Power Inspection Track Management and Dispatch System Based on BeiDou Positioning

Abstract

This application discloses a power line inspection trajectory management and scheduling system based on BeiDou positioning. It utilizes magnetic field strength as an adaptive adjustment factor to construct an electromagnetic sensing-based BeiDou trajectory smoothing model, dynamically adjusting the weights of positioning data to eliminate false jumps. Simultaneously, by calculating the effective work trajectory after noise removal and combining it with real-time current calculations of energy efficiency parameters, a risk prediction mechanism deeply coupled with trajectory quality and energy consumption is established. In this way, the system can effectively suppress system deviations caused by potential drift, achieving accurate quantification of the remaining task costs under strong interference environments. This not only significantly reduces the risk of power outages and power plant crashes due to optimistic energy consumption predictions but also ensures the continuity and safety of inspection operations through intelligent task trimming and scheduling instruction optimization.

Description

Technical Field

[0001] This application relates to the field of intelligent scheduling, and more specifically, to a power inspection trajectory management and scheduling system based on BeiDou positioning. Background Technology

[0002] With the widespread application of drone technology in power grid inspection, building an efficient and reliable inspection trajectory management and scheduling solution has become a key link in ensuring the safe and stable operation of the power grid. This demand stems from the limitations of traditional manual inspection methods in terms of efficiency, accuracy, and safety, as well as the urgent need for automated and refined monitoring of large-scale transmission line networks.

[0003] Currently, various power line inspection systems utilize global navigation satellite systems (such as BeiDou) for UAV positioning and navigation, and employ trajectory planning and energy management modules for task scheduling. However, existing solutions face significant challenges in dealing with the complex electromagnetic environment of ultra-high-voltage transmission lines. Particularly in high-voltage electromagnetic environments, reflections from the tower's metal structure and strong electromagnetic fields can cause multipath effects in BeiDou signals, leading to periodic drift and ambiguity in positioning data, resulting in distorted trajectory smoothing results. This further causes systematic biases in the scheduling system's estimation of remaining range and energy consumption based on distorted trajectory length and quality. A deeper technical bottleneck lies in the fact that existing filtering models are mostly static or general designs, unable to perceive the spatial distribution and intensity changes of interference sources, leading to amplified errors near the tower. Simultaneously, trajectory processing and energy consumption prediction modules are often decoupled, ignoring the additional energy consumption caused by high-frequency attitude adjustments of the UAV due to positioning drift, resulting in overly optimistic energy predictions. These problems not only affect inspection efficiency but can also lead to misjudgments of energy levels, mission interruptions, and even flight accidents, failing to meet the stringent requirements for inspection operation safety and scheduling reliability in environments with strong interference.

[0004] Therefore, we look forward to an optimized power line inspection trajectory management and scheduling system based on BeiDou positioning. Summary of the Invention

[0005] To address the aforementioned technical issues, this application provides a power line inspection trajectory management and scheduling system based on BeiDou positioning.

[0006] According to one aspect of this application, a power line inspection trajectory management and scheduling system based on BeiDou positioning is provided, comprising: The data acquisition module is used to acquire raw sensor data streams containing BeiDou raw observation data, airborne magnetometer data, airborne inertial navigation data, and battery status data. The data alignment module is used to perform time synchronization interpolation and multi-source data alignment on the original sensor data stream using the BeiDou timing signal as a reference to obtain an aligned state dataset. The smoothing filter module is used to perform adaptive BeiDou trajectory smoothing filter based on electromagnetic sensing on the BeiDou coordinate data in the aligned state dataset based on the magnetic field strength data in the aligned state dataset to obtain smooth trajectory points. The real-time energy efficiency parameter determination module is used to determine real-time energy efficiency parameters based on the instantaneous current data in the smooth trajectory points and the aligned state dataset. The task trajectory cost prediction module is used to calculate the remaining inspection path based on the smoothed trajectory points and the preset list of remaining inspection waypoints, and to predict the task trajectory cost of the remaining inspection path based on real-time energy efficiency parameters to obtain the scheduling risk index. The scheduling instruction generation module is used to compare the scheduling risk index with the preset risk threshold, and generate the final scheduling instruction containing actions such as continue execution, task pruning, or emergency return based on the comparison results.

[0007] Compared with existing technologies, this application provides a BeiDou-based power line inspection trajectory management and scheduling system. It utilizes magnetic field strength as an adaptive adjustment factor to construct an electromagnetic sensing-based BeiDou trajectory smoothing model, dynamically adjusting the weights of positioning data to eliminate false jumps. Simultaneously, by calculating the effective work trajectory after noise removal and combining it with real-time current calculations of energy efficiency parameters, a risk prediction mechanism deeply coupled with trajectory quality and energy consumption is established. In this way, the system can effectively suppress system deviations caused by potential drift, achieving accurate quantification of the remaining task costs under strong interference environments. This not only significantly reduces the risk of power outages and crashes due to optimistic energy consumption predictions but also ensures the continuity and safety of inspection operations through intelligent task trimming and scheduling instruction optimization. Attached Figure Description

[0008] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0009] Figure 1 This is a block diagram of a power line inspection trajectory management and scheduling system based on BeiDou positioning according to an embodiment of this application; Figure 2 This is a schematic diagram of the data flow of a power line inspection trajectory management and scheduling system based on BeiDou positioning according to an embodiment of this application; Figure 3 This is a block diagram of the data alignment module in a power line inspection trajectory management and scheduling system based on BeiDou positioning according to an embodiment of this application; Figure 4This is a block diagram of the real-time energy efficiency parameter determination module in the BeiDou-based power inspection trajectory management and scheduling system according to an embodiment of this application. Detailed Implementation

[0010] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0011] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0012] While this application makes various references to certain modules of the systems according to embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The modules described are merely illustrative, and different aspects of the systems and methods may use different modules.

