Data offline collection and intelligent synchronization method in weak network environment of national forest construction

Abstract

This invention discloses a method for offline data collection and intelligent synchronization in a weak network environment during national reserve forest construction, relating to the field of forestry information monitoring technology. The method includes: deploying multiple monitoring terminals within the forest construction area, each terminal integrating an edge processing module; the terminals collecting raw data, which the edge processing module then divides into routine monitoring data and event-triggered data; a drone cruising along a preset route, initiating a connection establishment process when it enters the theoretical communication range of the target monitoring terminal; after establishing a connection, the drone receiving data uploaded by the target monitoring terminal; and after data upload is complete, the monitoring terminal clearing the cache of synchronized data. This invention addresses the problem of unstable wireless communication caused by complex and obstructed environments in forest areas by proposing a connection establishment method where a ground monitoring terminal guides the drone to adjust its spatial position, significantly improving the initial connection success rate and communication quality.

Description

Technical Field

[0001] This invention relates to the field of forestry information monitoring technology, specifically to a method for offline data collection and intelligent synchronization in the context of weak network conditions during the construction of national reserve forests. Background Technology

[0002] The National Reserve Forest Project is a crucial strategic project for my country to ensure timber security and practice ecological civilization. Its construction and management require comprehensive and accurate real-time monitoring of tree growth, soil and weather conditions, pests and diseases, and fire risks. Currently, in vast forest areas with weak network infrastructure, the following two data collection methods are primarily relied upon:

[0003] Manual inspection method: Staff regularly go deep into the forest area to manually record or collect data on-site using handheld devices, and then organize and upload the data back at their base. This method has prominent problems such as high labor costs, high workload, poor data real-time performance, and limited coverage, making it difficult to meet the needs of refined and intelligent management of national reserve forests.

[0004] Traditional Wireless Sensor Networks (WSNs) involve deploying sensor nodes in forest areas, aggregating data to a gateway via a multi-hop self-organizing network, and then transmitting the data over long distances. However, the complex terrain and dense vegetation in forest areas cause severe wireless signal attenuation, resulting in low reliability of multi-hop transmission, difficulties in network maintenance, and uneven energy consumption leading to premature failure of some nodes. Using satellite communication for data transmission presents a bottleneck due to extremely high equipment and communication costs. Summary of the Invention

[0005] The purpose of this invention is to provide a method for automatic data acquisition and synchronization that can adapt to complex forest environments and achieve efficient and reliable communication.

[0006] To achieve the above objectives, this invention provides a method for offline data acquisition and intelligent synchronization in a weak network environment during the construction of national reserve forests. The method includes the following steps:

[0007] Within the forest construction area, multiple monitoring terminals are deployed in a distributed manner according to site conditions and monitoring needs. Each monitoring terminal integrates monitoring sensors and an edge processing module.

[0008] The monitoring sensor collects raw data, and the edge processing module processes the collected raw data, dividing it into regular monitoring data and event-triggered data based on the processing results.

[0009] The drone cruises along a preset route. When the drone enters the theoretical communication range of the target monitoring terminal, the connection establishment process is initiated.

[0010] Once the drone establishes a connection with the target monitoring terminal, the drone receives routine monitoring data and event-triggered data uploaded by the target monitoring terminal.

[0011] After the data upload is complete, the monitoring terminal clears the cache of the synchronized data and records the timestamp of this upload.

[0012] The principle of this invention is to decompose the monitoring task into two collaborative stages. The first is data acquisition and local preprocessing, accomplished through monitoring terminals deployed in a distributed manner within the forest area. Each terminal continuously monitors data and locally categorizes it into routine monitoring data and event-triggered data reflecting abnormalities or emergencies, and caches the data. The second is data aggregation, accomplished by a drone cruising along a preset route. When the drone approaches the target monitoring terminal, a communication connection is established, and the monitoring terminal uploads the cached data to the drone. After data synchronization, the monitoring terminal releases its storage space and records the synchronization time. This method utilizes the mobility of drones as mobile data relays, achieving automated data collection from widely distributed monitoring nodes without relying on fixed ground communication infrastructure. This overcomes the shortcomings of traditional manual inspection methods, such as low efficiency, high cost, and poor real-time performance, as well as the deployment difficulties and unreliable communication problems caused by severe obstruction in fixed wireless sensor networks.

[0013] Furthermore, the connection establishment process includes:

[0014] The drone transmits a broadcast detection signal containing its current three-dimensional spatial coordinates and flight attitude information;

[0015] After the target monitoring terminal receives the broadcast detection signal, the edge processing module calculates the theoretically optimal communication angle with the UAV based on the three-dimensional spatial coordinates of the monitoring terminal and using a pre-stored tree shading angle model.

[0016] The monitoring terminal sends the theoretically optimal communication angle to the drone;

[0017] After receiving the theoretically optimal communication angle, the UAV adjusts its position so that the axis of its communication antenna points to match the theoretically optimal communication angle, and then initiates a formal connection request.

[0018] In traditional methods, drones and terminals typically require multiple signal probes and handshakes, a time-consuming process with a success rate highly dependent on environmental factors. In this solution, as the drone approaches the target terminal, it actively broadcasts a signal containing its position and real-time flight attitude. Upon receiving this signal, the monitoring terminal invokes its pre-stored tree obstruction angle model and, combining its own coordinates with the drone's coordinates, calculates a theoretically optimal communication angle that effectively avoids major tree obstructions. Based on this angle, the drone adjusts its hovering position or altitude to align the main beam of its onboard communication antenna with the calculated elevation angle before initiating a formal communication connection request. This transforms the previously blind connection attempt process into a precise position alignment operation, significantly improving the initial connection success rate and reducing connection establishment latency and power consumption for both communicating parties.

[0019] Furthermore, the tree shading angle model is used to output the critical communication angle necessary to prevent tree shading under different azimuth angles of the monitoring terminal;

[0020] The forest shade angle model is constructed in the following way:

[0021] After initial deployment, the edge processing module divides the azimuth angle into several continuous zones.

[0022] The monitoring terminal continuously records external signal information from historical communication events. The external signal information includes signal reception strength, as well as the azimuth and elevation angles of the signal source relative to the monitoring terminal.

[0023] Within each azimuth zone, the relationship between signal received strength and the corresponding elevation angle is statistically fitted to obtain the fitted curve;

[0024] The elevation angle corresponding to the critical point in the fitted curve where the signal reception strength begins to fall below a preset reliable communication threshold is determined as the critical communication angle for that azimuth zone.

