Unmanned aerial vehicle elevation measurement system based on laser scanning reference

Abstract

The invention discloses an unmanned aerial vehicle elevation measurement system based on a laser scanning benchmark, and belongs to the technical field of elevation measurement. The system comprises a laser sweeping subsystem, an unmanned aerial vehicle flight control subsystem and an elevation data acquisition and processing subsystem, the laser swinger subsystem obtains an elevation datum plane according to a plane formed by rotating laser emitted by a laser swinger at a preset elevation benchmark of a to-be-measured area; the unmanned aerial vehicle flight control subsystem executes the flight control instruction and controls the position and attitude of the unmanned aerial vehicle in real time; the elevation data acquisition and processing subsystem acquires laser signals on the laser receiver, and obtains three-dimensional space information of a to-be-measured target point area measured by the depth camera; meanwhile, data processing is carried out on the collected laser signals and three-dimensional space information, the processed data are integrated, and elevation information of the to-be-measured target point is generated. Accurate measurement data is obtained through various sensors, and accurate elevation data measurement is realized in combination with a stable elevation datum plane.

Description

Technical Field

[0001] The present invention belongs to the technical field of elevation measurement, and more specifically, relates to an unmanned aerial vehicle (UAV) elevation measurement system based on a laser leveling reference. Background Art

[0002] Elevation measurement has important applications in many fields such as topographic surveying and mapping, highway construction, building engineering, precision agriculture, mineral resource exploration, and disaster emergency; traditional elevation measurement methods mainly rely on ground measurement instruments such as level instruments, total stations, and laser rangefinders, and professional technicians manually operate to record elevation data; although these methods have high measurement accuracy, due to completely relying on ground operations, they usually require a large amount of manpower and material resources, and the operation is complex, and is greatly restricted by terrain conditions. When conducting measurements in large areas, complex terrains, or remote areas, these traditional methods are time-consuming and costly.

[0003] In recent years, with the rapid development of UAV technology, UAVs have gradually shown unique advantages in the field of elevation measurement, making the elevation measurement system start to develop towards unmanned and automated directions; UAVs, with their advantages of high mobility, fast coverage, flexible operation, and adaptability to complex terrains, have significantly improved the efficiency in elevation measurement; by carrying a variety of sensors, UAVs can complete the elevation measurement of the target area in the air and generate a three-dimensional terrain model, providing data support for various engineering designs and terrain analyses; although the existing UAV elevation measurement technologies have made progress in many aspects, there are still many deficiencies in elevation accuracy, stability, and anti-interference; specifically, the deficiencies in the existing technologies are mainly reflected in the following aspects:

[0004] Most of the existing UAV elevation measurement systems rely on a single sensor, such as a height sensor, lidar, or optical camera; a single sensor may produce errors in a specific environment, such as a large measurement error may occur for a height sensor in an irregular measurement area, reflection deviation may occur for lidar on the water surface or a smooth surface, and data loss or measurement error will occur for an optical camera in a low-light or high-light environment; in addition, the measurement range and accuracy of a single sensor are limited, and it is difficult to meet the elevation accuracy requirements of large-area complex terrains; existing UAVs usually conduct measurements without a fixed reference object, and it is difficult to establish a high-precision and high-stability reference plane, resulting in challenges in real-time correction and precise positioning of data; this instability of the reference plane not only limits the measurement accuracy but also affects the reliability and consistency of the measurement data.

[0005] How to achieve the coordinated cooperation of multi-sensor data and ensure the accuracy and consistency of elevation data is one of the difficulties in the current elevation measurement technology. Summary of the Invention

[0006] Aiming at the defects of the related technologies, the purpose of the present invention is to provide an unmanned aerial vehicle (UAV) elevation measurement system based on a laser leveling reference, aiming to solve the problems of poor reliability and consistency of the elevation data obtained by the UAV, resulting in low accuracy of UAV elevation measurement.

[0007] To achieve the above object, the present invention provides a UAV elevation measurement system based on a laser leveling reference, including: a laser leveling subsystem, a UAV flight control subsystem, and an elevation data acquisition and processing subsystem;

[0008] The laser leveling subsystem is used to obtain an elevation reference plane based on the plane formed by the rotating laser emitted by the laser level at a preset elevation benchmark point in the area to be measured;

[0009] The UAV flight control subsystem is used to receive position information, generate flight control instructions, control the UAV to fly to the target points to be measured in sequence along the flight path for measurement, and perform real-time control on the position and attitude of the UAV;

[0010] The elevation data acquisition and processing subsystem includes a laser receiver, a depth camera, and an on-board computer disposed on the UAV; the laser receiver is used to receive the laser signal generated by the laser level scanning; the depth camera is used to measure the three-dimensional spatial information of the area of the target point to be measured; the on-board computer is used to perform data processing on the obtained laser signal and three-dimensional spatial information, obtain the relative position data between the UAV and the elevation reference plane according to the laser signal, obtain the relative height between the UAV and the target point to be measured according to the three-dimensional spatial information, and perform data integration on the relative position data and the relative height to generate the elevation information of the target point to be measured; where the elevation value H in the elevation information of the target point to be measured final = {(b + h2) - (a + h1)}, h1 is the relative position between the elevation reference plane and the laser receiver, h2 is the relative height from the depth camera to the ground of the target point to be measured; a is the distance from the starting base point of the laser receiver installation to the bottom of the UAV, and b is the distance from the starting base point of the depth camera installation to the bottom of the UAV.

[0011] Optionally, the laser receiver is fixed to the bottom of the UAV, and the receiving direction is perpendicular to the UAV body;

[0012] The depth camera is installed at the bottom of the UAV, and the viewing angle is vertically downward.

[0013] Optionally, the on-board computer uses a variational mode decomposition method with increased iterative feedback to extract the ideal laser signal from the distance data of the UAV to the elevation reference plane obtained by the laser receiver.

