Laser sensor measurement error surveying and mapping calibration method

Abstract

The invention relates to a laser sensor measurement error surveying and mapping calibration method which is suitable for the technical field of unmanned aerial vehicle laser remote sensing. The method comprises the following steps: determining the actual relative elevation of each ground control point relative to the unmanned aerial vehicle based on the three-dimensional coordinates of the unmanned aerial vehicle and the actual coordinate position of each ground control point; extracting data points corresponding to each ground control point in the ground surface point cloud data, and obtaining laser sensor attitude information when the laser sensor collects the data points corresponding to the ground control points; determining the measured relative elevation of each ground control point relative to the unmanned aerial vehicle based on the three-dimensional coordinates of the unmanned aerial vehicle and the data points corresponding to each ground control point; determining an elevation error corresponding to each ground control point based on the actual relative elevation and the measured relative elevation of each ground control point; obtaining an angle measurement regression fitting model based on the laser sensor attitude information and the elevation error corresponding to each ground control point; and calibrating the angle measurement error of the laser sensor based on the angle measurement regression fitting model.

Description

Technical Field

[0001] The present invention relates to a method for mapping and calibrating the measurement error of a laser sensor, which is applicable to the field of unmanned aerial vehicle (UAV) laser remote sensing technology. Background Art

[0002] The UAV-borne laser remote sensing technology is the result of the increasing maturity of UAV and laser sensor technologies, and is the most commonly used topographic survey method in the current surveying and mapping field. In recent years, with the development of the autonomous driving industry, the types of laser sensors have increased day by day. While the performance has been improved, the cost has gradually decreased. There have emerged many products in the industry that use vehicle-grade laser sensors integrated as airborne surveying and mapping laser payloads.

[0003] However, since the laser sensor itself is not designed for the surveying and mapping industry, such laser sensors have many disadvantages, such as large light spots, unclear details, difficult to accurately calibrate the prism parameters according to the surveying and mapping accuracy, and large deviations in angle measurement and distance measurement accuracy. Therefore, it can only be used for small-scale topographic map measurement with low accuracy requirements.

[0004] In existing surveying and mapping products, most integrations directly use the calibration results provided by laser sensor manufacturers without performing surveying and mapping-level calibration. The laser sensors produced by manufacturers are calibrated at short distances in the laboratory to meet the short-distance requirements of autonomous driving. The calibration results cannot be stably used for long-distance surveying and mapping. Some surveying instrument integrators build their own ground control fields and use the high-precision prior point clouds of buildings for equipment calibration. However, due to the divergence of scan lines in the distance and inability to converge, accurate and efficient calibration effects at long distances cannot be achieved. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: in view of the above problems, to provide a method for mapping and calibrating the measurement error of a laser sensor.

[0006] The technical solution adopted by the present invention is: a method for mapping and calibrating the measurement error of a laser sensor, comprising: Obtaining the three-dimensional coordinates of the UAV collected by a positioning device, the surface point cloud data collected by a laser sensor on the UAV, and the attitude information of the laser sensor collected by an inertial navigation device, wherein the surface point cloud data corresponds to the area where multiple ground control points are located; Based on the three-dimensional coordinates of the UAV and the actual coordinate positions of each ground control point, determining the actual relative elevation of each ground control point relative to the UAV; Extracting the data points corresponding to each ground control point in the surface point cloud data, and obtaining the attitude information of the laser sensor when collecting the data points corresponding to the ground control points; Based on the three-dimensional coordinates of the UAV and the data points corresponding to each ground control point, determining the measured relative elevation of each ground control point relative to the UAV; Based on the actual relative elevation and the measured relative elevation of each ground control point, determine the elevation error corresponding to each ground control point; Based on the attitude information and elevation error of the laser sensor corresponding to each ground control point, obtain an angular measurement regression fitting model; Based on the angular measurement regression fitting model, calibrate the angular measurement error of the laser sensor.

[0007] The obtaining of the angular measurement regression fitting model based on the attitude information and elevation error of the laser sensor corresponding to each ground control point includes: Based on the measured angle of the laser sensor corresponding to each ground control point , and the elevation error corresponding to each ground control point , fit the objective function; The objective function is: ; where D is the vertical distance from the laser sensor to the real ground; a and b are parameters to be fitted; and ε is the residual.

