intelligent monitoring equipment (battery level, storage space, working status) and drones (location, remaining battery level, flight progress), and triggers SMS / platform alarms when abnormalities occur (such as low battery of intelligent monitoring equipment, drone deviating from its path).

[0033] In one embodiment of the present invention, taking a national nature reserve in Northwest China (area 500 km², network coverage <10%, mainly monitoring rare species such as snow leopards and Tibetan antelopes) as an example, the system is deployed as follows: Infrared camera deployment: 50 infrared cameras are deployed at 5km intervals, focusing on snow leopard habitat (rocky slopes) and Tibetan antelope migration routes. Each camera integrates a dedicated communication module and GPS, with a storage capacity of 128GB. Drone configuration: A multi-rotor drone (55-minute flight time, 1.5kg payload) is selected, equipped with a dedicated communication module matched with an infrared camera and a 128GB solid-state drive. The base station is located at the management office of the protected area (with mains power and basic network signal). Cloud initialization: Import all infrared camera GPS coordinates, set the inspection cycle to 7 days, connect to the meteorological station data interface of the protected area (to obtain real-time wind speed and temperature), and preset the pixel features of snow leopards and Tibetan antelopes (for high-value data priority transmission judgment).

[0034] The implementation process is as follows: Infrared camera data acquisition: Infrared camera D12 (located in the core habitat of snow leopard) detected thermal signals (temperature -5℃, which is within the snow leopard's body temperature range) and movement triggers, and continuously captured 8 snow leopard images, named "202409060815_D12_001~008.jpg", while simultaneously recording the ambient temperature of -6℃ and the light intensity of 800 lux; Drone inspection: The cloud platform generates an inspection route (covering 10 infrared cameras, with a total flight distance of 18km), and determines the departure time at 8:00 AM based on meteorological data (wind speed level 3, no precipitation). Data interaction: At 8:35, the drone flew to a position 18m above the D12 camera, completed authentication and established a connection through a dedicated communication module, and transmitted 8 snow leopard images (48MB in total) within 6 minutes, with no packet loss in the MD5 check. Data transmission and analysis: The drone returned to the base station at 10:10 and transmitted 500MB of images from 10 cameras to the base station via a wired high-speed interface. After the base station uploaded the data to the cloud, the platform combined the shooting time of the D12 camera images (8:15, the snow leopard's active morning period) and the ambient temperature (-6℃, consistent with the snow leopard's cold-resistant habits) to help identify the images as snow leopards with an accuracy rate of 95%, and finally generated a "Snow Leopard Weekly Activity Report".

[0035] Based on the above implementation process, the data acquisition cycle was shortened from 30 days to 7 days, and the activity trajectories of individual snow leopards were successfully recorded 3 times (traditional manual card retrieval failed to capture them in time). Labor costs have been reduced from 200,000 yuan per year to 80,000 yuan, and no safety incidents have occurred involving personnel entering high-altitude and complex terrain. The species identification accuracy of nighttime blurred infrared images has been improved from 80% to 93%, reducing misidentification of snow leopards with other felines; The infrared camera can work continuously for 60 days on a single charge, and the drone can complete data transmission for 10 cameras in a single flight. The dedicated communication module ensures a transmission success rate of over 98%, and the system stability meets the needs of long-term monitoring.

[0036] Compared to existing wildlife monitoring technologies, the core innovations of this invention are reflected in the following four dimensions, and it does not rely on other external technical solutions, focusing solely on the technological breakthrough of the "intelligent monitoring equipment-drone" collaborative transmission system: 1. Break through the dependence on "physical card retrieval" and build a closed loop for mobile data transmission. While existing monitoring technologies can optimize data at the edge, they still rely on manual card retrieval or fixed network transmission in scenarios without network access. This invention uses drones as mobile transmission carriers and completely eliminates the limitations of physical card retrieval through a link of "intelligent monitoring equipment local storage - drone inspection reception - base station relay - cloud analysis". The data acquisition cycle is shortened from "once a month" to "once a week", and key data can be reported 24 hours in advance, solving the "last mile" problem of data transmission in remote areas.