[0025] Based on the critical communication angles corresponding to each azimuth zoning zone, a polar coordinate model centered on the monitoring terminal is constructed as the forest shading angle model.

[0026] Each monitoring terminal, after deployment, autonomously learns over a long period to establish its own tree shading angle model. The terminal continuously records all captured external wireless signal events (from passing drones or other beacons) and divides all horizontal directions into several consecutive azimuth zones. Within each zone, all historically recorded signal reception strength-elevation angle data points are statistically fitted to generate a trend curve reflecting the signal strength change with elevation angle in that direction. By analyzing this fitted curve, the critical communication angle for each azimuth zone is determined. Finally, using the terminal as the origin, the horizontal azimuth angle as the angular coordinate, and the critical communication angle as the radial representation value, a polar coordinate model, i.e., the tree shading angle model, is constructed. This method enables communication guidance decisions to be based on realistic and dynamic environmental perception.

[0027] Furthermore, based on the three-dimensional spatial coordinates of the monitoring terminal, and using a pre-stored forest shading angle model, the theoretically optimal communication angle for communicating with the UAV is calculated, including:

[0028] Based on the three-dimensional spatial coordinates of the monitoring terminal and the UAV, the horizontal azimuth angle α and the horizontal geometric distance d of the UAV relative to the monitoring terminal are calculated. h and height difference Δh;

[0029] Match the horizontal azimuth angle α to the tree shading angle model and output the critical communication angle β of the azimuth angle partition corresponding to α;

[0030] Based on the critical angle of faith β, the height difference Δh, and the geometric distance d on the horizontal plane h The theoretical optimal communication angle is obtained by iteratively calculating the flight attitude information of the UAV.

[0031] Furthermore, the goal of the iterative calculation is to find a candidate beacon angle γ that simultaneously satisfies constraints A and B. c As the optimal perspective for the aforementioned theory:

[0032] Among them, constraint A: based on the candidate belief angle γ c With respect to the horizontal azimuth angle α, define a desired communication link from the monitoring terminal to the expected communication point in the air;

[0033] Based on the UAV's flight attitude information, the direction vector of the desired communication link is transformed from the ground coordinate system to the UAV's body coordinate system, and the actual effective communication angle γ of the desired communication link in the UAV's body coordinate system is calculated. e ;

[0034] Actual effective faith angle γ e Greater than the critical belief angle β;

[0035] Constraint B: In the UAV body coordinate system, the direction of the desired communication link is within the effective radiation main lobe range of the UAV onboard communication antenna in the current real-time flight attitude.

[0036] Among them, constraint A focuses on the reliability of the signal propagation path. First, based on the candidate elevation angle γ... c A virtual communication link is defined in spatial geometry, pointing from the monitoring terminal to the expected hovering position of the UAV, along with the horizontal azimuth angle α. Then, considering that the UAV's current pitch, roll, and other attitudes will cause the actual pointing of the airborne antenna to deviate from its geometric axis, the direction vector of this geometric link is transformed from the ground reference system to the UAV's body coordinate system through coordinate transformation. This allows the calculation of the actual effective communication angle γ of the terminal relative to the actual orientation of the UAV antenna. e This constraint ultimately requires the calculation of γ. e The angle γ must be greater than the critical communication angle β found in the model. This ensures that, under the actual antenna pointing direction, the communication link remains spatially higher than environmental obstacles, satisfying the geometric conditions for line-of-sight or quasi-line-of-sight propagation. Constraint B focuses on the validity of the antenna radiation pattern. It requires that, after the above coordinate transformation, the direction of the desired communication link defined in the body coordinate system must fall within the effective main lobe angle range of the UAV's onboard communication antenna in its current specific attitude. The antenna's main lobe is the direction in which its radiated energy is most concentrated; only signal receiving points within this range can obtain sufficiently strong signal strength. The optimal angle γ is found iteratively by satisfying both constraints. c This enabled the establishment of a high-quality link.

[0037] Furthermore, the forest construction area is divided into several independent sub-monitoring areas. In any sub-monitoring area, a unique main monitoring terminal and several auxiliary monitoring terminals are deployed, and the main monitoring terminal and the auxiliary monitoring terminals are connected in communication.

[0038] The method further includes:

[0039] The drone passes through all sub-monitoring areas sequentially according to the preset main cruise path;

[0040] Within any sub-monitoring area, the UAV establishes a connection with the main monitoring terminal of that sub-monitoring area and receives a local state set returned by the main monitoring terminal. The local state set includes the state information of the main monitoring terminal itself and its affiliated monitoring terminals.

[0041] Based on the local state set, the UAV updates its built-in value probability map, which is used to record the location coordinates of all monitoring terminals in the forest construction area, the comprehensive collection of value estimates and the confidence level of the estimates.

[0042] Based on the updated value probability map, its own real-time location, and remaining battery power, the drone plans a sub-cruise path covering multiple auxiliary monitoring terminals within the current sub-monitoring area.

[0043] The drone receives data information collected by the auxiliary monitoring terminal based on the sub-cruise path.

[0044] The forest area is vast, with numerous terminals and limited drone endurance. Simply having drones traverse all terminals along fixed routes would be inefficient and fail to prioritize the retrieval of critical data. This solution divides the entire forest area into several sub-monitoring zones, setting up a main monitoring terminal and multiple auxiliary monitoring terminals within each zone. Local communication connections are established between the main and auxiliary terminals, forming a local network with the main terminal as the information aggregation point. Drone patrols are divided into two levels: the upper level uses a pre-set main patrol path to guide the drone to visit each sub-zone sequentially; the lower level involves the drone first establishing a connection with the corresponding main monitoring terminal within each sub-zone. The drone obtains a local state set from the main terminal. Based on this set, the drone updates its internally maintained value probability map. Subsequently, based on the real-time updated value probability map, its current location, and remaining battery power, the drone dynamically calculates and generates a sub-pattern covering multiple auxiliary monitoring terminals within the current sub-zone, and then visits these terminals sequentially along this path to collect data. This method transforms the data collection behavior of drones from a static, pre-programmed mode to an intelligent mode based on dynamic planning of real-time regional state information, thereby significantly improving the overall efficiency and value of data collection within a limited flight time.

[0045] Furthermore, the status information includes: the remaining battery power of the monitoring terminal, the total amount of data cached, the time elapsed since the last data upload, and the amount of regular monitoring data and event-triggered data cached by the monitoring terminal;

[0046] The calculation method for the comprehensive collection value assessment includes:

[0047] The data urgency value is calculated based on the remaining battery power of the monitoring terminal, the total amount of data cache, and the time since the last data upload.