[0014] Optionally, the on-board computer constructs a cylindrical three-dimensional space processing area with a radius of r based on the three-dimensional space information collected by the depth camera, divides it into four sub-spaces S1, S2, S3, and S4, calculates the average depth value of the area to be measured through the point cloud projection of the four space regions, and obtains the relative height between the depth camera and the target point to be measured through the geometric projection relationship.

[0015] Optionally, the real-time adjustment of the UAV by the UAV flight control subsystem includes:

[0016] Obtain the real-time data of the inertial measurement unit inside the flight controller to get the pitch angle θ and roll angle at the current moment

[0017] According to the formula Calculate the included angle α between the real-time center of gravity direction and the direction of gravitational acceleration of the UAV to perform real-time dynamic adjustment of the attitude angle of the UAV;

[0018] Obtain the real-time position information of the real-time kinematic differential positioning module to perform real-time dynamic adjustment of the position of the UAV.

[0019] Optionally, the elevation data acquisition and processing subsystem uses the included angle α between the center of gravity direction and the direction of gravitational acceleration of the UAV to perform real-time dynamic compensation on the height data in the laser data and the height data in the three-dimensional space data during the data integration process;

[0020] The height data in the laser data is corrected to the height after angle compensation as:

[0021]

[0022] where, H relative is the height initially measured by the laser receiver, and Δα is the error value of the included angle between the center of gravity direction and the direction of gravitational acceleration;

[0023] The height data in the three-dimensional space data is corrected to the height after angle compensation as:

[0024]

[0025] where, H depth is the height initially measured by the depth camera;

[0026] The elevation information of the target point to be measured is:

[0027] H final ={(b + h2)-(a + h1)}

[0028] where, h1 is the relative position H′ of the laser receiver and the elevation reference plane relative, h2 is the relative height H′ from the depth camera to the ground of the target point to be measured depth , a and b are respectively the distances from the installation starting base points of the laser receiver and the depth camera to the bottom of the drone.

[0029] Optionally, the laser leveling subsystem includes a level, a tripod, and a laser leveler;

[0030] The level is used to establish a preset elevation level point in the area to be measured by the method of gradually transmitting the height difference through known elevation reference points;

[0031] The laser leveler is fixedly installed on the tripod and is used to horizontally arrange at a preset standard height on the starting elevation level point in the area to be measured, so that the rotating laser emitted by it forms an elevation reference plane, which serves as the elevation measurement reference for the drone in the area to be measured.

[0032] Optionally, the drone flight control subsystem includes an on-board computer, a flight controller, and a real-time kinematic differential positioning module;

[0033] The on-board computer is used to receive the position information of the area to be measured, generate an optimal flight path, and send the flight control instructions to the flight controller;

[0034] The real-time kinematic differential positioning module is used to obtain the position information of the drone during flight;

[0035] The flight controller is used to receive the flight control instructions, control the drone to fly to each target position to be measured in sequence; and is also used for the real-time adjustment of the position and attitude of the drone during the flight process and the data acquisition process.

[0036] Through the above technical solutions conceived by the present invention, compared with the prior art, the following beneficial effects can be achieved:

[0037] 1. A drone elevation measurement system based on a laser leveling reference provided by the present invention realizes the acquisition of high-precision elevation data through the combined action of multiple sensors such as a laser leveler, a laser receiver, and a depth camera; specifically, the reference plane generated by the laser leveler provides a stable elevation reference, the relative position data between the drone and the elevation reference plane is obtained through the laser receiver, and the height information between the drone and the target point to be measured is obtained in combination with the depth camera to realize accurate elevation data measurement; this data acquisition method greatly improves the accuracy and reliability of the elevation data.

[0038] 2. The present invention provides an unmanned aerial vehicle (UAV) elevation measurement system based on a laser sweeping benchmark, which improves the measurement accuracy of the system by designing an efficient data processing algorithm. The ideal laser signal is extracted from the laser signal collected by the laser receiver by adding a variational mode decomposition method with iterative feedback, and the relative position of the elevation reference surface on the receiver is accurately calculated. The three-dimensional spatial information collected by the depth camera is constructed into a cylindrical three-dimensional spatial processing area with a radius of r, and the area is divided into four subspaces, S1, S2, S3 and S4. The average depth of the area to be measured is calculated by projecting point clouds of the four spatial areas. These two processing algorithms optimize the data processing process, improve the measurement accuracy of the system, and enable the UAV to achieve more accurate real-time data processing during the data collection process.

[0039] 3. The present invention provides a UAV elevation measurement system based on a laser sweeping benchmark, which performs real-time position and attitude adjustment by combining a real-time dynamic differential positioning module and an inertial measurement unit in the flight controller. By fusing data from the real-time dynamic differential positioning module and the inertial measurement unit in the flight controller, the flight control system can calculate the position and attitude adjustment of the UAV in real time, automatically adjust the position, pitch, roll, and yaw angles, offset deviations caused by factors such as airflow and terrain changes, and ensure that the UAV hovers stably above the target point to be measured; at the same time, the altitude data obtained by the laser receiver and depth camera are compensated in real time according to the attitude adjustment amount; this attitude compensation technology not only significantly improves the UAV's anti-disturbance capability, but also ensures the accuracy of elevation measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is the elevation measurement working state of a UAV elevation measurement system based on a laser leveling benchmark provided by the present invention in a certain engineering scene;

[0041] Figure 2 This is a workflow diagram of a UAV elevation measurement system based on laser leveling benchmark provided by the present invention;

[0042] Figure 3 A schematic diagram of the elevation calculation of a target point to be measured by an unmanned aerial vehicle elevation measurement system based on a laser sweeping benchmark provided by the present invention;

[0043] Figure 4 This is a flow chart of extracting the ideal laser signal using the variational mode decomposition method with added iterative feedback;

[0044] Figure 5 It is a schematic diagram of the point cloud division of the depth camera's measurement data in the target area;

[0045] Figure 6 It is a schematic diagram of the drone coordinate system;

[0046] Figure 7It is a schematic diagram of height data compensation.