[0037] 2. Design an "environment-adaptive" collaborative communication mechanism to ensure transmission stability in complex scenarios. Unlike commonly used fixed sensor communication logic, this invention is designed for the complex terrain (mountains, dense forests) and meteorological conditions (low temperature, strong winds) of protected areas. It is equipped with a dedicated communication module adapted for the field, combined with a 10-20m height coordinated positioning design of "UAV-intelligent monitoring equipment" (avoiding obstructions), and a segmented transmission and retransmission mechanism to achieve a transmission success rate of over 98%. At the same time, the communication process uses a preset encryption key to verify identity, avoiding data leakage or unauthorized device access, thus balancing stability and security.

[0038] 3. Implement "prioritizing the transmission of high-value data" to optimize data management efficiency. Existing technologies mostly rely on single-side screening; this invention adds a "pixel feature prediction" function to the intelligent monitoring device, which prioritizes the transmission of high-value images of suspected rare species and abnormal behaviors based on basic features such as target size and outline, and then transmits ordinary data simultaneously, increasing the proportion of effective data from 15% in traditional manual screening to 90%; at the same time, the intelligent monitoring device marks the transmitted data as "synchronized" to avoid duplicate transmission and reduce storage and bandwidth waste.

[0039] 4. Construct a "multi-device collaborative management and control" system to adapt to long-term unattended operation scenarios. This invention establishes a multi-device collaborative management system consisting of "intelligent monitoring equipment - drone - base station - cloud": the cloud platform can monitor the power and storage space of the intelligent monitoring equipment, the remaining power and flight trajectory of the drone in real time, and trigger alarms when abnormalities occur; the intelligent monitoring equipment adopts dual power supply of solar energy and lithium battery, supports low power consumption mode (power consumption <0.5W when there is no animal activity), and the drone has automatic obstacle avoidance and low power return-to-home functions, which are well adapted to the extreme environmental needs of long-term unattended operation in protected areas.

[0040] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the invention.

Claims

1. A data transmission method for intelligent wildlife monitoring equipment based on UAV collaboration, characterized in that, Includes the following steps: S1: Data collection and local storage are performed using intelligent monitoring equipment; S2: Plan and prepare inspection routes for drones; S3: Data exchange and transmission between drones and intelligent monitoring equipment; S4: Transmit the data acquired by the drone back to the cloud for analysis.

2. The data transmission method for intelligent wildlife monitoring equipment based on UAV collaboration according to claim 1, characterized in that, S1 includes: S11: The intelligent monitoring equipment is deployed in the target monitoring area of ​​the protected area. The equipment adopts a dual triggering mechanism of thermal sensing and motion. When the animal's body temperature is detected to be >25℃ or the image pixel change rate is >5%, it will automatically start shooting and generate a 1080P resolution color or infrared image. It will also automatically switch to infrared mode at night. S12: The intelligent monitoring device has a built-in industrial-grade storage unit of at least 128GB. It names the captured images in the format of "capture timestamp-device number-serial number". At the same time, it records the ambient temperature and light intensity at the time of capture through integrated temperature and light sensors and stores them together with the images. S13: The intelligent monitoring device supports data tagging function, automatically classifies continuously captured images of the same target, and sets up a storage protection mechanism. When the remaining power is <10%, it prioritizes the retention of stored data and stops capturing new images.

3. The data transmission method for intelligent wildlife monitoring equipment based on UAV collaboration according to claim 1, characterized in that, S2 includes: S21: The cloud management platform pre-imports the GPS location information of all intelligent monitoring devices in the protected area, and combines the drone parameters with the terrain data of the protected area to generate the optimal inspection route; S22: The drone is equipped with a dedicated communication module, at least 128GB of high-speed storage module and GPS navigation module. Before flight, the optimal inspection route, target intelligent monitoring equipment list and encryption key are loaded through the cloud management platform.