[0048] The inherent value of the data is calculated based on the amount of regular monitoring data and event-triggered data cached by the monitoring terminal.

[0049] The comprehensive collection value estimate is obtained by weighting and summing the data urgency value with the data intrinsic value.

[0050] In path planning decisions, when faced with the technical problem of how to obtain a comprehensive data collection value estimate to assess the priority of data collection for each terminal, traditional methods only consider a single factor (such as the amount of data), which cannot fully reflect the urgency and value of the collection. This invention first calculates the data urgency value, which is determined by three factors: (1) the remaining battery power of the terminal. The lower the battery power, the higher the risk of permanent data loss due to power failure, and the greater the urgency; (2) the total amount of data cache. The fuller the cache, the higher the risk of new data not being able to be stored or old data being overwritten due to storage overflow, and the greater the urgency; (3) the time since the last successful data upload. The longer the time interval, the worse the data timeliness, and the higher the risk that the terminal may be in an abnormal state, and the greater the urgency. Secondly, the inherent value of the data is calculated. This value is calculated differently based on the type and quantity of data cached by the terminal: event-triggered data has higher priority and value; routine monitoring data has basic value. Finally, the calculated data urgency value and the inherent value of the data are weighted and summed to obtain the final comprehensive data collection value estimate. This estimate provides a comprehensive decision-making basis for the intelligent path planning of UAVs.

[0051] Furthermore, the valuation confidence level is updated in the following ways:

[0052] Before the patrol mission begins, the drone assigns an initial confidence level C to any auxiliary monitoring terminal within the forest construction area. in ;

[0053] When the auxiliary monitoring terminal communicates directly with the drone, its corresponding estimated confidence level is updated to the preset highest confidence level C. max ;

[0054] When the status information of the auxiliary monitoring terminal is sent by the main monitoring terminal, the corresponding estimated confidence level is updated to the confidence level currently stored by the monitoring terminal and the preset indirect confidence level C. ind The larger value in, where C max >C ind >C in ;

[0055] For auxiliary monitoring terminals not involved in the local state set, the corresponding estimated confidence level is updated by multiplying it by a decay factor.

[0056] In dynamic environments, to prevent UAVs from making inefficient flight decisions based on outdated or unreliable value information, it is necessary to properly manage the timeliness and reliability of the comprehensive value assessment. To address this technical problem, this invention proposes a confidence update mechanism, which measures the reliability of the assessment at the current moment. The update rules are divided into three categories based on the information acquisition method: (1) Direct communication update, which means obtaining first-hand, up-to-date, and accurate status information from the monitoring terminal; therefore, the assessment confidence is updated to the preset highest confidence level C. max This indicates that the information is completely reliable. (2) Indirectly obtain the update. At this time, update the estimated confidence level corresponding to the auxiliary terminal to its currently stored confidence level and a preset indirect confidence level C. ind The larger of the two. This rule means that the system considers indirect information to have a certain degree of credibility, but will not use it to reduce the confidence level of a potentially more reliable information obtained from earlier direct communication. (3) The decay update of unmentioned terminals: For those monitoring terminals in the value probability map that were not mentioned in this local state set, their status information was not updated. This is likely because these terminals have run out of power or are malfunctioning. The system believes that the credibility of their information decreases over time, so it multiplies their corresponding estimated confidence level by a decay factor less than 1 to decay it. This mechanism makes the reference significance of the value estimate of terminals that have not been accessed or queried for a long time gradually decrease, driving UAVs to prioritize exploring areas with updated or more reliable information.

[0057] Furthermore, based on the updated value probability map, its own real-time location, and remaining battery power, a sub-cruise route covering multiple auxiliary monitoring terminals within the sub-monitoring area is planned, including:

[0058] Based on the updated value probability map, the geographical locations, comprehensive collection value estimates, and estimate confidence levels of all auxiliary monitoring terminals within the current sub-monitoring area are extracted.

[0059] Using the connection point between the UAV and the current main monitoring terminal as the starting point of the sub-cruise path, the available cruise power budget within the current sub-monitoring area is calculated based on the UAV's real-time remaining power and the preset energy consumption per unit distance.

[0060] With the goal of maximizing the value of data recovered within the cruise power budget, and constrained by sequentially accessing multiple auxiliary monitoring terminals from the starting point and finally returning to the starting point, a sub-regional path optimization model is constructed.

[0061] Solve the sub-region path optimization model to obtain a sequential sequence of visits to multiple auxiliary monitoring terminals, which serves as the sub-cruise path.

[0062] This invention transforms the planning of a specific flight path sequence into a mathematical model and solves it. The specific steps include: extracting the geographical locations, comprehensive data acquisition value estimates, and confidence levels of all auxiliary monitoring terminals within the current sub-monitoring area; then determining the constraint boundary, using the connection point between the UAV and the main monitoring terminal as the starting point of the sub-cruise (which is also the planned return point); and calculating the maximum power consumption the UAV can consume after leaving this starting point to complete sub-area data acquisition and safely return to this point, based on the UAV's current real-time remaining battery power and the preset energy consumption per unit flight distance. This is the cruise power budget. Subsequently, an optimization model is constructed. The optimization objective of this model is to maximize the total data value (measured by the comprehensive data acquisition value estimate of the terminals) obtained by the UAV from the starting point, sequentially visiting a series of auxiliary monitoring terminals, and finally returning to the starting point, without exceeding the cruise power budget. The decision variables of the model are which terminals are visited and the order in which they are visited. Finally, a suitable optimization algorithm (such as heuristic search, dynamic programming, etc.) is used to solve the model, obtaining a specific sequence of auxiliary monitoring terminal visits, which is the desired sub-cruise path. Through this process, after entering each sub-region, the drone can dynamically generate a locally optimal data collection route based on the latest information and its own status, thereby maximizing the value of data collection under the hard constraint of endurance.

[0063] Furthermore, when constructing the sub-regional path optimization model, the estimated cost of the UAV traveling to any auxiliary monitoring terminal is determined in the following way:

[0064] Calculate the Euclidean distance from the previous node on the path to the geographical location of the auxiliary monitoring terminal, and use it as the base distance cost;

[0065] Obtain the current valuation confidence level of the auxiliary monitoring terminal in the value probability map;

[0066] A confidence cost coefficient is calculated based on the current valuation confidence level, wherein the confidence cost coefficient is negatively correlated with the current valuation confidence level;

[0067] The estimated cost is obtained by multiplying the base distance cost by the confidence cost coefficient.