[0047] Explanation of reference numerals:

[0048] Laser level 5, tripod 6, flight controller 7, real-time kinematic differential positioning module 8, laser receiver 9, depth camera 10, on-board computer 11. Detailed implementation manners

[0049] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0050] The following describes the content involved in the above embodiments in conjunction with a preferred embodiment.

[0051] Refer to Figure 1 , a drone elevation measurement system based on a laser leveling reference, comprising: a laser leveling subsystem, a drone flight control subsystem, and an elevation data acquisition and processing subsystem;

[0052] The laser leveling subsystem is used to obtain an elevation reference plane according to the plane formed by the rotating laser emitted by the laser level at a preset elevation benchmark point in the area to be measured;

[0053] The drone flight control subsystem is used to receive position information, generate flight control instructions, control the drone to fly to the target points to be measured in sequence along the flight path for measurement, and perform real-time control on the position and attitude of the drone;

[0054] The elevation data acquisition and processing subsystem includes a laser receiver, a depth camera, and an on-board computer arranged on the drone; the laser receiver is used to receive the laser signal generated by the laser level scanning; the depth camera is used to measure the three-dimensional space information of the area of the target point to be measured; the on-board computer is used to perform data processing on the obtained laser signal and three-dimensional space information, obtain the relative position data between the drone and the elevation reference plane according to the laser signal, obtain the relative height between the drone and the target point to be measured according to the three-dimensional space information, and perform data integration on the relative position data and the relative height to generate the elevation information of the target point to be measured; where the elevation value H in the elevation information of the target point to be measured final={(b + h2) - (a + h1)}, where h1 is the relative position between the elevation reference plane and the laser receiver, h2 is the relative height from the depth camera to the ground of the target point to be measured; a is the distance from the starting base point of the laser receiver installation to the bottom of the drone, and b is the distance from the starting base point of the depth camera installation to the bottom of the drone.

[0055] As Figure 2 shown, the complete measurement process of the drone elevation measurement system for measuring the target point to be measured, processing multi-sensor measurement data, and obtaining the elevation information of the target point to be measured includes: calibrating the elevation benchmark points in the target area to be measured; forming an elevation reference plane through the laser leveling subsystem; the drone going to the target area according to the flight control instruction to measure the elevation of each target point in the target point area of the target area to be measured; the data processing system performing data processing and integration on the collected laser signals and three-dimensional space information to obtain the elevation information of the target point to be measured.

[0056] (1) Calibration of the elevation benchmark points in the area to be measured, and the specific steps are as follows:

[0057] (1.1) Select a reference point with a known elevation as the starting point of calibration, and set its elevation as H0.

[0058] (1.2) In the area to be measured, use a level to measure the elevation difference between the starting reference point and the next calibration point; assuming the elevation difference of the first calibration point is Δh1, then the elevation H1 of the first calibration point is:

[0059] H1 = H0 + Δh1

[0060] where Δh1 can be positive or negative, depending on the height difference between the calibration point and the starting point.

[0061] (1.3) According to the required elevation reference density in the area to be measured, sequentially measure the elevation differences of the subsequent calibration points. For the i-th calibration point, the elevation H i can be expressed as:

[0062] H i = H i-1 + Δh i

[0063] where H i-1 is the elevation value of the (i - 1)-th leveling point, and Δh i is the elevation difference between adjacent calibration points.

[0064] Measure the same elevation difference multiple times and take the average value, or measure multiple leveling points in a closed loop path to detect and correct systematic errors; the cumulative elevation difference of the closed leveling line should be zero, which is used to check the accuracy of the elevation calibration. As Figure 1 shown, B1 in it is an elevation benchmark point selected from the calibrated benchmark points in the area to be measured in a certain project.

[0065] (2) The laser leveling subsystem obtains the elevation reference plane based on the plane formed by the rotating laser emitted by the laser leveling instrument at the preset elevation level point in the area to be measured.

[0066] The laser leveling subsystem includes a level, a tripod 6 and a laser leveling instrument 5;

[0067] The level is used to establish a preset elevation level point in the area to be measured by transferring the elevation difference step by step from the known elevation reference point.

[0068] The laser leveler 5 is fixedly mounted on a tripod and is used to be horizontally arranged at a preset standard height on the starting elevation level point of the area to be measured, so that the rotating laser it emits forms an elevation reference plane, which serves as the elevation measurement reference of the drone in the area to be measured.

[0069] (2.1) Accurately fix the laser leveler 5 on the tripod 6 at the leveling point B1 to ensure that the installation position of the instrument is stable and avoid position deviation.

[0070] (2.2) Use the precision adjustment mechanism and level of the laser leveler 5 to adjust the angle and horizontal position of the instrument at the installation location. Use an electronic bubble level or mechanical level to confirm the overall horizontal position of the instrument to ensure that the laser emission surface is parallel to the horizontal plane.

[0071] (2.3) Start the laser leveler 5 so that the laser beam it emits forms a 360-degree rotating light plane. The rotating light plane should use the elevation of the leveling point B1 as the leveling point elevation.

[0072] (2.4) Based on the known B1 elevation, fine-tune the height of the laser leveler 5 to ensure that the center of the rotating laser plane is aligned with the elevation of the leveling point. This step is accomplished by gradually adjusting the vertical height between the horizontal reference plane and B1 to avoid elevation deviations caused by height errors.

[0073] (2.5) Use auxiliary height detection devices, such as electronic levels or elevation sensors, to reconfirm the elevation reference of the laser plane to ensure accuracy.

[0074] (3) The UAV goes to the area to be measured according to the flight control command to measure the elevation of each target point in the area to be measured. The specific steps are as follows:

[0075] (3.1) The drone receives the position information of the target points to be measured. An elevation data acquisition and processing subsystem and a drone flight control subsystem are set on the drone. The drone flight control subsystem includes an on-board computer 11, a flight controller 7, and a real-time kinematic differential positioning module 8. The elevation data acquisition and processing subsystem includes a laser receiver 9, a depth camera 10, and an on-board computer 11. Among them, the on-board computer 11 is shared by the two subsystems.

[0076] The on-board computer 11 is used to receive the position information of the area to be measured, generate the optimal flight path, and send the flight control instructions to the flight controller 7.