[0008] It further includes: Based on the three-dimensional coordinates of the UAV, determine the actual relative elevation of the UAV relative to the ground below; Based on the distance collected when the laser sensor is perpendicular to the ground, combined with the actual relative elevation of the UAV relative to the ground below, determine the elevation difference , and according to the angle between the 0 axis of the laser sensor and the ground vertical line at this time determine the ranging systematic error of the laser sensor; Based on the ranging systematic error of the laser sensor, calibrate the ranging error of the laser sensor.

[0009] The ranging systematic error of the laser sensor is: ; where is the ranging systematic error.

[0010] A device for mapping and calibrating the measurement error of a laser sensor includes: A data acquisition module, configured to acquire the three-dimensional coordinates of the UAV collected by a positioning device, the surface point cloud data collected by a laser sensor on the UAV, and the attitude information of the laser sensor collected by an inertial navigation device, where the surface point cloud data corresponds to the area where multiple ground control points are located; An actual elevation calculation module, configured to determine the actual relative elevation of each ground control point relative to the UAV based on the three-dimensional coordinates of the UAV and the actual coordinate positions of each ground control point; An information extraction module, configured to extract data points corresponding to each ground control point in the ground point cloud data, and obtain the attitude information of the laser sensor when the laser sensor collects the data points corresponding to the ground control points; A measured elevation calculation module, configured to determine the measured relative elevation of each ground control point relative to the unmanned aerial vehicle based on the three-dimensional coordinates of the unmanned aerial vehicle and the data points corresponding to each ground control point; An elevation error calculation module, configured to determine the elevation error corresponding to each ground control point based on the actual relative elevation and the measured relative elevation of each ground control point; A data fitting module, configured to obtain an angular regression fitting model based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point; An error calibration module, configured to perform angular error calibration of the laser sensor based on the angular regression fitting model.

[0011] A storage medium, on which a calculator program executable by a processor is stored, and when the computer program is executed, the steps of the method for mapping and calibrating the measurement error of the laser sensor are implemented.

[0012] A device for mapping and calibrating the measurement error of a laser sensor, having a memory and a processor, and a calculator program executable by the processor is stored on the memory, and when the computer program is executed, the steps of the method for mapping and calibrating the measurement error of the laser sensor are implemented.

[0013] A system for mapping and calibrating the measurement error of a laser sensor, comprising: An unmanned aerial vehicle, on which a laser sensor is mounted; A positioning device, mounted on the unmanned aerial vehicle, configured to collect the three-dimensional coordinates of the unmanned aerial vehicle; An inertial navigation device, mounted corresponding to the laser sensor, configured to collect the attitude information of the laser sensor; An on-board computer, configured to synchronize the time of the positioning device, the inertial navigation device, and the laser sensor, and record the data collected by the positioning device, the inertial navigation device, and the laser sensor; A data processing device, configured to be able to implement the steps of the method for mapping and calibrating the measurement error of the laser sensor.

[0014] The beneficial effects of the present invention are as follows: The present invention performs fitting based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point to obtain an angular regression fitting model, and thus performs angular error calibration of the laser sensor based on the angular regression fitting model, improving the accuracy of the equipment results to achieve higher-precision mapping operations.

[0015] Based on the distance collected when the laser sensor is perpendicular to the ground, combined with the actual relative elevation of the drone relative to the ground below, the elevation difference is determined, and the ranging system error of the laser sensor is determined according to the angle between the 0 axis of the laser sensor and the ground vertical line. Furthermore, based on the ranging system error of the laser sensor, the ranging error calibration of the laser sensor is carried out to improve the accuracy of the equipment results, so as to achieve higher-precision surveying and mapping operations. Brief Description of the Drawings

[0016] Figure 1 It is a schematic diagram of the internal prism and optical path propagation of the laser sensor.

[0017] Figure 2 It is a schematic diagram of the result error caused by the laser sensor error.