[0068] If only physical distance is considered as cost, a drone might be attracted to a terminal with high historical value but extremely low confidence (potentially due to equipment malfunction), resulting in ineffective flight. This solution works by weighting and adjusting traditional distance costs to reflect the risks associated with information uncertainty. Cost calculation consists of three parts: first, the Euclidean distance from the previous node on the planned path to the current target terminal's geographical location, representing the basic physical energy consumption required for flight; second, the confidence cost coefficient, which is negatively correlated with the current estimated confidence level—the lower the confidence level, the higher the cost coefficient; and third, multiplying the above two to obtain the final estimated cost. In this invention, for a terminal with the same estimated value but lower confidence, its cost in the path optimization model will be higher. Therefore, when optimizing the total value, the algorithm tends to select terminals with a higher product of value and confidence, improving the robustness and efficiency of path planning in real-world dynamic environments.

[0069] One or more technical solutions provided by this invention have at least the following technical effects or advantages:

[0070] 1. This invention addresses the challenge of unstable wireless communication caused by complex obstruction environments in forest areas by proposing a connection establishment method where a ground monitoring terminal guides a UAV to adjust its spatial position. By constructing a polar coordinate tree obstruction angle model locally on each monitoring terminal based on historical communication data, the terminal can calculate the theoretically optimal alignment angle required to avoid known obstacles in real time when the UAV approaches, combining the UAV's position with its own model, and then feed this calculation back to the UAV. The UAV then precisely adjusts its hovering position based on this angle, aligning its antenna's main beam in that direction. This method transforms the traditional passive connection process, which relies on repeated signal strength probing, into a precise spatial alignment operation based on environmental awareness. This fundamentally solves the problems of difficult, time-consuming, and energy-intensive link establishment under non-line-of-sight conditions, significantly improving the initial connection success rate and communication quality, and laying a reliable physical link foundation for subsequent efficient data synchronization.

[0071] 2. This invention proposes an intelligent path planning mechanism based on dynamic value assessment and confidence management, effectively solving the optimization decision-making problem of prioritizing the collection of the most urgent and important data under the constraint of limited UAV endurance. This method, through the collaboration of a main monitoring terminal and auxiliary monitoring terminals, enables the UAV to acquire global regional status information in a single interaction, and constructs and updates a value probability map that integrates data value and timeliness reliability. Within a sub-region, the UAV no longer follows a fixed path, but dynamically solves a sub-cruise path that maximizes the total value of data recovery within the power budget, based on the map, real-time location, and remaining battery power. Uniquely, it transforms information confidence into a path cost coefficient, enabling the planning algorithm to automatically weigh the risks of utilizing known high-value targets versus exploring outdated targets, thereby ensuring that flight resources are always allocated to high-value and high-reliability data collection tasks. This significantly improves the data collection efficiency and system robustness of a single cruise. Attached Figure Description

[0072] The accompanying drawings, which are provided to further illustrate embodiments of the invention and constitute a part of this invention, are not intended to limit the scope of the invention.

[0073] Figure 1 This is a flowchart illustrating the method for offline data collection and intelligent synchronization in the context of weak network conditions during the construction of national reserve forests, as described in this invention. Detailed Implementation

[0074] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, where there is no conflict, the embodiments of the present invention and the features thereof can be combined with each other.

[0075] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0076] Example 1

[0077] Please refer to Figure 1 Embodiment 1 of the present invention provides a method for offline data collection and intelligent synchronization in a weak network environment during the construction of national reserve forests. The method includes the following steps:

[0078] Within the forest construction area, multiple monitoring terminals are deployed in a distributed manner according to site conditions and monitoring needs. Each monitoring terminal integrates monitoring sensors and an edge processing module.

[0079] The monitoring sensor collects raw data, and the edge processing module processes the collected raw data, dividing it into regular monitoring data and event-triggered data based on the processing results.

[0080] The drone cruises along a preset route. When the drone enters the theoretical communication range of the target monitoring terminal, the connection establishment process is initiated.

[0081] Once the drone establishes a connection with the target monitoring terminal, the drone receives routine monitoring data and event-triggered data uploaded by the target monitoring terminal.

[0082] After the data upload is complete, the monitoring terminal clears the cache of the synchronized data and records the timestamp of this upload.

[0083] In practical implementation, monitoring terminals can be deployed in forest areas at a density of 1 to 5 terminals per hectare, depending on the tree species, topography, and other site conditions. Each monitoring terminal integrates multiple monitoring sensors, such as tree runoff sensors, soil temperature and humidity sensors, miniature weather stations, and infrared thermal cameras. The edge processing module can use a low-power STM32 series MCU. The raw data collected by the sensors is processed by the edge processing module, including data filtering, format standardization, and data classification based on preset rules (such as temperature sudden changes exceeding 10°C, image recognition of fire points or pest characteristics). This data is divided into routine monitoring data reflecting normal conditions (such as growth data) and event-triggered data reflecting anomalies (such as fire alarms and pest alarms). Both types of data are cached in partitions in the terminal's local flash memory.

[0084] The drone, equipped with an onboard computer and communication module, cruises along a pre-planned route in the flight control system. When the drone reaches the theoretical communication range above a target monitoring terminal (this range depends on the power of the communication module used and can be set according to actual conditions), the connection establishment process is initiated. After successful connection establishment, the monitoring terminal packages and uploads its cached routine monitoring data and event-triggered data to the drone. After data synchronization is complete, the monitoring terminal clears the cache space of the uploaded data and records the timestamp of this synchronization for management purposes. After the cruise ends, the drone flies to an area with network signal and uploads all collected data in batches to the cloud data center.

[0085] The connection establishment process includes:

[0086] The drone transmits a broadcast detection signal containing its current three-dimensional spatial coordinates and flight attitude information;

[0087] After the target monitoring terminal receives the broadcast detection signal, the edge processing module calculates the theoretically optimal communication angle with the UAV based on the three-dimensional spatial coordinates of the monitoring terminal and using a pre-stored tree shading angle model.

[0088] The monitoring terminal sends the theoretically optimal communication angle to the drone;

[0089] After receiving the theoretically optimal communication angle, the UAV adjusts its position so that the axis of its communication antenna points to match the theoretically optimal communication angle, and then initiates a formal connection request.

[0090] In practice, the communication module on the UAV periodically broadcasts a detection signal. This signal encapsulates the UAV's real-time three-dimensional spatial coordinates obtained by the GPS / RTK module and the real-time flight attitude information obtained by the IMU, including pitch angle, roll angle and yaw angle.