[0077] The real-time kinematic differential positioning module 8 is used to obtain the high-precision position information during the flight of the drone.

[0078] The flight controller 7 is used to receive the flight control instructions, control the drone to fly to each target position to be measured in sequence, and is also used for the real-time adjustment of the position and attitude of the drone during the flight process and the data acquisition process.

[0079] Number the position information of each known target point to be measured, specify the measurement order, and ensure the accuracy of each point's position, so that the drone can go to and execute the measurement tasks in sequence. To ensure the consistency and easy parsing of the data format of the target position information, the binary encoding is used to format the target position information.

[0080] Input the position information of the target to be measured into the drone. Specifically: Transmit the data containing the number, coordinates, and elevation information of the target point to the drone through the ground control system.

[0081] After receiving the data, the on-board computer 11 in the drone flight control subsystem performs verification to ensure that the position information of the target to be measured stored in the drone is exactly the same as the input data. The drone sends a reception confirmation message to the ground station and uploads the list of positions to be measured to the flight controller 7.

[0082] After the information is input, the on-board real-time kinematic differential positioning module 8 of the drone and the confirmation program proofread the target position information to confirm that all coordinate information is valid and reachable within the flight area of the drone. If the target position information changes during the task, the on-board computer 11 supports the dynamic update of the position information and real-time upload to the flight controller 7. After the new information is input, the drone automatically adjusts the flight path and updates to the latest list of target position information to ensure the flexibility and real-time adaptability of the flight path.

[0083] (3.2) The UAV flies to the target point to be measured according to the position information of the target point to be measured. The specific steps are as follows: The on-board computer 11 of the UAV generates an optimal flight path according to the received position information of the target point to be measured and its measurement order; The path planning takes into account the take-off point of the UAV, the shortest path between each target point, the flight altitude and the obstacle avoidance strategy to ensure the efficiency and safety of the entire measurement task; The UAV flies to the target point to be measured in turn according to the optimal flight path.

[0084] The real-time kinematic differential positioning module 8 provides centimeter-level three-dimensional coordinate information to the flight controller 7, enabling the UAV to accurately fly along the established path and locate to each target point to be measured; For the positioning of each target point, according to the input coordinates of the target point to be measured, such as Figure 1 the target point M7(X0, Y0, Z0) to be measured in, the UAV continuously adjusts its own position to ensure that the coordinate error between it and the target point M7 is maintained within the centimeter level; The flight controller 7 adjusts the flight path according to the real-time updated RTK data. When the UAV approaches the target point, the speed gradually decreases to achieve stable and accurate positioning; After reaching the coordinate area (X0 + Δx, Y0 + Δy, Z0 + Δz) of the target point, the UAV enters the hover mode to ensure accurate measurement.

[0085] When the UAV is in the hover state, the flight controller 7 continuously detects the minute changes in the pitch angle and roll angle and performs attitude compensation through motor power adjustment; For example, if the pitch angle changes due to wind (for example, increases from 0° to +2°), the flight controller 7 immediately identifies the tilt and compensates by adjusting the thrust of the front and rear motors to return the pitch angle of the UAV to the target angle; Similarly, the offset of the roll angle will also trigger corresponding compensation operations. For example, when the roll angle offsets to -1.5°, the system corrects it in the reverse direction by adjusting the thrust of the left and right motors to ensure that the UAV always maintains a horizontal attitude; The entire compensation process is completed within milliseconds to ensure that the UAV maintains an accurate vertical hover state during the measurement process and is not affected by environmental micro-disturbances.

[0086] (3.3) The UAV hovering above the target point to be measured moves vertically to find the elevation reference plane. The specific steps are as follows:

[0087] The laser receiver 9 is fixed at the bottom of the UAV, and the receiving direction is perpendicular to the UAV body; The laser receiver 9 has a wide light receiving range to ensure that it can quickly detect the existence of the laser reference plane (elevation reference plane) during the vertical search process; The laser receiver 9 conducts real-time data interaction with the flight control system of the UAV through a USB data cable. This connection method not only supports efficient data transmission but also ensures the real-time processing and position feedback of the signal of the laser receiver 9 to maintain the accurate hover of the UAV at the elevation reference plane position.

[0088] The UAV flight control system controls the UAV to move slowly in the vertical direction according to the feedback data of the laser receiver 9. When the laser receiver 9 detects the elevation reference plane signal, its internal circuit transmits the received laser signal to the filtering circuit, amplification circuit and peak holding circuit through photoelectric conversion, and records the current signal intensity, so as to determine that the laser receiver 9 detects the elevation reference plane.

[0089] After capturing the elevation reference plane, the UAV maintains precise hovering on the elevation reference plane through continuous attitude control; the flight controller 7 combines attitude monitoring and real-time compensation mechanisms to automatically correct attitude or position offsets caused by environmental factors, ensuring that the laser receiver 9 always maintains its position on the reference plane and is not affected by external disturbances, thus ensuring accurate measurement of the height of the elevation reference plane in the subsequent process.

[0090] [[ID=***6***]](3.4) The UAV detects the elevation reference plane, maintains precise hovering and measures the target point to be measured.

[0091] The laser receiver 9 is fixed at the bottom of the UAV, and the receiving direction is perpendicular to the UAV body, and is used to detect the laser signal generated by the elevation reference plane on the laser receiver 9; the laser receiver 9 is connected to the on-board computer 11 through a USB data cable;

[0092] The depth camera 10 is installed at the bottom of the UAV, and the viewing angle is vertically downward, and is used to obtain the three-dimensional spatial information of the area of the target point to be measured.