[0018] Figure 3 It is a schematic diagram of the design of the calibration site control scheme.

[0019] Figure 4 It is a schematic diagram of the laser sensor calibration device.

[0020] Figure 5 It is a structural diagram of the laser sensor calibration device.

[0021] Figure 6 It is a schematic diagram of the calibration flight route.

[0022] Figure 7 It is a flowchart of the method for calibrating the measurement error of the laser sensor for surveying. Detailed Description of the Invention

[0023] Embodiment 1: In this embodiment, the laser sensor 6 inside is as Figure 1 shown, where 1 and 2 are two groups of wedge-shaped lenses, and the two groups of lenses are respectively driven to rotate by two groups of motors. The laser emitted by the laser 4 is converged by the lens 3 and then parallelly emitted towards the lens 2. After the lens 2 refracts the light, it enters the lens 1. Different rotation angles between the lenses 1 and 2 can control the laser to be emitted in different directions. The rotation relationship and mirror surface flatness between the lenses 1 and 2 will affect the accuracy of the laser for measuring the angle of the ground. At the same time, as a ranging device, the laser will also have a ranging error. Usually, the manufacturer will provide a calibration result of the indoor distance to ensure the use of short-distance measurement for autonomous driving.

[0024] Figure 1 In [figure] 5 is the control circuit part of the laser sensor, and the drawing of the receiving chip APD is omitted in the schematic diagram.

[0025] As Figure 2As shown in the figure, due to the errors in angle measurement and distance measurement, when the laser sensor 6 performs mapping measurement in the repeated scanning mode, there will be an error between the measured ground 10 and the actual ground 7. Due to the characteristics of the laser lens, this error usually manifests as the measured ground bending upward or downward, and there is an error between the actual field of view angle edge 9 and the measurement edge 8.

[0026] Based on the principle and existence form of the above errors, in this embodiment, in a certain frame or several frames of data collected by the laser sensor 6, by measuring the coordinates of the actual ground, the measurement error can be calculated, and this error is the Z-axis error in the elevation direction in the initial state. Through regression fitting, the Z-axis error is brought into the regression fitting equation, and the elevation error directly below the device is used as the ranging system error, that is, the angle measurement error and distance measurement error of the device can be solved.

[0027] As Figure 3 shown, in this embodiment, a plurality of ground control points 12 are arranged on the ground. The plurality of ground control points 12 are arranged in a straight line, and the layout distance is 1 m. The coordinates of the control points are in the CGCS2000 coordinate system, and the acquisition accuracy should be less than 1 cm. When calibrating, the UAV flight path is perpendicular to the straight line where the ground control points 12 are located, and a certain frame of laser sensor point cloud data 11 collected by the laser sensor 6 covers a plurality of ground control points 12.

[0028] For the flight calibration of the laser sensor, as Figure 4 、 Figure 5 shown, in this embodiment, the inertial navigation device 15, the airborne computer 16, and the laser sensor 6 are fixedly installed and mounted on the UAV 13, and a positioning device 14 is installed on the UAV 13.

[0029] In this example, the positioning device 14 records the three-dimensional coordinate position of the UAV platform at a rate of 20 Hz, the inertial navigation device 15 records the attitude information of the laser sensor 6 at a rate of 300 Hz, and the airborne computer 16 realizes the time synchronization of the positioning device 14, the inertial navigation device 15, and the laser sensor 6, and records all data, and the time synchronization accuracy is less than 1 ms.

[0030] Before calibrating the laser sensor, calibrate the relative relationship between the inertial navigation device 15 and the laser sensor 6, that is, complete the exterior orientation calibration of the entire system.

[0031] In this embodiment, the UAV 13 flies according to 3 designed parallel flight paths. As Figure 6 shown, when the device is calibrated, the middle flight path of the flight path is used for calibration. The purpose of flying three flight paths is to judge whether the entire system has completed the exterior orientation calibration by the data layering situation in the overlapping area between the three flight paths. The exterior orientation calibration of the inertial navigation device and the laser sensor is the basis of this calibration scheme.