[0091] The monitoring terminal's communication module continuously listens to the channel. Upon receiving a broadcast signal from the drone, its edge processing module is immediately activated and calls upon a tree shading angle model pre-stored in the module's internal flash memory, representing the tree shading situation around the terminal. Combining its own fixed three-dimensional coordinates (obtained during deployment) and the analyzed drone coordinates and attitude, the edge processing module runs a calculation program to solve for a theoretically optimal communication angle.

[0092] Subsequently, the monitoring terminal sends the angle value back to the drone. Upon receiving this angle, the drone's onboard computer, combined with its current coordinates, calculates the target spatial position it needs to adjust to; for example, it needs to climb to a position with an elevation angle of 35 degrees relative to the terminal and a constant azimuth angle. The drone then automatically adjusts its trajectory, flies to the target position, and hovers, with its onboard communication antenna axis pointing as close as possible to the calculated theoretically optimal communication angle. After adjustment, the drone initiates a formal communication connection request, such as a TCP handshake, thus greatly improving the initial connection success rate in complex, obstructed environments.

[0093] The tree shading angle model is used to output the critical communication angle necessary to prevent tree shading under different azimuth angles of the monitoring terminal.

[0094] The forest shade angle model is constructed in the following way:

[0095] After initial deployment, the edge processing module divides the azimuth angle into several continuous zones.

[0096] The monitoring terminal continuously records external signal information from historical communication events. The external signal information includes signal reception strength, as well as the azimuth and elevation angles of the signal source relative to the monitoring terminal.

[0097] Within each azimuth zone, the relationship between signal received strength and the corresponding elevation angle is statistically fitted to obtain the fitted curve;

[0098] The elevation angle corresponding to the critical point in the fitted curve where the signal reception strength begins to fall below a preset reliable communication threshold is determined as the critical communication angle for that azimuth zone.

[0099] Based on the critical communication angles corresponding to each azimuth zoning zone, a polar coordinate model centered on the monitoring terminal is constructed as the forest shading angle model.

[0100] In practice, each monitoring terminal enters an initial learning phase immediately after deployment. During this period, its edge processing module divides the horizontal 360-degree azimuth into 72 consecutive sectors, each sector being 5 degrees. Whenever it receives an external signal from a passing drone or other terminal, it records the received signal strength (RSSI value, in dBm) and calculates the azimuth and elevation angles of the source relative to itself based on the signal source coordinates.

[0101] After the initial learning period, the edge processing module performs statistical analysis on the data within each azimuth sector. For example, for records in the due north direction (0-5 degree sector), a scatter plot of signal received strength versus elevation angle is plotted, and curve fitting is performed using the least squares method. Assume that the fitted curve of a monitoring terminal shows that when the elevation angle is below 25 degrees, the signal strength is generally below -90dBm; when the elevation angle is above 30 degrees, the signal strength stabilizes above -70dBm. The edge processing module uses algorithms, such as finding abrupt changes in the curve slope or intersections with the threshold line, to determine that the critical communication azimuth angle for this sector is 28 degrees. This angle means that in this direction, the communication link elevation angle must be greater than 28 degrees to effectively avoid tree obstruction.

[0102] This process is repeated for all 72 sectors to obtain a series of azimuth-critical communication angle correspondence tables. Using the monitoring terminal itself as the origin, the azimuth as the polar angle, and the critical communication angle as the polar radius, a digital polar coordinate model, i.e., a forest shading angle model, can be constructed. This model is stored in the terminal for subsequent communication guidance. The model can be updated periodically to adapt to environmental changes caused by forest growth or logging.

[0103] Among them, the calculation of the theoretically optimal communication angle for communication with the UAV based on the three-dimensional spatial coordinates of the monitoring terminal and using a pre-stored forest shading angle model includes:

[0104] Based on the three-dimensional spatial coordinates of the monitoring terminal and the three-dimensional spatial coordinates of the UAV, the horizontal azimuth angle α, the horizontal geometric distance dh, and the height difference Δh of the UAV relative to the monitoring terminal are calculated.

[0105] Match the horizontal azimuth angle α to the tree shading angle model and output the critical communication angle β of the azimuth angle partition corresponding to α;

[0106] Based on the critical communication angle β, altitude difference Δh, horizontal geometric distance dh, and UAV flight attitude information, the theoretical optimal communication angle is solved through iterative calculation.

[0107] The calculation of the horizontal azimuth angle α, the horizontal geometric distance dh, and the height difference Δh of the UAV relative to the monitoring terminal is existing technology, and this invention will not elaborate on it further.

[0108] Specifically, the goal of the iterative calculation is to find a candidate beacon angle γ that simultaneously satisfies constraints A and B. c As the optimal perspective for the aforementioned theory:

[0109] Among them, constraint A: based on the candidate belief angle γ c With respect to the horizontal azimuth angle α, define a desired communication link from the monitoring terminal to the expected communication point in the air;

[0110] Based on the UAV's flight attitude information, the direction vector of the desired communication link is transformed from the ground coordinate system to the UAV's body coordinate system, and the actual effective communication angle γ of the desired communication link in the UAV's body coordinate system is calculated. e ;

[0111] Actual effective faith angle γ e Greater than the critical belief angle β;

[0112] Constraint B: In the UAV body coordinate system, the direction of the desired communication link is within the effective radiation main lobe range of the UAV onboard communication antenna in the current real-time flight attitude.

[0113] In practice, after obtaining the horizontal azimuth angle α, the horizontal geometric distance dh, the height difference Δh, and the critical beacon angle β, the edge processing module initiates an iterative solution program. It initializes a candidate beacon angle γ. c = β + 5°. Based on γ c And α, we can define a desired communication link vector. Combining the UAV attitude, we calculate the actual effective elevation angle γ of this link in the UAV body coordinate system using a coordinate transformation matrix. e Determine γ e Is γ greater than β, and determine if the link direction is within the main lobe of the UAV antenna. If both conditions are met, then γ... c This is the theoretically optimal angle of belief; if it is not satisfied, then increase γ. c (If the step size is 1°) Calculate again until a solution that meets the conditions is found, or the maximum search range is exceeded.

[0114] For example, if the candidate angle γ c =30°, α=45°.

[0115] Constraint A: First, in the ground coordinate system (NED coordinate system), construct the unit direction vector Vground = [cos(30°)cos(45°), cos(30°)sin(45°), -sin(30°)] pointing from the terminal to the desired UAV position. Calculate the rotation matrix R using the UAV's current attitude angles (assuming pitch angle φ=5°, roll angle θ=0°, yaw angle ψ=60°). Multiply Vground by R on the left to obtain the direction vector Vbody in the UAV body coordinate system (FRD coordinate system). Calculate the angle between Vbody and the horizontal plane (XY plane) of the UAV body to obtain the actual effective communication angle γ. e Assuming γ is calculated e = 28°. Due to γ e (28°) > β (25°), which satisfies constraint A, indicating that the communication link can avoid tree obstruction in space.