[0093] The elevation data acquisition and processing subsystem obtains the laser signal generated by the elevation reference plane on the laser receiver 9 at this moment, and obtains the three-dimensional spatial information collected by the depth camera 10 at this time. The specific steps are as follows:

[0094] When the UAV is precisely hovering, when the horizontal rotating beam emitted by the laser level 5 reaches a certain height of the laser receiver 9, the laser receiver 9 will generate a laser signal with a corresponding intensity; at the same time, the depth camera 10 starts to capture the three-dimensional spatial information of the area directly below it; the depth camera 10 is installed at the bottom of the UAV, the viewing angle of the depth camera 10 is vertically downward, and its optical center is on the same vertical line as the center of the real-time kinematic differential positioning module 8; when the UAV is directly above the target point to be measured, the depth camera 10, the real-time kinematic differential positioning module 8 and the target point always remain collinear to ensure the accuracy of measurement and the consistency of positioning.

[0095] (4) The data acquisition and processing subsystem processes the collected laser signal and three-dimensional spatial information to obtain the elevation information of the target point to be measured, specifically including:

[0096] (4.1) Process the laser signal data collected by the laser receiver 9.

[0097] Note: There seems to be a formatting issue in the original Chinese text where the numbering in item (3.4) is incorrect. It should likely be (3.5) or something else for proper sequential numbering. I've translated it as it is but just noted this for clarity. Also, I've corrected the numbering in the English translation for item (3.4) to (3.5) in the translation for better logical flow. If this is not what you intended, please let me know.Optionally, a variational mode decomposition method with added iterative feedback is used to extract an ideal laser signal from the distance data from the UAV to the elevation reference surface acquired by the laser receiver 9.

[0098] External disturbances, such as drone jitter and sweeper mechanical vibration, can introduce high-frequency and low-frequency periodic oscillations or irregular fluctuations into the collected laser signal, resulting in suboptimal laser signals collected by the laser receiver 9. These unstable factors not only affect the quality of the laser signal but also reduce measurement accuracy and reliability. The onboard computer 11 uses a variational mode decomposition method with iterative feedback to separate the effects of drone jitter, sweeper vibration, and other factors, and separate the ideal laser signal from the laser signal data. Based on the peak position of the obtained ideal laser signal, the relative position H of the elevation reference surface on the receiver is accurately calculated. relative , and quantify it into millimeter units to facilitate subsequent precise position processing; Figure 4 As shown in FIG, detailed steps of using the variational mode decomposition method with added iterative feedback to process the laser signal collected by the laser receiver 9 are as follows:

[0099] Acquire the collected data and obtain the laser signal y(t) of the laser receiver 9 from the collected data. The laser signal contains the following three components:

[0100] y idesal (t): Ideal laser signal, frequency is 10HZ.

[0101] y drone (t): Laser signal changes caused by UAV jitter.

[0102] y vibration (t): Laser signal change caused by the mechanical vibration of the sweeper.

[0103] y(t)=y ideal (t)+y drone (t)+y vibration (t)

[0104] Extract the ideal laser signal y from the laser signal y(t) ideal (t).

[0105] Before using variational mode decomposition to process the acquired laser signal, the original signal is preprocessed to remove the trend term in the laser signal and perform normalization to eliminate baseline offset and avoid numerical instability:

[0106]

[0107] According to the known interference components (drone jitter and level vibrator vibration), the number of layers K of variational mode decomposition for the initial decomposition of the laser signal is set to 3. To ensure that the modal bandwidth of the ideal signal at 10HZ is narrow and avoid interference with adjacent frequencies, the initial value of the initial penalty factor α is set to 2000; during the iterative decomposition process of the laser signal, this parameter is dynamically adjusted according to the extraction effect of the ideal laser signal. If there is modal spectrum overlap, increase α to tighten the bandwidth; if the mode is too fragmented, then decrease α; set the relative error tolerance and maximum number of iterations of variational mode decomposition, and call the variational mode decomposition algorithm to decompose the preprocessed laser signal for the first time to obtain 3 intrinsic mode functions (IMFs):

[0108]

[0109] Perform Fourier transform on the 3 intrinsic mode functions (IMFs) obtained by decomposition respectively, and calculate the spectrum U k (f), and identify its center frequency f k ; According to the known frequency of the ideal laser signal is 10HZ, retain the intrinsic mode functions where 10HZ ≤ f k ≤ 11HZ (allowing ±1HZ fluctuation), check whether the waveform of the retained mode is a smooth periodic signal (excluding residual high-frequency glitches or low-frequency drifts), and superimpose the selected intrinsic mode functions to obtain the ideal laser signal

[0110]

[0111] Calculate the residual Perform wavelet packet decomposition on the residual r(t) to identify high-frequency and low-frequency subbands; Dynamically adjust the number of decomposition layers (such as K new = K + the number of interference subbands) and the value of the penalty factor α according to the subband energy; Continuously iterate this decomposition process according to the subband energy of the residual signal until the ideal laser signal is obtained; This method realizes the separation of various interference signals through the iterative optimization process of variational mode decomposition guided by the residual signal (negative feedback), so as to obtain the ideal laser signal y ideal (t).

[0112] According to the peak position of the obtained ideal laser signal, accurately calculate the relative position H of the elevation reference plane on the receiver relative , the relative height data H relativeIt is transmitted to the flight control system of the drone to achieve real-time data interaction through a USB data cable; the flight control system of the drone automatically adjusts the hovering height according to this relative height data to ensure that the position of the elevation reference plane always remains within the set height window, so as to achieve precise hovering of the drone on this reference plane; in order to cope with possible minor disturbances during flight, the laser receiver 9 periodically samples the relative height data to form a continuous set of height samples and reduces the height fluctuation by calculating the sliding average value; the calculation formula for the average relative height is as follows:

[0113]

[0114] where t i represents the timestamp of the i-th sampling, and H relative (t i ) represents the relative height data at that time.

[0115] (4.2) Process the three-dimensional space information between the measured target point collected by the depth camera 10 and the camera.

[0116] Optionally, the on-board computer 11 constructs a cylindrical three-dimensional space processing area with a radius of r based on the three-dimensional space information collected by the depth camera 10, divides it into four sub-spaces S1, S2, S3, and S4, calculates the average depth value of the measured area by means of point cloud projection of the four space areas, and obtains the relative height between the depth camera 10 and the measured target point through geometric projection relationship.