4. The data transmission method for intelligent wildlife monitoring equipment based on UAV collaboration according to claim 3, characterized in that, The generation of the optimal inspection path includes: S211: Construct a 1:1 digital twin scenario by integrating the topography and ecological cycle of the protected area, simulate the theoretically feasible path library of drones under physical constraints and ecologically sensitive areas, and mark no-fly zones and ecologically high-value areas; S212: The intelligent monitoring equipment transmits wind speed and species trace data in real time through micro weather stations and edge ecology identification, dynamically corrects the digital twin scene, and the drone scans the environment with binocular vision and millimeter-wave radar when it is running, and compares the digital twin model to correct obstacles and energy consumption data. S213: An improved Q-learning algorithm is adopted to quantify inspection efficiency and ecological value into a reward function, train the path decision model, and after each inspection is completed, the data is sent back to optimize the path decision model in reverse, and output the main path and backup path.

5. The data transmission method for intelligent wildlife monitoring equipment based on UAV collaboration according to claim 1, characterized in that, S3 includes: S31: The UAV flies to the target intelligent monitoring equipment according to the planned path, and sends a data transmission request signal to the intelligent monitoring equipment through a dedicated communication module. After receiving the signal, the intelligent monitoring equipment verifies the identity of the UAV through a preset encryption key. After successful verification, a two-way communication connection is established. S32: The intelligent monitoring device transmits a data index table to the drone, which contains metadata such as the file name, shooting time, file size, ambient temperature and light intensity of the image to be transmitted. After receiving the data, the drone verifies the integrity of the index table using the MD5 checksum algorithm. If any data is missing, it requests retransmission. S33: Based on the index table, the UAV and intelligent monitoring equipment use a chunked transmission protocol to transmit image data; S34: After the data transmission of a single intelligent monitoring device is completed, the intelligent monitoring device marks the transmitted image in its local storage and sends a transmission completion signal to the drone. The drone records the next inspection time of the intelligent monitoring device and then flies to the next target intelligent monitoring device.

6. The data transmission method for intelligent wildlife monitoring equipment based on UAV collaboration according to claim 1, characterized in that, S4 includes: S41: After the drone completes the data reception of all target intelligent monitoring equipment, it returns to the base station in the protected area according to the planned path. During the return process, if it passes through an area with basic network coverage, it uploads high-value images to the cloud platform in advance. S42: After the drone returns to the base station, it transmits all the stored image data to the base station server via a wired high-speed interface. The base station server performs a second verification on the data to eliminate transmission errors. S43: The base station server uploads complete data to the cloud management platform via a wired network or a wireless communication module adapted for the field. The cloud management platform classifies and processes the image data, establishes a data catalog according to "intelligent monitoring device number - shooting date", and optimizes species identification by combining image shooting timestamps and environmental data to generate a wildlife monitoring report.

7. The data transmission method for intelligent wildlife monitoring equipment based on UAV collaboration according to claim 1, characterized in that, The method of combining image capture timestamps and environmental data to optimize species identification includes: S431: The shooting time is parsed into a time period label and the temperature is mapped into a temperature zone label using a multi-dimensional feature fusion model. Image features and environmental features are weighted and fused through an attention mechanism. The multi-dimensional feature fusion model adopts a dual-branch structure of CNN image recognition and environmental label encoding. S432: An improved Softmax classification is adopted, and environmental matching degree is introduced as a temperature coefficient to filter out species that do not match the time period and temperature at all. The output results include an explanation of environmental matching degree. S433: Construct a species library with environmental attributes based on historical observation data of the protected area, and use transfer learning to continuously fine-tune the multi-dimensional feature fusion model and dynamically update the time period and temperature association rules.

8. A data transmission system for intelligent wildlife monitoring equipment based on drone collaboration, used to implement the method described in any one of claims 1-7, characterized in that, include: The intelligent monitoring equipment data acquisition module is used for data acquisition and local storage; The drone collaborative transmission module is used for drone inspection path planning and preparation. The base station data relay module is used for data exchange and transmission between the drone and the intelligent monitoring equipment; The cloud management platform module is used to transmit and analyze the data acquired by the drone in the cloud.

Images