[0013] Flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0014] The technical solution of this application proposes a power line inspection trajectory management and scheduling system based on BeiDou positioning. Figure 1 This is a block diagram of a power line inspection trajectory management and scheduling system based on BeiDou positioning according to an embodiment of this application. Figure 2 This is a schematic diagram of the data flow in a power line inspection trajectory management and scheduling system based on BeiDou positioning, according to an embodiment of this application. Figure 1 and Figure 2As shown, the power line inspection trajectory management and scheduling system 300 based on BeiDou positioning according to an embodiment of this application includes: a data acquisition module 310, used to acquire raw sensor data streams containing raw BeiDou observation data, airborne magnetometer data, airborne inertial navigation data, and battery status data; a data alignment module 320, used to perform time synchronization interpolation and multi-source data alignment on the raw sensor data streams using BeiDou timing signals as a reference to obtain an aligned state dataset; and a smoothing filtering module 330, used to perform adaptive filtering based on electromagnetic sensing on the BeiDou coordinate data in the aligned state dataset based on the magnetic field strength data in the aligned state dataset. The BeiDou trajectory smoothing filter is used to obtain smooth trajectory points; the real-time energy efficiency parameter determination module 340 is used to determine real-time energy efficiency parameters based on the smooth trajectory points and the instantaneous current data in the aligned state dataset; the mission trajectory cost prediction module 350 is used to calculate the remaining inspection path based on the smooth trajectory points and the preset remaining inspection waypoint list, and to predict the mission trajectory cost of the remaining inspection path based on the real-time energy efficiency parameters to obtain the scheduling risk index; the scheduling instruction generation module 360 ​​is used to compare the scheduling risk index with the preset risk threshold, and to generate the final scheduling instruction containing the actions of continuing execution, mission pruning or emergency return based on the comparison result.

[0015] Specifically, the data acquisition module 310 is used to acquire a raw sensor data stream containing BeiDou raw observation data, airborne magnetometer data, airborne inertial navigation data, and battery status data. It should be understood that when power line inspection drones traverse strong electromagnetic environments such as ultra-high voltage transmission lines, single sensor data is highly susceptible to environmental interference, resulting in severe observational biases. To achieve high-precision trajectory management and intelligent scheduling decisions, the technical solution of this application constructs a multi-dimensional, heterogeneous, and complementary sensing foundation. BeiDou raw observation data provides a spatiotemporal reference for the global navigation satellite system; airborne magnetometer data is used to sense the surrounding electromagnetic field intensity distribution; airborne inertial navigation data provides pose compensation through high-frequency sampling; and battery status data is the energy cornerstone for quantifying mission energy efficiency. By integrating these data into a unified raw sensor data stream, essential raw physical quantity inputs are provided for subsequent electromagnetic sensing filtering, real-time energy efficiency calculation, and path cost prediction, thereby fundamentally solving the problems of positioning offset and inaccurate energy consumption prediction in complex environments.

[0016] In practice, first, after confirming that the system is powered on, initialize the data communication interfaces of the Beidou positioning receiver, the airborne three-axis magnetometer, the airborne inertial navigation system (usually including a gyroscope and accelerometer), and the battery management system (BMS), such as UART, I2C, or SPI, to ensure that each sensor is ready and starts outputting data.

[0017] Next, the system reads data packets from the initialized sensor interfaces in parallel using multi-threading or interrupt service routines. The reading of each type of data packet is independent to ensure maximum real-time performance and prevent delays in reading data from one sensor from affecting the timeliness of other data. During this process, the system continuously acquires raw BeiDou observation data containing pseudorange and carrier phase information from the BeiDou receiver, three-axis magnetic field component data from the airborne magnetometer, angular velocity and acceleration data from the airborne inertial navigation system, and battery status data such as battery voltage, instantaneous current, and remaining charge from the BMS. After being read, this data is temporarily stored in its respective circular buffer.

[0018] Then, the system marks each data packet with a timestamp generated by the system clock when it is successfully read and stored in the buffer. Subsequently, a data acquisition management thread retrieves data packets from each buffer according to a certain strategy (such as first-in, first-out) and combines them into a structured data frame or data packet, which is a basic unit of the raw sensor data stream.

[0019] Specifically, the data alignment module 320 is used to perform time synchronization interpolation and multi-source data alignment on the original sensor data stream using the BeiDou timing signal as a reference to obtain an aligned state dataset. It should be understood that various sensors carried by UAVs, such as BeiDou receivers, airborne inertial navigation systems, and magnetometers, often have inherent differences in their physical hardware frequencies and sampling times, and their time scales are not entirely the same. If unaligned asynchronous data is used directly, the positioning trajectory will not be able to match attitude oscillations, energy consumption, and environmental electromagnetic interference, leading to severe phase differences in subsequent smoothing filtering and causing drastic fluctuations in energy efficiency prediction. Therefore, in the technical solution of this application, a unified temporal and spatial reference is established to resample multi-source sensor observations onto the BeiDou timing sequence, forming a strictly time-aligned dataset to ensure the accuracy of subsequent algorithms.

[0020] Figure 3 This is a block diagram of the data alignment module in a BeiDou-based power line inspection trajectory management and scheduling system according to an embodiment of this application. Figure 3As shown, the data alignment module 320 includes: a multi-source data synchronization unit 321, used to parse the GPS week and second from the BeiDou observation data of the original sensor data stream as the system master clock, and to perform linear interpolation resampling on the airborne inertial navigation data and airborne magnetometer data in the original sensor data stream based on the discrete time sequence of the system master clock to obtain synchronized multi-source data; a data enhancement unit 322, used to extract the three-axis magnetic field components in the synchronized multi-source data to calculate the magnetic field strength scalar, and subtract a preset geomagnetic background reference value to obtain the environmental electromagnetic interference intensity characteristics, and to merge the environmental electromagnetic interference intensity characteristics as a new dimension into the synchronized multi-source data to obtain an enhanced intermediate data packet; and a dimensional standardization unit 323, used to standardize the dimensions of each physical quantity in the enhanced intermediate data packet to obtain an aligned state dataset.