[0117] In the elevation measurement scenario, the measurement environment in the area where the measured target point is located is not absolutely ideal. The ground in the area where the measured target point is located may be uneven, have an inclination angle, have weeds or stones; when the ground is uneven or has an inclination angle, the measurement path of the common height sensor will change during measurement, resulting in the measured height value deviating from the true value; in the area where the ground is sunken, the sensor measures a lower height, while in the raised area, the measurement result is on the high side; when the ground fluctuates greatly, this error will be aggravated and cannot accurately reflect the true height of the measured target point;

[0118] To overcome the influence of uneven ground, inclination angles, weeds or stones in the area to be measured on height measurement, a depth camera 10 is used to calculate the height of the target point to be measured; the depth camera 10 can capture the three-dimensional point cloud information of the target point and its surrounding area, providing more accurate depth data; by processing the point cloud in the area near the target point to be measured to calculate the true height of the target point to be measured, not only can the influence brought by ground undulation be reduced, but also the limitation of single height sensor measurement can be eliminated, ensuring more stable and accurate height results; the advantage of the depth camera 10 in dealing with complex terrain lies in its ability to comprehensively capture environmental details, thus providing more accurate measurement data.

[0119] During point cloud processing, with the target point to be measured as the center, a cylindrical processing area with a radius of r is constructed as the point cloud analysis space; during point cloud processing, the PROSAC algorithm based on probability is used for plane fitting, a suitable candidate point set is selected to estimate the initial plane model, and the plane parameters are iteratively optimized to improve the accuracy of point cloud plane fitting; at the same time, the elevation feature extraction is carried out through the projection point density analysis method, the depth distribution feature of the point cloud in the target space area is calculated, and according to the average depth value of the area to be measured, the relative height between the depth camera 10 and the target point to be measured is obtained through the geometric projection relationship.

[0120] Determine the target point P to be measured target (x target , y target , z target ) of the position, and collect the point cloud of the area around the point to be measured; select a cylindrical area with a radius of r with the target point to be measured as the center as the spatial area for point cloud processing; for all the collected point clouds P i (x i , y i , z i ), judge whether it is located in this cylindrical space area according to the following formula:

[0121] (x i - x target ) 2 + (y i - y target ) 2 ≤ r 2

[0122] For each point P in the point cloud i (x i , y i , z i ), calculate its normal vector; if the normal of the point is significantly different from the normals of the surrounding points, then this point is considered a noise point and is removed; the outlier points are identified through clustering algorithms such as DBSCAN, and the points that do not conform to the expected shape, such as stones, weeds, etc., are removed.

[0123] Using the PROSAC plane fitting algorithm to further locate the point cloud of the target area; taking the point cloud information with lower depth information in the cylindrical target area as the preset threshold, selecting appropriate candidate points, and estimating the initial plane; using a plane equation form to describe the plane:

[0124] Ax + By + Cz + D = 0

[0125] where A, B, C, and D are the parameters of the plane equation.

[0126] Gradually select points that better meet the fitting requirements for plane fitting, select a subset from the randomly sampled points for plane fitting, evaluate the "fitness" of each point according to the fitting result, and preferentially select the points that are most consistent with the current fitting plane; in each round of sampling, calculate the fitness according to the distance from the point to the fitting plane:

[0127]

[0128] If the distance is less than the set threshold ∈, then the point is considered to belong to the plane.

[0129] Introduce more qualified points in each iteration, gradually approach the optimal plane by increasing the sampling of appropriate points; each iteration is evaluated based on the distance between the points and the current fitting plane to optimize the fitting result; continue to iterate until the point cloud sampling result meets the minimum error condition or reaches the maximum number of iterations; after meeting the convergence condition, output the final model, remove the point cloud outside the target point area, and extract all points that meet the fitting conditions.

[0130] After extracting the point cloud from the target space area, further extract the point cloud based on the elevation extraction algorithm of the projection point density and remove the irrelevant interference points. First, divide the processed point cloud of the target area into four subspaces S1, S2, S3, and S4, and project each point P in the point clouds of the four space regions i onto the planes where P1, P2, P3, and P4 are located respectively to obtain the projection point P' i .

[0131] Refer to Figure 5 , the target space area is divided into four subspaces S1, S2, S3, and S4, and the corresponding projection planes of the four subspaces are P1, P2, P3, and P4.

[0132] Divide the point cloud projected onto the planes P1, P2, P3, and P4 into multiple small grid regions, and the size of each grid is determined by the density of the collected point cloud and the required accuracy; assume the size of the grid is Δu×Δz, then the number of point clouds in each grid can be obtained by counting the number of points in this area.

[0133] For each grid region G ij , calculate the number of points N within this region ij :

[0134]

[0135] where is the indicator function, when the point P′ k is located within the grid G ij , the function value is 1, otherwise it is 0.

[0136] Calculate the number of points within the grids corresponding to the planes of P1, P2, P3, and P4 respectively, obtain the point cloud of the vertical facade in the target region, and further remove the low-density points to obtain the point cloud set S related to the vertical facades of P1, P2, P3, and P4 vertical :

[0137] S vertical,i ={P′ k |N ij (P′ k )>ρ threshold}

[0138] Based on the point cloud set S vertical,i , obtain the projection heights h of the point cloud in the subspace region on the vertical facades of P1, P2, P3, and P4 respectively target,i , and calculate the average depth value of this region from the projection height values obtained from the four subspace regions From The relative height H between the depth camera 10 and the target point to be measured can be obtained through the geometric projection relationship depth .

[0139] (4.3) Perform real-time dynamic compensation on the relative height data of the drone to the elevation reference plane measured by the laser receiver 9 and the height data of the drone to the target point to be measured measured by the depth camera 10.

[0140] Optionally, after the drone flight control subsystem adjusts the position and attitude of the drone, the drone flight control subsystem obtains the real-time data of the inertial measurement unit inside the flight controller, and obtains the pitch angle θ and roll angle at the current moment

[0141] According to the formula calculate the included angle α between the real-time center of gravity direction of the drone and the direction of gravitational acceleration to perform real-time dynamic adjustment of the attitude angle of the drone;

[0142] Obtain the real-time position information of the real-time dynamic differential positioning module to perform real-time dynamic adjustment of the position of the drone.