[0116] Constraint B: Calculate the angle between Vbody and the preset maximum gain direction of the UAV's onboard antenna (assumed to be the Z-axis of the aircraft, i.e., pointing towards the ground). Assume the antenna's half-power beamwidth (Θ) 3dB The angle is 60°. If the calculated included angle is 20°, it is less than Θ. 3dB If / 2, then it is determined that the desired communication link direction is within the effective radiation main lobe range of the antenna, satisfying constraint B.

[0117] Only when both A and B are satisfied can γ c Only then will it be adopted as the final theoretical optimal angle of belief. If γ c =30° does not satisfy B, so γ needs to be increased. c Recalculate until an angle that satisfies both conditions is found.

[0118] Example 2

[0119] Based on Example 1, the forest construction area is divided into several independent sub-monitoring areas. A unique main monitoring terminal and several auxiliary monitoring terminals are deployed in any sub-monitoring area, and the main monitoring terminal and the auxiliary monitoring terminals are connected in communication.

[0120] The method further includes:

[0121] The drone passes through all sub-monitoring areas sequentially according to the preset main cruise path;

[0122] Within any sub-monitoring area, the UAV establishes a connection with the main monitoring terminal of that sub-monitoring area and receives a local state set returned by the main monitoring terminal. The local state set includes the state information of the main monitoring terminal itself and its affiliated monitoring terminals.

[0123] Based on the local state set, the UAV updates its built-in value probability map, which is used to record the location coordinates of all monitoring terminals in the forest construction area, the comprehensive collection of value estimates and the confidence level of the estimates.

[0124] Based on the updated value probability map, its own real-time location, and remaining battery power, the drone plans a sub-cruise path covering multiple auxiliary monitoring terminals within the current sub-monitoring area.

[0125] The drone receives data information collected by the auxiliary monitoring terminal based on the sub-cruise path.

[0126] In practice, for example, a 1000-hectare forest area might be divided into 10 independent sub-monitoring areas. A main monitoring terminal is deployed at the center of each area, with 20 auxiliary monitoring terminals distributed throughout the area. The main and auxiliary monitoring terminals maintain a periodic heartbeat connection via a low-power wide-area network (such as LoRa), and the main terminal can aggregate the status of the auxiliary terminals.

[0127] Before the cruise begins, a main cruise path is established connecting the main monitoring terminals in each area, and the UAV flies along this main cruise path. When entering a sub-monitoring area, the UAV first establishes a connection with the main monitoring terminal in that area. After connection, the UAV not only receives data from the main terminal itself, but also sends a status query request to it. The main monitoring terminal then packages the status information of its subordinate terminals into a local status set and returns it to the UAV.

[0128] The onboard computer of the drone maintains a global value probability map. Upon receiving a local state set, the drone updates the map with the comprehensive collection value estimate and confidence level of all monitoring terminals (including main and auxiliary) in that sub-monitoring area.

[0129] After updating the map, the drone, based on its location and remaining battery power, uses a path planning algorithm to generate a sub-cruise path covering multiple high-value auxiliary terminals within the sub-region. The drone then detaches from the main terminal and flies along the sub-cruise path to each of these auxiliary terminals, establishing connections and collecting data. After completing data collection, the drone returns to the main cruise path and flies to the next area.

[0130] The status information includes: the remaining battery power of the monitoring terminal, the total amount of data cached, the time since the last data upload, and the amount of regular monitoring data and event-triggered data cached by the monitoring terminal;

[0131] The calculation method for the comprehensive collection value assessment includes:

[0132] The data urgency value is calculated based on the remaining battery power of the monitoring terminal, the total amount of data cache, and the time since the last data upload.

[0133] The inherent value of the data is calculated based on the amount of regular monitoring data and event-triggered data cached by the monitoring terminal.

[0134] The comprehensive collection value estimate is obtained by weighting and summing the data urgency value with the data intrinsic value.

[0135] When calculating the urgency value of the data, the remaining battery power, the total amount of data cached, and the time since the last data upload are normalized and then weighted for calculation, with weights set according to the actual situation. The calculation follows these principles: the lower the remaining battery power, the greater the urgency; the fuller the total data cache, the greater the urgency; and the longer the time interval since the last successful data upload, the greater the urgency. The specific calculation method is existing technology in this field, and this invention does not limit it.

[0136] When calculating the inherent value of data, for example, a monitoring terminal has 5 cached event trigger data entries (Devent = 5 entries, including 1 fire alarm Lv.3 and 4 pest and disease Lv.2 entries) and 100 regular data entries (Dregular = 100 entries).

[0137] If the weight of fire alarm Lv.3 is set to 10, the weight of pest and disease Lv.2 is set to 5, and the value of regular data unit is set to 0.1, then the inherent value of the monitoring terminal data = (1 * 10 + 4 * 5) + 100 * 0.1 = 30 + 10 = 40.

[0138] The valuation confidence level is updated in the following way:

[0139] Before the patrol mission begins, the drone assigns an initial confidence level C to any auxiliary monitoring terminal within the forest construction area. in ;

[0140] When the auxiliary monitoring terminal communicates directly with the drone, its corresponding estimated confidence level is updated to the preset highest confidence level C. max ;

[0141] When the status information of the auxiliary monitoring terminal is sent by the main monitoring terminal, the corresponding estimated confidence level is updated to the confidence level currently stored by the monitoring terminal and the preset indirect confidence level C. ind The larger value in, where C max >C ind >C in ;

[0142] For auxiliary monitoring terminals not involved in the local state set, the corresponding estimated confidence level is updated by multiplying it by a decay factor.

[0143] For example, assuming the reliability range is between 0 and 1, the preset C max = 1.0, C ind = 0.6, C in = 0.3, attenuation factor ε=0.9.

[0144] For example, when a drone receives status information about its auxiliary monitoring terminal Y from the main monitoring terminal in a certain sub-monitoring area, if the current confidence level of auxiliary monitoring terminal Y is 0.3, it is updated to max(0.3, C_ind)=0.6; if its current confidence level is 0.8 (from an earlier direct communication), it is updated to max(0.8, 0.6)=0.8.