[0021] Specifically, the multi-source data synchronization unit 321 is used to extract the GPS week and second from the BeiDou observation data in the original sensor data stream as the system master clock, and to perform linear interpolation resampling on the airborne inertial navigation data and airborne magnetometer data in the original sensor data stream based on the discrete time sequence of the system master clock to obtain synchronized multi-source data. In this process, firstly, the GPS week and second are extracted from the BeiDou observation data in the original sensor data stream as the system master clock. This time information comes from the atomic clock of the BeiDou satellite and has extremely high accuracy and stability. Next, based on the discrete time sequence of the system master clock, linear interpolation resampling is performed on the airborne inertial navigation data and airborne magnetometer data in the original sensor data stream, which typically have higher frequencies. For any sensor data stream requiring interpolation, at each master clock time point, the two data points immediately before and after that time in the sensor data stream are found. Then, applying the calculation principle of linear interpolation, that is, assuming that the data changes linearly with time between the two known points, the synchronized data value of the sensor at the reference time is calculated proportionally based on the time position of the target time relative to these two known points. This operation aligns all sensor data to the BeiDou time reference, ultimately resulting in synchronized multi-source data.

[0022] Specifically, the data augmentation unit 322 is used to extract the three-axis magnetic field components from the synchronous multi-source data to calculate the magnetic field strength scalar, and subtract a preset geomagnetic background reference value to obtain the environmental electromagnetic interference intensity characteristics. These environmental electromagnetic interference intensity characteristics are then incorporated into the synchronous multi-source data as a new dimension to obtain an enhanced intermediate data packet. This step aims to extract new features characterizing the environmental state from the synchronous data. In this process, firstly, the three-axis magnetic field components are extracted from the synchronous multi-source data, and according to the calculation logic of the magnetic field strength scalar, the square root of the sum of the squares of the three axial components is taken to calculate the magnetic field strength scalar value. Next, this scalar value is subtracted from the preset geomagnetic background reference value, which represents the local magnetic field strength under interference-free conditions, thereby obtaining the environmental electromagnetic interference intensity characteristics. This characteristic value directly reflects the magnetic field disturbance intensity of the current environment relative to the clean background. Furthermore, the calculated environmental electromagnetic interference intensity characteristics are incorporated into the original synchronous multi-source data as a new data dimension, thus forming a more information-rich enhanced intermediate data packet.

[0023] Specifically, the dimensional standardization unit 323 is used to standardize the dimensions of each physical quantity in the enhanced intermediate data packet to obtain an aligned state dataset. This step aims to eliminate the impact of differences in dimensions and orders of magnitude between different physical quantities on subsequent algorithms. In this process, for each physical quantity in the dataset, the mean and standard deviation of the physical quantity for all samples are first calculated. Then, for each sample value, the mean is subtracted and the value is divided by the standard deviation. After this step, the values ​​of each dimension of the data will become dimensionless standard scores with a mean of approximately zero and a standard deviation of one. The data after this step constitutes the final aligned state dataset that can be used for subsequent modules. This dataset not only has strictly aligned timestamps and includes environmental electromagnetic interference characteristics, but also has data in each dimension on the same numerical scale.

[0024] Specifically, the smoothing filter module 330 is used to perform adaptive BeiDou trajectory smoothing filtering based on electromagnetic sensing on the BeiDou coordinate data in the aligned state dataset, based on the magnetic field strength data in the aligned state dataset, to obtain smooth trajectory points. It should be understood that the strong magnetic field environment around ultra-high voltage transmission lines and the multipath effect caused by the metal structure of the tower will cause the measurement noise of the BeiDou satellite navigation system to exhibit non-stationary attenuation characteristics, reflected in the original coordinates as false jumps of several meters or even tens of meters. Traditional static filtering models, due to their fixed parameters, cannot perceive the current electromagnetic risk level of the UAV, resulting in an inability to effectively suppress drift when approaching interference sources, thus leading to cumulative trajectory length deviations and misjudgments of energy efficiency. Therefore, in the technical solution of this application, by performing adaptive filtering based on electromagnetic sensing, the system can dynamically adjust the trust level of BeiDou data according to the perceived magnetic field strength. When the interference is strong, it relies more on inertial navigation data for estimation and prediction; when the interference is weak, it uses BeiDou data for correction, thereby outputting a smooth trajectory that truly reflects the actual posture of the UAV, laying a high-precision foundation for subsequent energy efficiency management.

[0025] In practice, the electromagnetic interference intensity characteristics in the aligned state dataset are first analyzed. A dynamic expansion coefficient is calculated based on a pre-defined baseline observation noise covariance matrix, and an adaptive observation noise covariance matrix is ​​constructed where the diagonal elements increase non-linearly with the interference intensity. Specifically, by analyzing the electromagnetic interference intensity characteristics in the aligned state dataset and using a pre-defined baseline observation noise covariance matrix (which defines the noise statistics of BeiDou coordinate observations under interference-free conditions), a dynamic expansion coefficient that varies with the interference intensity is calculated. A typical calculation logic is to make this coefficient proportional to the square of the interference intensity; that is, doubling the interference intensity will amplify the noise variance by four times, thus achieving a non-linear response. Then, an adaptive observation noise covariance matrix is ​​constructed. Specifically, the values ​​of the diagonal elements in this matrix (which directly represent the system's uncertainty estimate of the position observations) are dynamically and non-linearly increased with the increase of electromagnetic interference intensity. This construction process implies that the stronger the electromagnetic interference perceived by the system, the greater the perceived noise in the BeiDou coordinate observations, and the lower the trust weight given to it in subsequent filtering calculations.