[0143] During the data integration process, the elevation data acquisition and processing subsystem uses the angle ɑ between the center-of-gravity direction of the UAV and the direction of gravitational acceleration to perform real-time dynamic compensation on the height data in the laser data and the height data in the three-dimensional space data.

[0144] After angle compensation, the corrected height of the height data in the laser data is:

[0145]

[0146] where H relative is the height initially measured by the laser receiver, and Δα is the error value of the angle between the center-of-gravity direction and the direction of gravitational acceleration;

[0147] After angle compensation, the corrected height of the height data in the three-dimensional space data is:

[0148]

[0149] where H depth is the height initially measured by the depth camera;

[0150] The elevation information of the target point to be measured is:

[0151] H final = {(b + h2) - (a + h1)}

[0152] where h1 is the relative position H′ of the laser receiver with respect to the elevation reference plane relative , h2 is the relative height H′ of the depth camera to the ground of the target point to be measured depth , and a and b are the distances from the installation starting base points of the laser receiver and the depth camera to the bottom of the UAV, respectively.

[0153] As Figure 6 shown, O E XYZ is the inertial coordinate system, and O′ E X′Y′Z′ is the UAV coordinate system; in the ideal state, the UAV coordinate system coincides with the inertial coordinate system. At this time, the center-of-gravity direction of the UAV coincides completely with the direction of gravitational acceleration, and the UAV is in a horizontally stable attitude state; in the actual state, the UAV coordinate system and the inertial coordinate system do not completely coincide. At this time, α is the angle between the center-of-gravity direction of the UAV and the direction of gravitational acceleration, and θ and are the pitch angle and roll angle generated by the UAV during flight, respectively.

[0154] Perform real-time dynamic compensation on the relative height collected by the laser receiver 9 and the height from the ground measured by the depth camera 10 according to the error angle of real-time dynamic compensation; specifically include: converting the body coordinate system of the UAV to the inertial coordinate system, and determining the relationship between the angle α between the center-of-gravity direction of the UAV and the direction of gravitational acceleration and the pitch angle and roll angle based on this conversion; (AsFigure 6 , O E XYZ is an inertial coordinate system, and O′ E X′Y′Z′ is the UAV coordinate system, and the pitch angle θ describes the angle of the UAV rotating around the transverse axis; The angle describing the UAV rotating around the longitudinal axis) The direction matrix between the inertial coordinate system and the UAV coordinate system is expressed as:

[0155]

[0156] Among them, ψ is the yaw angle, which can be ignored in the horizontal attitude angle compensation.

[0157] The direction vector of the gravitational acceleration in the inertial coordinate system is The direction vector of the UAV's center of gravity in the UAV coordinate system is Transform the direction of the UAV's center of gravity from the UAV coordinate system to the inertial coordinate system. According to the direction matrix Get ψ can be ignored in the horizontal attitude angle compensation. Let ψ = 0, then the direction vector of the UAV's center of gravity in the inertial coordinate system is expressed as:

[0158]

[0159] The relationship between the included angle α between the direction of the UAV's center of gravity and the direction of the gravitational acceleration is:

[0160]

[0161] The included angle α between the direction of the UAV's center of gravity and the direction of the gravitational acceleration is:

[0162]

[0163] The real-time dynamic compensation for the error value of the included angle between the direction of the UAV's center of gravity and the direction of the gravitational acceleration is actually the real-time dynamic compensation for the error values of the pitch angle and roll angle generated by the UAV; According to the inertial measurement unit inside the flight controller, the pitch angle θ and roll angle of the UAV at the data acquisition moments of the laser receiver 9 and the depth camera 10 are obtained This angle is the existing error angle. Perform real-time dynamic compensation for this error angle, and perform real-time correction on the relative height collected by the laser receiver 9 and the height from the ground measured by the depth camera 10 according to this error angle, as follows:

[0164] Assume that when the UAV collects height data at a certain moment, the pitch angle measured by using the inertial measurement unit inside the flight controller at this time is θ t , and the roll angle is The initial height measured by the depth camera 10 is H depth , and the initial height collected by the laser receiver 9 is Hrelative ; As shown in Figure 7 , considering the influence of attitude angle error on altitude data, the error angle and altitude correction are dynamically compensated in real time through the following formula:

[0165] According to the collected θ t and , calculate the pitch angle and roll angle that need to be compensated

[0166] Δθ = θ t

[0167]

[0168] At this time, the error value of the included angle between the center of gravity direction and the direction of gravitational acceleration:

[0169]

[0170] The initial measured altitude of the depth camera 10 is H depth , and the corrected altitude after angle compensation is:

[0171]

[0172] The initial measured altitude of the laser receiver 9 is H relative , and the corrected altitude after angle compensation is:

[0173]

[0174] (4.4) The elevation data acquisition and processing subsystem integrates the processed data to calculate the elevation information of the target point to be measured.

[0175] According to Figure 3 , the marked h1 and h2 are the relative positions H ′ relative of the processed elevation reference plane at the laser receiver 9 at a certain moment and the relative height H' from the depth camera 10 to the ground of the target point to be measured depth ; a and b are the distances from the starting base point of the laser receiver 9 installation to the bottom of the UAV and the distance from the starting base point of the depth camera 10 installation to the bottom of the UAV respectively; then Figure 3 the elevation value of the target point to be measured in

[0176] H final = {(b + h2) - (a + h1)}

[0177] By integrating the data of the laser receiver 9 and the data of the depth camera 10, the elevation data generated by the system has high precision and stability, and can output complete elevation information. Further, the elevation information can be used to construct a three-dimensional elevation distribution model of the target area. This model has full coverage and high-precision elevation measurement results, can generate a complete three-dimensional elevation map of the target area, and provides reliable data support for various application scenarios that require accurate elevation information.