[0145] Based on the updated value probability map, its own real-time location, and remaining battery power, a sub-cruise route is planned to cover multiple auxiliary monitoring terminals within the sub-monitoring area, including:

[0146] Based on the updated value probability map, the geographical locations, comprehensive collection value estimates, and estimate confidence levels of all auxiliary monitoring terminals within the current sub-monitoring area are extracted.

[0147] Using the connection point between the UAV and the current main monitoring terminal as the starting point of the sub-cruise path, the available cruise power budget within the current sub-monitoring area is calculated based on the UAV's real-time remaining power and the preset energy consumption per unit distance.

[0148] With the goal of maximizing the value of data recovered within the cruise power budget, and constrained by sequentially accessing multiple auxiliary monitoring terminals from the starting point and finally returning to the starting point, a sub-regional path optimization model is constructed.

[0149] Solve the sub-region path optimization model to obtain a sequential sequence of visits to multiple auxiliary monitoring terminals, which serves as the sub-cruise path.

[0150] For example, after the drone updates the value probability map at the main monitoring terminal in a certain sub-monitoring area, it extracts information on all 20 auxiliary monitoring terminals in that area: geographical location (P i ), comprehensive value valuation (V) i ), confidence level (C) i ).

[0151] Assume the drone's remaining battery power allows it to fly a distance of Dmax = 10 kilometers, and the energy consumption per unit distance is constant. The goal of path planning is to start from the main terminal location (starting point S), visit several auxiliary terminals, and return to S, maximizing the sum of the estimated comprehensive value of the visited terminals, while ensuring that the total cost of the path does not exceed Dmax.

[0152] In this embodiment, the total cost of the path is not simply the sum of geometric distances, but rather the accumulation of resources consumed by each terminal accessed, i.e., the estimated cost. This cost includes the flight distance and the decision-making risk premium due to the uncertainty of the target terminal information. Therefore, the planning problem can be formalized as a constrained combinatorial optimization problem:

[0153] Objective function: MaximizeΣ(V i * x i )

[0154] Constraints:

[0155] Total path cost ΣCost(j→i)≤Dmax;

[0156] The path starts from the starting point S and eventually returns to S.

[0157] Other path continuity constraints.

[0158] Where, x i The binary decision variable represents whether terminal i is accessed; Cost(j→i) is the estimated cost from the previous node j to the monitoring terminal i.

[0159] The UAV's onboard computer uses heuristic algorithms (such as simulated annealing) to solve the above model. When searching for access sequences, the algorithm continuously calculates the cumulative cost of the current partial path and compares it with Dmax to ensure that the power constraints are not violated. Finally, the algorithm outputs a terminal access sequence that maximizes the total value of the recovered data within the power budget, for example: S→Terminal 5→Terminal 12→Terminal 8→Terminal 1→S. This sequence is the sub-cruise path under the current power constraints.

[0160] When constructing the sub-regional path optimization model, the estimated cost of the UAV traveling to any auxiliary monitoring terminal is determined in the following way:

[0161] Calculate the Euclidean distance from the previous node on the path to the geographical location of the auxiliary monitoring terminal, and use it as the base distance cost;

[0162] Obtain the current valuation confidence level of the auxiliary monitoring terminal in the value probability map;

[0163] A confidence cost coefficient is calculated based on the current valuation confidence level, wherein the confidence cost coefficient is negatively correlated with the current valuation confidence level;

[0164] The estimated cost is obtained by multiplying the base distance cost by the confidence cost coefficient.

[0165] In this embodiment, Cost(j→i) is determined by two parts:

[0166] Base distance cost: This is the Euclidean distance Dist(j, i) from the previous node j on the path to the geographical location of the target monitoring terminal i. This part directly corresponds to the energy consumed by the UAV during flight.

[0167] Confidence Cost Coefficient: This coefficient is related to the current estimated confidence level C of the target monitoring terminal i. i Negative correlation. The coefficient is calculated using the formula (1 + K * (1 - C)). i ), where K is a preset adjustment coefficient (e.g., K=1). Confidence level C i The lower the value, the larger the coefficient.

[0168] Finally, the formula for calculating the estimated cost is:

[0169] Cost(j→i) = Dist(j, i) * (1 + K * (1 - C i ))

[0170] For example, monitoring terminal C is closer, but due to its extremely low confidence level, the cost of accessing it is higher than that of monitoring terminal D, which is farther away but has a higher confidence level. During path optimization, choosing terminal B can yield benefits with less cost, thus potentially allowing access to more high-value, high-confidence terminals within the limited Dmax, ultimately maximizing the total value.

[0171] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0172] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for offline data collection and intelligent synchronization under weak network conditions during national reserve forest construction, characterized in that: The method includes the following steps: Within the forest construction area, multiple monitoring terminals are deployed in a distributed manner according to site conditions and monitoring needs. Each monitoring terminal integrates monitoring sensors and an edge processing module. The monitoring sensor collects raw data, and the edge processing module processes the collected raw data, dividing it into regular monitoring data and event-triggered data based on the processing results. The drone cruises along a preset route. When the drone enters the theoretical communication range of the target monitoring terminal, the connection establishment process is initiated. Once the drone establishes a connection with the target monitoring terminal, the drone receives routine monitoring data and event-triggered data uploaded by the target monitoring terminal. After the data upload is complete, the monitoring terminal clears the cache of the synchronized data and records the timestamp of this upload.

2. The method for offline data acquisition and intelligent synchronization under weak network conditions during national reserve forest construction, as described in claim 1, is characterized in that... The connection establishment process includes: The drone transmits a broadcast detection signal containing its current three-dimensional spatial coordinates and flight attitude information; After the target monitoring terminal receives the broadcast detection signal, the edge processing module calculates the theoretically optimal communication angle with the UAV based on the three-dimensional spatial coordinates of the monitoring terminal and using a pre-stored tree shading angle model. The monitoring terminal sends the theoretically optimal communication angle to the drone; After receiving the theoretically optimal communication angle, the UAV adjusts its position so that the axis of its communication antenna points to match the theoretically optimal communication angle, and then initiates a formal connection request.

3. The method for offline data acquisition and intelligent synchronization under weak network conditions during national reserve forest construction, as described in claim 2, is characterized in that... The tree shading angle model is used to output the critical communication angle necessary to prevent tree shading under different azimuth angles of the monitoring terminal. The forest shade angle model is constructed in the following way: After initial deployment, the edge processing module divides the azimuth angle into several continuous zones. The monitoring terminal continuously records external signal information from historical communication events. The external signal information includes signal reception strength, as well as the azimuth and elevation angles of the signal source relative to the monitoring terminal. Within each azimuth zone, the relationship between signal received strength and the corresponding elevation angle is statistically fitted to obtain the fitted curve; The elevation angle corresponding to the critical point in the fitted curve where the signal reception strength begins to fall below a preset reliable communication threshold is determined as the critical communication angle for that azimuth zone. Based on the critical communication angles corresponding to each azimuth zoning zone, a polar coordinate model centered on the monitoring terminal is constructed as the forest shading angle model.