[0026] Next, using the optimal estimated state from the previous moment in the aligned state dataset and inertial navigation data, a state transition equation is constructed to deduce the prior state estimation vector and prior error covariance matrix at the current moment, thus obtaining the prior state prediction packet. In other words, the system's internal model (state transition equation) is used to predict the current state. In this process, firstly, the optimal estimated state from the previous moment in the aligned state dataset (containing information such as position and velocity) and the current airborne inertial navigation data (such as acceleration and angular velocity) are used to construct the state transition equation. Then, based on the laws of physical motion (e.g., the current velocity combined with the time interval can predict the position at the next moment, and the current acceleration can update the velocity), the prior state estimate of the system at the current moment and the error covariance matrix of this prediction are calculated. The obtained prior state estimate and its error covariance together constitute the prior state prediction packet, which represents the best prediction of the current state based on the system's internal model, but has not yet incorporated current actual observation information.

[0027] Furthermore, the Kalman gain matrix is ​​calculated using the adaptive observation noise covariance matrix and the prior state prediction packet. Then, the prior state estimation vector is updated and corrected using measurements extracted from the aligned state dataset to obtain smooth trajectory points. In this process, firstly, the Kalman gain matrix is ​​calculated based on the adaptive observation noise covariance matrix and the prior error covariance matrix. This matrix is ​​essentially a weight allocator: when the reliability of the observations is high (i.e., the observation noise covariance matrix value is small) and the reliability of the predictions is relatively low (i.e., the prediction error covariance matrix value is large), the gain matrix will be larger, making the filter more inclined to believe the observations to correct the predictions; conversely, when the observation noise is large (i.e., the adaptive observation noise covariance matrix increases significantly due to strong interference), the calculated gain matrix will be smaller. Next, the prior state estimation vector is updated and corrected using measurements extracted from the aligned state dataset to obtain smooth trajectory points. In practice, the previous predicted state is used as a baseline, plus a correction increment consisting of the product of the Kalman gain and the observation residual, where the observation residual represents the difference between the actual acquired BeiDou coordinates and the predicted position. In this way, the system achieves accurate correction of pose estimation under different electromagnetic pressures.

[0028] Specifically, the real-time energy efficiency parameter determination module 340 is used to determine real-time energy efficiency parameters based on instantaneous current data in the smoothed trajectory points and aligned state dataset. It should be understood that in strong electromagnetic environments, power line inspection drones perform high-frequency attitude corrections and speed compensations to combat positioning drift and airflow disturbances. This frequent noise reduction consumes a significant amount of electrical energy, but this wasted effort is often ignored in traditional scheduling schemes. If the smoothed true displacement state is not correlated with real-time current, voltage, and attitude oscillations, scheduling based solely on the remaining battery percentage will lead to an overestimation of the remaining inspection path's endurance. Therefore, calculating real-time energy efficiency parameters reflecting the true power consumption per unit distance through this step can eliminate ineffective energy efficiency interference, providing a high-precision dynamic energy consumption benchmark for the scheduling system, thereby preventing mid-air power outages and crashes.

[0029] Figure 4 This is a block diagram of the real-time energy efficiency parameter determination module in a BeiDou-based power line inspection trajectory management and scheduling system according to an embodiment of this application. Figure 4 As shown, the real-time energy efficiency parameter determination module 340 includes: an instantaneous power decomposition and attitude oscillation loss quantization unit 341, used to perform instantaneous power decomposition and attitude oscillation loss quantization on the battery voltage, instantaneous current and angular velocity data in the aligned state dataset to obtain a power feature data packet; an effective displacement velocity calculation unit 342, used to calculate the effective displacement velocity of the smooth trajectory points to obtain an effective velocity scalar; and a real-time energy efficiency parameter calculation unit 343, used to subtract a preset theoretical hovering power from the instantaneous total power in the power feature data packet to obtain the working power, and calculate the ratio of the working power to the effective velocity scalar to obtain the real-time energy efficiency parameter.

[0030] Specifically, the instantaneous power decomposition and attitude oscillation loss quantization unit 341 is used to perform instantaneous power decomposition and attitude oscillation loss quantization on the battery voltage, instantaneous current, and angular velocity data in the aligned state dataset to obtain a power feature data packet. This step aims to identify the portion of the total power consumption related to effective displacement. In this process, the battery voltage, instantaneous current, and angular velocity data in the aligned state dataset are processed. The instantaneous total power is equal to the product of the battery voltage and the instantaneous current. To quantify energy consumption more precisely, the system also estimates the attitude oscillation loss. Specifically, it is assumed that maintaining drastic attitude adjustments (manifested as high angular velocity) requires additional power, which is proportional to the square of the angular velocity. By subtracting the estimated attitude oscillation loss from the total power, a more accurate effective power component for generating displacement can be obtained. These calculated power values ​​collectively constitute the power feature data packet.

[0031] Specifically, the effective displacement velocity calculation unit 342 is used to calculate the effective displacement velocity of the smoothed trajectory points to obtain an effective velocity scalar. This step aims to extract a scalar velocity reflecting the actual speed of the UAV's movement from the smoothed trajectory points. Specifically, the motion of the UAV in three-dimensional space can be decomposed into velocity components in three directions: north, east, and sky. The effective velocity scalar is the square root of the sum of the squares of these three velocity components, i.e., the magnitude of the three-dimensional velocity vector. This scalar velocity value comprehensively reflects the overall movement rate of the UAV in space and serves as the benchmark for evaluating its displacement efficiency. Specifically, the effective displacement velocity of the smoothed trajectory points is calculated using the following formula: in, , and These are the three-axis velocity components in the smooth trajectory points.