[0178] In the embodiment of the present invention, through the combined action of multiple sensors such as the laser level 5, the laser receiver 9, and the depth camera 10, the acquisition of high-precision elevation data is realized. The reference plane generated by the laser level provides a stable elevation reference. The relative position data of the reference plane is obtained through the laser receiver 9, and combined with the height information of the measurement point obtained by the depth camera 10, more accurate elevation data measurement is realized. The data of the laser receiver 9 and the depth camera 10 are processed by an algorithm to remove environmental interference. The included angle between the gravity center direction of the unmanned aerial vehicle and the direction of gravitational acceleration is used to perform real-time dynamic compensation on the height data of the laser receiver 9 and the depth camera 10. Finally, the data is integrated and high-precision three-dimensional elevation information is generated, solving the problems of the existing unmanned aerial vehicle elevation measurement technology and providing high-precision, high-reliability, and anti-interference elevation measurement results.

[0179] It is easy for those skilled in the art to understand that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An unmanned aerial vehicle elevation measurement system based on a laser leveling reference, characterized in that, Including: A laser leveling subsystem, a UAV flight control subsystem, and an elevation data acquisition and processing subsystem; The laser leveling subsystem is used to obtain an elevation reference plane based on the plane formed by the rotating laser emitted by the laser level at the preset elevation benchmark point in the area to be measured; The UAV flight control subsystem is used to receive position information, generate flight control commands, control the UAV to fly to the target points to be measured in sequence along the flight path for measurement, and perform real-time control on the position and attitude of the UAV; The elevation data acquisition and processing subsystem includes a laser receiver, a depth camera, and an airborne computer disposed on the unmanned aerial vehicle; the laser receiver is used to receive the laser signal generated by the laser level; the depth camera is used to measure the three-dimensional spatial information of the area of the target point to be measured; the airborne computer is used to perform data processing on the acquired laser signal and three-dimensional spatial information, obtain the relative position data between the unmanned aerial vehicle and the elevation reference plane according to the laser signal, obtain the relative height between the unmanned aerial vehicle and the target point to be measured according to the three-dimensional spatial information, and integrate the relative position data and the relative height to generate the elevation information of the target point to be measured; wherein the elevation value H in the elevation information of the target point to be measured final = {(b + h2) - (a + h1)}, where h1 is the relative position between the elevation reference plane and the laser receiver, h2 is the relative height from the depth camera to the ground of the target point to be measured; a is the distance from the starting base point of the laser receiver installation to the bottom of the unmanned aerial vehicle, and b is the distance from the starting base point of the depth camera installation to the bottom of the unmanned aerial vehicle.

2. The UAV elevation measurement system according to claim 1, wherein The laser receiver is fixed at the bottom of the UAV, and the receiving direction is perpendicular to the UAV body; The depth camera is installed at the bottom of the UAV, and the viewing angle is vertically downward.

3. The UAV height measurement system according to claim 1, characterized in that: The on-board computer uses the variational mode decomposition method with increased iterative feedback to extract the ideal laser signal from the distance data of the UAV to the elevation reference plane obtained by the laser receiver.

4. The drone elevation measurement system described in claim 1, characterized in that, The on-board computer constructs a cylindrical three-dimensional space processing area with a radius of r according to the three-dimensional space information collected by the depth camera, and divides it into four sub-spaces S1, S2, S3, and S4. By means of point cloud projection in the four space regions, the average depth value of the area to be measured is calculated, and the relative height between the depth camera and the target point to be measured is obtained through the geometric projection relationship.

5. The UAV height measurement system according to claim 1, characterized in that: The real-time adjustment of the position and attitude of the UAV by the UAV flight control subsystem includes: Obtain the real-time data of the inertial measurement unit inside the flight controller to get the pitch angle θ and roll angle at the current moment According to the formula calculate the angle α between the real-time center of gravity direction and the gravitational acceleration direction of the drone to perform real-time dynamic adjustment of the attitude angle of the drone; Obtaining the real-time position information of the real-time kinematic differential positioning module to perform real-time dynamic adjustment on the position of the UAV.

6. The drone elevation measurement system according to claim 5, characterized in that, During the data integration process, the elevation data acquisition and processing subsystem uses the included angle α between the gravity center direction of the UAV and the direction of gravitational acceleration to perform real-time dynamic compensation on the height data in the laser data and the height data in the three-dimensional space data; The height data in the laser data is corrected after angle compensation to: Among them, H relative is the height initially measured by the laser receiver, and Δα is the error value of the angle between the center-of-gravity direction and the direction of gravitational acceleration; The height data in the three-dimensional space data is corrected after angle compensation to: Among them, H depth is the height initially measured by the depth camera; The elevation information of the target point to be measured is: H final = {(b + h2) - (a + h1)} Where h1 is the relative position H′ between the laser receiver and the elevation reference surface relative , h2 is the relative height H′ from the depth camera to the ground of the target point to be measured depth , a and b are the distances from the installation starting points of the laser receiver and depth camera to the bottom of the drone, respectively.

7. The drone elevation measurement system according to claim 1, characterized in that, The laser leveling subsystem includes a level, a tripod, and a laser level; The level is used to establish a preset elevation benchmark point in the area to be measured by the method of gradually transmitting the height difference through the known elevation benchmark point; The laser level is fixedly installed on the tripod, and is used to be horizontally arranged at the preset standard height on the starting elevation benchmark point in the area to be measured, so that the rotating laser emitted forms an elevation reference plane, which serves as the elevation measurement benchmark for the UAV in the area to be measured.

8. The drone elevation measurement system according to claim 1, characterized in that, The UAV flight control subsystem includes an on-board computer, a flight controller, and a real-time kinematic differential positioning module; The on-board computer is used to receive the position information of the area to be measured, generate the optimal flight path, and send the flight control commands to the flight controller; The real-time kinematic differential positioning module is used to obtain the position information of the UAV during the flight process; The flight controller is used to receive the flight control commands and control the UAV to fly to each target position to be measured in sequence; at the same time, it is used for the real-time adjustment of the position and attitude of the UAV during the flight and data acquisition process.

Images