4. The method for offline data acquisition and intelligent synchronization under weak network conditions during national reserve forest construction, as described in claim 3, is characterized in that... Based on the three-dimensional spatial coordinates of the monitoring terminal, and using a pre-stored forest shading angle model, the theoretically optimal communication angle for communication with the UAV is calculated, including: Based on the three-dimensional spatial coordinates of the monitoring terminal and the UAV, the horizontal azimuth angle α and the horizontal geometric distance d of the UAV relative to the monitoring terminal are calculated. h and height difference Δh; Match the horizontal azimuth angle α to the tree shading angle model and output the critical communication angle β of the azimuth angle partition corresponding to α; Based on the critical angle of faith β, the height difference Δh, and the geometric distance d on the horizontal plane h The theoretical optimal communication angle is obtained by iteratively calculating the flight attitude information of the UAV.

5. The method for offline data acquisition and intelligent synchronization in a weak network environment during national reserve forest construction, as described in claim 4, is characterized in that... The goal of the iterative calculation is to find a candidate beacon angle γ that simultaneously satisfies constraints A and B. c As the optimal perspective for the aforementioned theory: Among them, constraint A: based on the candidate belief angle γ c With respect to the horizontal azimuth angle α, define a desired communication link from the monitoring terminal to the expected communication point in the air; Based on the UAV's flight attitude information, the direction vector of the desired communication link is transformed from the ground coordinate system to the UAV's body coordinate system, and the actual effective communication angle γ of the desired communication link in the UAV's body coordinate system is calculated. e ; Actual effective communication angle γ e Greater than the critical belief angle β; Constraint B: In the UAV body coordinate system, the direction of the desired communication link is within the effective radiation main lobe range of the UAV onboard communication antenna in the current real-time flight attitude.

6. The method for offline data acquisition and intelligent synchronization under weak network conditions during national reserve forest construction, as described in claim 1, is characterized in that... The forest construction area is divided into several independent sub-monitoring areas. In any sub-monitoring area, a single main monitoring terminal and several auxiliary monitoring terminals are deployed, and the main monitoring terminal and the auxiliary monitoring terminals are connected to each other. The method further includes: The drone passes through all sub-monitoring areas sequentially according to the preset main cruise path; Within any sub-monitoring area, the UAV establishes a connection with the main monitoring terminal of that sub-monitoring area and receives a local state set returned by the main monitoring terminal. The local state set includes the state information of the main monitoring terminal itself and its affiliated monitoring terminals. Based on the local state set, the UAV updates its built-in value probability map, which is used to record the location coordinates of all monitoring terminals in the forest construction area, the comprehensive collection of value estimates and the confidence level of the estimates. Based on the updated value probability map, its own real-time location, and remaining battery power, the drone plans a sub-cruise path covering multiple auxiliary monitoring terminals within the current sub-monitoring area. The drone receives data information collected by the auxiliary monitoring terminal based on the sub-cruise path.

7. The method for offline data acquisition and intelligent synchronization in a weak network environment during national reserve forest construction, as described in claim 6, is characterized in that... The status information includes: the remaining battery power of the monitoring terminal, the total amount of data cached, the time since the last data upload, and the amount of regular monitoring data and event-triggered data cached by the monitoring terminal; The calculation method for the comprehensive collection value assessment includes: The data urgency value is calculated based on the remaining battery power of the monitoring terminal, the total amount of data cache, and the time since the last data upload. The inherent value of the data is calculated based on the amount of regular monitoring data and event-triggered data cached by the monitoring terminal. The comprehensive collection value estimate is obtained by weighting and summing the data urgency value with the data intrinsic value.

8. The method for offline data acquisition and intelligent synchronization in a weak network environment during national reserve forest construction, as described in claim 6, is characterized in that... The valuation confidence level is updated in the following ways: Before the patrol mission begins, the drone assigns an initial confidence level C to any auxiliary monitoring terminal within the forest construction area. in ; When the auxiliary monitoring terminal communicates directly with the drone, its corresponding estimated confidence level is updated to the preset highest confidence level C. max ; When the status information of the auxiliary monitoring terminal is sent by the main monitoring terminal, the corresponding estimated confidence level is updated to the confidence level currently stored by the monitoring terminal and the preset indirect confidence level C. ind The larger value in, where C max >C ind >C in ; For auxiliary monitoring terminals not involved in the local state set, the corresponding estimated confidence level is updated by multiplying it by a decay factor.

9. The method for offline data acquisition and intelligent synchronization in a weak network environment during national reserve forest construction, as described in claim 6, is characterized in that... Based on the updated value probability map, its own real-time location, and remaining battery power, a sub-cruise route is planned to cover multiple auxiliary monitoring terminals within the sub-monitoring area, including: Based on the updated value probability map, the geographical locations, comprehensive collection value estimates, and estimate confidence levels of all auxiliary monitoring terminals within the current sub-monitoring area are extracted. Using the connection point between the UAV and the current main monitoring terminal as the starting point of the sub-cruise path, the available cruise power budget within the current sub-monitoring area is calculated based on the UAV's real-time remaining power and the preset energy consumption per unit distance. With the goal of maximizing the value of data recovered within the cruise power budget, and constrained by sequentially accessing multiple auxiliary monitoring terminals from the starting point and finally returning to the starting point, a sub-regional path optimization model is constructed. Solve the sub-region path optimization model to obtain a sequential sequence of visits to multiple auxiliary monitoring terminals, which serves as the sub-cruise path.

10. The method for offline data acquisition and intelligent synchronization under weak network conditions during national reserve forest construction, as described in claim 9, is characterized in that... When constructing the sub-regional path optimization model, the estimated cost of the UAV traveling to any auxiliary monitoring terminal is determined in the following way: Calculate the Euclidean distance from the previous node on the path to the geographical location of the auxiliary monitoring terminal, and use it as the base distance cost; Obtain the current valuation confidence level of the auxiliary monitoring terminal in the value probability map; A confidence cost coefficient is calculated based on the current valuation confidence level, wherein the confidence cost coefficient is negatively correlated with the current valuation confidence level; The estimated cost is obtained by multiplying the base distance cost by the confidence cost coefficient.

Images