[0032] Specifically, the real-time energy efficiency parameter calculation unit 343 is used to subtract a preset theoretical hovering power from the instantaneous total power in the power feature data packet to obtain the working power, and to calculate the ratio of the working power to the effective velocity scalar to obtain the real-time energy efficiency parameter. In this process, firstly, the instantaneous total power is read from the power feature data packet, and the preset theoretical hovering power is read from the pre-stored configuration parameters. This theoretical hovering power is a constant value obtained through ground experiment calibration, representing the average power consumption of the UAV when hovering stationary in a windless environment under standard atmospheric conditions. Next, the instantaneous total power is subtracted from the theoretical hovering power to obtain the working power. Here, the working power represents the additional power output by the UAV's power system at the current moment after exceeding the requirements for maintaining basic hovering. This power is mainly used to drive the UAV to generate acceleration, overcome air resistance for horizontal movement, and adjust its attitude in response to external interference (such as wind and electromagnetic interference). Then, the effective velocity scalar is obtained, and the ratio of the working power to the effective velocity scalar is calculated to obtain the final real-time energy efficiency parameter. In other words, the power consumption per unit displacement rate is obtained by dividing the extra power consumed to generate displacement by the displacement rate achieved by that power. This ratio is the real-time energy efficiency parameter defined by the system. It intuitively represents the extra power consumed by the UAV (exceeding the hovering base) to achieve a unit speed of forward motion per meter per second under the current flight conditions. The lower the parameter value, the higher the flight energy efficiency; conversely, a higher parameter value indicates that the current flight state is inefficient, such as during acceleration, headwind flight, or anti-vibration maneuvers under strong electromagnetic interference.

[0033] Specifically, the task trajectory cost prediction module 350 is used to calculate the remaining inspection path based on the smoothed trajectory points and a preset list of remaining inspection waypoints, and to predict the task trajectory cost of the remaining inspection path based on real-time energy efficiency parameters to obtain a scheduling risk index. It should be understood that traditional power line inspection systems typically make return-to-base decisions based on the percentage of remaining battery power under ideal conditions or simple straight-line distances. This model ignores the dynamic variables in the inspection environment. In actual operation, the electromagnetic interference intensity varies across different road sections, and the battery's voltage support capacity and risk tolerance exhibit a significant non-linear decrease as the depth of discharge increases. By executing this step, the system can utilize the real energy efficiency parameters calculated from the previous steps to simulate energy loss for paths that have not yet been executed, thereby predicting the potential risk of an in-flight crash. This cost prediction for future trajectories organically combines electromagnetic environmental pressure, UAV dynamic response, and battery chemical characteristics, providing the scheduling system with a proactive risk indicator to ensure that inspection tasks operate efficiently within safe thresholds.

[0034] In practice, firstly, using the current coordinates in the smoothed trajectory point as the starting point, the remaining waypoints in the list of unvisited waypoints and the final return point are connected sequentially to calculate the total Euclidean distance of the remaining inspection path. The remaining inspection path refers to the topological line of the UAV starting from the current adaptively smoothed physical coordinates, traversing the remaining waypoint sequence, and finally returning to the base. During this process, the component differences between adjacent waypoints in the three-axis coordinate system are continuously extracted from the path sequence. The squares of the differences on each axis are calculated and summed. Then, the square root of this sum of squares is taken to obtain the spatial straight-line distance between adjacent waypoints. Finally, all segmented distances from the current point to the return point are accumulated to obtain the total geometric length of the entire path.

[0035] Next, the system identifies segments of the remaining inspection path that cross high-interference areas and applies a safety redundancy coefficient. It then uses the unit effective displacement energy consumption coefficient from real-time energy efficiency parameters and the hovering reference power to perform a weighted integral on the total Euclidean distance to obtain the predicted total power consumption. In this process, the system first identifies segments of the remaining inspection path known to cross high electromagnetic interference areas (e.g., segments near specific high-voltage towers) and applies a safety redundancy coefficient greater than 1 to these segments to simulate the additional distance or time loss caused by avoidance and cautious flight. Then, it uses real-time energy efficiency parameters (physically defined as the unit effective displacement energy consumption coefficient, i.e., the extra energy consumed per meter of movement) and the hovering reference power to perform a weighted integral on the total Euclidean distance to obtain the predicted total power consumption. In practice, the total energy consumption is decomposed into two parts: one is the mobile energy consumption, which is proportional to the flight distance and is calculated by multiplying the real-time energy efficiency parameter by the length of the corresponding route segment; the other is the hovering energy consumption, which is proportional to the flight time and is calculated by multiplying the hovering power by the estimated time required to complete the route segment (segment length divided by the estimated effective speed of the segment). For segments identified as high-risk, both parts of the energy consumption are further multiplied by a safety redundancy factor for amplification. Finally, the energy consumption of these two parts for all segments along the entire path is calculated separately and then summed to obtain the total predicted power demand.

[0036] Next, the current remaining battery power in the aligned state dataset is obtained, and the ratio of the predicted total required battery power to the current remaining battery power is calculated. A battery discharge depth penalty factor is then introduced to perform a nonlinear correction to obtain the scheduling risk index. In this process, firstly, a simple ratio of the predicted total required battery power to the current remaining battery power is calculated. Then, a battery discharge depth penalty factor is introduced to nonlinearly correct this original ratio to obtain the final scheduling risk index. Considering the physical characteristics of lithium batteries, in the later stages of discharge, the actual energy that can be released at the same remaining battery power percentage decreases; that is, the deeper the battery discharge, the lower its usable energy efficiency. Therefore, the penalty factor is designed as a function related to the current remaining battery power: when the remaining battery power is high, the penalty factor is close to 1, having little impact on the original ratio; when the remaining battery power is low, the penalty factor is significantly greater than 1, thus amplifying the original ratio and making the calculated risk index higher, reflecting the risk that the actual driving range at low battery power is weaker than the nominal value. The final result after this nonlinear correction is the scheduling risk index.

[0037] It is worth noting that the power demand weighted forecasting method based on real-time energy efficiency parameters provided in the above implementation scheme only introduces a simple scalar coefficient when predicting the remaining power consumption for tasks. To linearly amplify energy consumption in high-risk areas, and roughly assume that the drone maintains a fixed average cruising speed throughout the entire flight. Flight. In the actual scenario of inspecting ultra-high voltage transmission lines, there is a strong nonlinear coupling effect between electromagnetic fields and UAV dynamics. This simplified treatment leads to serious prediction bias. First, the prediction method fails to capture the negative correlation between field strength and flight speed. Specifically, when the UAV approaches strong electromagnetic interference sources such as high-voltage tower heads, in order to maintain the convergence of BeiDou / RTK positioning calculations and ensure obstacle avoidance safety, the flight control system or operator usually forcibly triggers a safety deceleration strategy. This means that in areas with stronger interference, the time required for the UAV to fly a unit distance is longer, resulting in a multiplied, rather than nonlinear, increase in the energy consumption of the hovering base, which is a major energy consumer. Second, the prediction method ignores the spurious work gradient caused by positioning noise. Specifically, the intensity of BeiDou signal noise caused by electromagnetic interference increases exponentially with distance. The decaying noise forces the UAV to make high-frequency attitude corrections to maintain its predetermined coordinates. This spurious work generated to counteract noise is not a fixed coefficient, but a continuous energy dissipation field integrated along the inspection path. Due to the failure to establish a precise mathematical mapping between the electromagnetic field strength distribution and the UAV's dynamic response, the above prediction method often underestimates the remaining power consumption by 15%-20% in complex electromagnetic environments, especially when involving detailed inspections near towers. This systematic bias can easily mislead the dispatch system into making aggressive decisions, increasing the risk of in-flight power outages or crashes.

[0038] To address the aforementioned shortcomings, this application proposes a preferred embodiment. First, a path field strength distribution sequence is constructed based on smooth trajectory points and a pre-defined list of remaining inspection waypoints. Specifically, using the known pre-defined list of remaining inspection waypoints (containing the three-dimensional coordinates of the power towers) and the total remaining path distance, the continuous inspection path is discretized into a series of small step sizes (e.g., ...). The micro-element road segment. Based on the laws of electromagnetic field propagation, a system is established with each power tower as the core. The electromagnetic interference potential energy function centered on the path is used to calculate the potential energy of each discrete sampling point along the path. The scalar of the combined interference field strength at the location can be expressed by the following formula: in, Defined as the first The scalar of the combined electromagnetic interference field strength at each path element can quantify the comprehensive electromagnetic pressure experienced at that point. Defined as the center coordinates of the nearest power tower; Defined as a reference constant for the electromagnetic radiation intensity of power towers, it depends on the voltage level; Defined as the electromagnetic field attenuation index (usually taken as 2 or 3), it reflects how quickly the interference attenuates with distance; Defined as a regularization term to prevent the denominator from being zero. By refining the macroscopic path into a microscopic field strength distribution sequence, the system can accurately capture the real electromagnetic environment pressure faced by the UAV during each meter of flight, thus providing a refined input benchmark for subsequent computational speed degradation and additional power consumption.

[0039] Next, a dynamic response sequence incorporating adaptive safety planning velocity and virtual electromagnetic viscous drag power was obtained by performing a field strength coupling-based dynamic response calculation on the path field strength distribution sequence. This step aims to reconstruct the motion state of the UAV within each micro-segment based on the physical fact that stronger disturbances result in slower speeds and greater oscillations. Specifically, the adaptive safety planning velocity for each micro-segment was calculated using an exponential decay model. Field strength The higher the value, the lower the planning speed, to simulate the safe deceleration logic in real-world scenarios. Simultaneously, the virtual electromagnetic viscous drag power caused by anti-interference attitude correction is calculated. This power is positively correlated with the field strength, representing the extra energy the UAV expends to resist positioning drift. This process can be expressed by the formula: in, Defined as the maximum cruising speed under undisturbed conditions; Defined as the speed suppression coefficient, it reflects the sensitivity of the flight control system to interference. The larger the value, the more pronounced the deceleration when encountering interference; Defined as the first Additional attitude correction power generated by the point due to counteracting localization noise; Defined as the oscillation energy consumption conversion coefficient. This step transforms abstract electromagnetic interference into concrete speed loss and power penalty, establishing an analytical constraint between environmental field strength and UAV flight control behavior, thereby enabling the realistic reproduction of the UAV's cautious, high-frequency flight state in the near-tower area.

[0040] Furthermore, based on the hovering reference power and basic mobile energy consumption coefficient in the real-time energy efficiency parameters, the differential energy consumption of the micro-element road segment, including hovering time extension loss and anti-interference additional energy consumption, is integrated using the dynamic response sequence to obtain the predicted total energy consumption. That is, after obtaining the dynamic response of each micro-element segment, numerical integration is performed over the entire remaining path. Specifically, for each micro-element... Using the formula Calculate its actual flight time, and then calculate the total energy consumption within that segment: including the hovering base energy consumption accumulated over time. Additional energy consumption due to interference over time and the base mobile energy consumption accumulated with distance ( Finally, the total amount of electricity required for a high-precision prediction is obtained by summing the results. This process can be expressed by the formula: in, Defined as the total power required for the improved high-precision prediction; Defined as the hovering base power of the drone; Defined as the infinitesimal step size in path discretization; Defined as the energy consumption per unit distance traveled in a clean environment; while This study profoundly reveals the physical mechanism by which electromagnetic interference causes speed reduction and, consequently, nonlinearly extends hovering time. This approach eliminates errors arising from overly idealized speed assumptions, enabling pixel-level accurate prediction of remaining range and providing a reliable basis for scheduling systems.

[0041] Subsequently, based on the predicted total power required, a scheduling risk index is determined. Specifically, firstly, the current remaining power data provided by the battery management system is read from the aligned state dataset, and the total power required to complete the remaining inspection path is obtained. The predicted total power required is divided by the current remaining power to obtain an original ratio. This ratio intuitively reflects the degree to which the current remaining power meets the power required to complete the entire task under an ideal battery model. If this ratio is less than 1, it indicates that the power is theoretically sufficient; if it is equal to 1, it is in a critical state; if it is greater than 1, it indicates that there is a power shortage. Furthermore, a battery discharge depth penalty factor is introduced for nonlinear correction. Considering the physical characteristics of lithium batteries, in the later stages of discharge (i.e., when the remaining power is low), there is a phenomenon of accelerated voltage drop, increased internal resistance, and less actual releaseable energy than the nominal remaining power. To more realistically assess the risk, the system introduces a battery discharge depth penalty factor. This factor is a function related to the current remaining power. When the remaining power is high, the penalty factor is close to 1, and has almost no amplification effect on the original ratio; when the remaining power is low, the penalty factor will be significantly greater than 1. The system multiplies the original ratio calculated in the previous step with this penalty factor, thereby weighting and amplifying the risk in low-battery conditions to obtain the final scheduling risk index. After this nonlinear correction, the final risk index can more accurately reflect the battery's true range potential, especially when the battery is low, the system will make more conservative decisions.

[0042] Specifically, this preferred example establishes an analytical mapping between the electromagnetic environment and the UAV's dynamic response by introducing a virtual potential energy field, thereby solving the problem of insufficient accuracy in endurance prediction under complex electromagnetic environments. This enables the system to accurately quantify the implicit causal chain of interference leading to deceleration, deceleration leading to prolonged hovering time, and ultimately resulting in a nonlinear surge in total energy consumption, improving the granularity of the prediction model from the entire path to micro-level segments. Ultimately, this refined modeling significantly reduces the risk of in-flight power outages due to energy consumption prediction errors, while avoiding wasted inspection efficiency caused by overly conservative estimates, ultimately achieving a dual improvement in both safety and efficiency for power inspection and scheduling.

[0043] Specifically, the scheduling instruction generation module 360 ​​is used to compare the scheduling risk index with a preset risk threshold, and generate a final scheduling instruction containing actions such as continue execution, task pruning, or emergency return to base based on the comparison result. It should be understood that although the preceding steps have accurately quantified the potential risks during the inspection process through multi-source data fusion and energy efficiency modeling, the numerical and abstract risk index cannot be directly understood by the UAV flight control system. The on-site environment for power line inspections is extremely complex; positioning fluctuations and surges in energy consumption caused by strong electromagnetic interference often occur within milliseconds. If the subjective experience of ground command personnel is relied upon entirely to determine whether the inspection needs to be interrupted, response delays often lead to severe power depletion or loss of control and crashes. By executing this step, the system can automatically trigger a link response from the airborne sensors to the flight control layer at the first moment of risk perception, based on preset safety boundaries. This not only greatly improves the self-healing survivability of the UAV under extreme conditions but also ensures that the inspection mission can guarantee asset safety even when energy is limited, thereby maximizing efficiency and controlling risks.

[0044] In practice, the first step is to select a scheduling strategy based on a risk threshold mapping of the scheduling risk index to obtain a scheduling strategy identifier corresponding to the continue execution mode, task pruning mode, or emergency return mode. During this process, the calculated scheduling risk index is compared with two preset risk thresholds in the system, typically a low-risk threshold and a high-risk threshold. Based on the comparison results, a corresponding scheduling strategy identifier is selected. The decision logic is as follows: when the risk index is lower than or equal to the low-risk threshold, the system considers the power supply sufficient and chooses to continue the original plan; when the risk index is between the low-risk and high-risk thresholds, the system considers the power supply tight but still manageable and chooses to prune some non-core tasks to ensure a safe return; when the risk index is higher than the high-risk threshold, the system considers the power supply insufficient to ensure safety and an immediate return is necessary.

[0045] Next, based on the scheduling policy identifier, the remaining inspection waypoint list is restructured and shortcuts are planned to obtain the final execution waypoint sequence. Specifically, based on the scheduling policy identifier obtained in the previous step, the module processes the preset remaining inspection waypoint list accordingly. If the policy is to continue execution, the final waypoint sequence remains consistent with the original plan. If the policy is to cut tasks, one or more non-critical waypoints need to be deleted from the remaining waypoint list according to preset rules (such as waypoint priority and geographical proximity), and the optimal connection path from the current position to the remaining critical waypoints and finally to the return point is recalculated. If the policy is to return urgently, the waypoint sequence is simplified to the shortest path directly from the current position to the return point, ignoring all remaining inspection task points.

[0046] Next, the final waypoint sequence is converted into underlying flight control protocol code, and a recommended cruise speed calculated based on real-time energy efficiency parameters is appended to obtain the final dispatch command. Specifically, firstly, each three-dimensional coordinate point in the final waypoint sequence is converted into corresponding command code according to the specifications of the underlying flight control protocol. Simultaneously, a recommended cruise speed is calculated based on factors such as real-time energy efficiency parameters, and this speed value is appended as a parameter to the corresponding waypoint command. Finally, all these commands are packaged into a complete final dispatch command package containing waypoint and speed information, and sent to the UAV's flight control system for execution via data link.

[0047] As described above, the BeiDou-based power line inspection trajectory management and scheduling system 300 according to the embodiments of this application can be implemented in various wireless terminals, such as servers with BeiDou-based power line inspection trajectory management and scheduling algorithms. In one possible implementation, the BeiDou-based power line inspection trajectory management and scheduling system 300 according to the embodiments of this application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the BeiDou-based power line inspection trajectory management and scheduling system 300 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the BeiDou-based power line inspection trajectory management and scheduling system 300 can also be one of many hardware modules of the wireless terminal.

[0048] Alternatively, in another example, the BeiDou-based power inspection trajectory management and scheduling system 300 and the wireless terminal can also be separate devices, and the BeiDou-based power inspection trajectory management and scheduling system 300 can be connected to the wireless terminal via wired and / or wireless networks, and transmit interactive information in accordance with the agreed data format.

[0049] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A power line inspection trajectory management and scheduling system based on BeiDou positioning, characterized in that, include: The data acquisition module is used to acquire raw sensor data streams containing BeiDou raw observation data, airborne magnetometer data, airborne inertial navigation data, and battery status data. The data alignment module is used to perform time synchronization interpolation and multi-source data alignment on the original sensor data stream using the BeiDou timing signal as a reference to obtain an aligned state dataset. The smoothing filter module is used to perform adaptive BeiDou trajectory smoothing filter based on electromagnetic sensing on the BeiDou coordinate data in the aligned state dataset based on the magnetic field strength data in the aligned state dataset to obtain smooth trajectory points. The real-time energy efficiency parameter determination module is used to determine real-time energy efficiency parameters based on the instantaneous current data in the smooth trajectory points and the aligned state dataset. The task trajectory cost prediction module is used to calculate the remaining inspection path based on the smoothed trajectory points and the preset list of remaining inspection waypoints, and to predict the task trajectory cost of the remaining inspection path based on real-time energy efficiency parameters to obtain the scheduling risk index. The scheduling instruction generation module is used to compare the scheduling risk index with the preset risk threshold, and generate the final scheduling instruction containing actions such as continue execution, task pruning, or emergency return based on the comparison results.

2. The power line inspection trajectory management and scheduling system based on BeiDou positioning according to claim 1, characterized in that, The data alignment module includes: The multi-source data synchronization unit is used to extract the GPS week and second from the BeiDou observation data in the original sensor data stream as the system master clock, and to perform linear interpolation resampling on the airborne inertial navigation data and airborne magnetometer data in the original sensor data stream based on the discrete time sequence of the system master clock to obtain synchronized multi-source data. The data augmentation unit is used to extract the three-axis magnetic field components from the synchronous multi-source data to calculate the magnetic field strength scalar, and subtract the preset geomagnetic background reference value to obtain the environmental electromagnetic interference intensity characteristics. The environmental electromagnetic interference intensity characteristics are then merged into the synchronous multi-source data as a new dimension to obtain the enhanced intermediate data packet. The dimensional normalization unit is used to normalize the dimensions of each physical quantity in the enhanced intermediate data packet to obtain an aligned state dataset.

3. The power line inspection trajectory management and scheduling system based on BeiDou positioning according to claim 1, characterized in that, The smoothing filter module is used for: The electromagnetic interference intensity characteristics in the aligned state dataset are analyzed, the dynamic expansion coefficient is calculated based on the preset basic observation noise covariance matrix, and an adaptive observation noise covariance matrix with diagonal elements that increase nonlinearly with the interference intensity is constructed. Using the optimal estimated state from the previous time step in the aligned state dataset and the inertial navigation data, a state transition equation is constructed to deduce the prior state estimation vector and the prior error covariance matrix at the current time step to obtain the prior state prediction packet. The Kalman gain matrix is ​​calculated using the adaptive observation noise covariance matrix and the prior state prediction packet. The prior state estimation vector is then updated and corrected by combining the BeiDou observation coordinates extracted from the aligned state dataset to obtain smooth trajectory points.

4. The power line inspection trajectory management and scheduling system based on BeiDou positioning according to claim 1, characterized in that, The real-time energy efficiency parameter determination module includes: The instantaneous power decomposition and attitude oscillation loss quantization unit is used to perform instantaneous power decomposition and attitude oscillation loss quantization on the battery voltage, instantaneous current and angular velocity data in the aligned state dataset to obtain power feature data packets; The effective displacement velocity calculation unit is used to calculate the effective displacement velocity of smooth trajectory points to obtain the effective velocity scalar. The real-time energy efficiency parameter calculation unit is used to subtract the preset theoretical hovering power from the instantaneous total power in the power characteristic data packet to obtain the working power, and to calculate the ratio of the working power to the effective speed scalar to obtain the real-time energy efficiency parameters.

5. The power line inspection trajectory management and scheduling system based on BeiDou positioning according to claim 4, characterized in that, The effective displacement velocity calculation unit is used to calculate the effective displacement velocity of points on a smooth trajectory using the following formula: in, , and These are the three-axis velocity components in the smooth trajectory points.

6. The power line inspection trajectory management and scheduling system based on BeiDou positioning according to claim 1, characterized in that, The task trajectory cost prediction module is used for: Starting from the current coordinates in the smooth trajectory point, connect the unvisited waypoints in the remaining inspection waypoint list and the final return point in sequence to calculate the total Euclidean distance of the remaining inspection path. Identify road segments that cross high-interference areas in the remaining inspection path to apply a safety redundancy coefficient, and use the unit effective displacement energy consumption coefficient and hovering reference power in the real-time energy efficiency parameters to perform a weighted integral on the total Euclidean distance to obtain the predicted total power consumption. Obtain the current remaining power in the aligned state dataset, calculate the ratio of the predicted total power required to the current remaining power, and introduce a battery discharge depth penalty factor to perform nonlinear correction to obtain the scheduling risk index.

7. The power line inspection trajectory management and scheduling system based on BeiDou positioning according to claim 1, characterized in that, The scheduling instruction generation module is used for: The scheduling risk index is used to select a decision strategy based on risk threshold mapping to obtain a scheduling strategy identifier corresponding to the continue execution mode, task pruning mode or emergency return mode. Based on the scheduling strategy identifier, the remaining inspection waypoint list is restructured and shortcuts are planned to obtain the final execution waypoint sequence. The final waypoint sequence is converted into underlying flight control protocol code, and a recommended cruise speed calculated based on real-time energy efficiency parameters is added to obtain the final dispatch instructions.

8. The power line inspection trajectory management and scheduling system based on BeiDou positioning according to claim 1, characterized in that, The task trajectory cost prediction module is also used for: Based on the smooth trajectory points and the preset list of remaining inspection waypoints, a path field strength distribution sequence is constructed; Dynamic response sequences containing adaptive safety planning velocity and virtual electromagnetic viscous drag power have been obtained by performing dynamic response calculations based on field strength coupling on the path field strength distribution sequence. Based on the hovering reference power and basic mobile energy consumption coefficient in the real-time energy efficiency parameters, the differential energy consumption of the micro-element road segment, including the hovering time extension loss and anti-interference additional energy consumption, is used to obtain the predicted total power consumption. The scheduling risk index is determined based on the predicted total electricity demand.

Images