[0032] As shown Figure 7 in the figure, this embodiment provides a method for mapping and calibrating the measurement error of a laser sensor, which specifically includes the following steps:

[0033] S1. Obtain the three-dimensional coordinates of the drone collected by the positioning device, the ground point cloud data collected by the laser sensor on the drone, and the attitude information of the laser sensor collected by the inertial navigation device, where the ground point cloud data corresponds to the area where multiple ground control points are located.

[0034] S2. Calibrate the angle measurement error of the laser sensor.

[0035] S201. Based on the three-dimensional coordinates of the drone and the actual coordinate positions of each ground control point, determine the actual relative elevation of each ground control point relative to the drone.

[0036] S202. Extract the data points corresponding to each ground control point in the ground point cloud data, and obtain the attitude information of the laser sensor when collecting the data points corresponding to the ground control points.

[0037] S203. Based on the three-dimensional coordinates of the drone and the data points corresponding to each ground control point, determine the measured relative elevation of each ground control point relative to the drone.

[0038] S204. Based on the actual relative elevation and the measured relative elevation of each ground control point, determine the elevation error corresponding to each ground control point.

[0039] S205. Based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point, obtain an angle measurement regression fitting model.

[0040] In this embodiment, based on the measured angle of the laser sensor corresponding to each ground control point , and the elevation error corresponding to each ground control point , fit the objective function.

[0041] According to the error formation principle of the laser, establish the error model as follows. The relationship between the measured elevation error and the angle measurement error is: ; where: D is the vertical distance from the laser to the true ground; is the measured angle of the laser (with 0 degrees vertically downward as the reference, symmetric left and right); is the angle measurement error function, which is an odd function and satisfies .

[0042] Assume that the angle measurement error is represented by an odd polynomial (symmetry): ; where a, b, c, … are parameters to be fitted.

[0043] Simplified error model formula

[0044] Retain the main low-order terms (such as the first-order and third-order terms), and substitute them to get: ; Equivalently: ; According to the arranged ground control points, calculate different angles θ of the control point positions and the corresponding elevation errors .

[0045] Fit parameters a and b through multiple linear regression, and the objective function is: ; where ε is the residual.

[0046] S206. Calibrate the angle measurement error of the laser sensor based on the angle regression fitting model.

[0047] Embodiment 2: This embodiment is a method for mapping and calibrating the measurement error of a laser sensor, which specifically includes the following steps: A. Obtain the three-dimensional coordinates of the unmanned aerial vehicle collected by the positioning device, the surface point cloud data collected by the laser sensor on the unmanned aerial vehicle, and the attitude information of the laser sensor collected by the inertial navigation device, where the surface point cloud data corresponds to the area where multiple ground control points are located; B. Calibrate the ranging error of the laser sensor.

[0048] B1. Based on the three-dimensional coordinates of the unmanned aerial vehicle, determine the actual relative elevation of the unmanned aerial vehicle relative to the ground below;

[0049] B2. Based on the distance collected when the laser sensor is perpendicular to the ground, combined with the actual relative elevation of the unmanned aerial vehicle relative to the ground below, determine the elevation difference , and determine the ranging system error of the laser sensor according to the angle between the 0 axis of the laser sensor and the vertical line of the ground at this time.

[0050] The ranging system error of the laser sensor is: ; where is the ranging system error.

[0051] C. Calibrate the angle measurement error of the laser sensor.

[0052] C1. Determine the actual relative elevation of each ground control point relative to the UAV based on the three-dimensional coordinates of the UAV and the actual coordinate positions of each ground control point.

[0053] C2. Extract the data points corresponding to each ground control point in the surface point cloud data, and obtain the attitude information of the laser sensor when collecting the data points corresponding to the ground control points.

[0054] C3. Determine the measured relative elevation of each ground control point relative to the UAV based on the three-dimensional coordinates of the UAV and the data points corresponding to each ground control point.

[0055] C4. Determine the elevation error corresponding to each ground control point based on the actual relative elevation and the measured relative elevation of each ground control point.

[0056] C5. Obtain the angle measurement regression fitting model based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point.

[0057] In this embodiment, based on the measured angles of the laser sensor corresponding to each ground control point , and the elevation errors corresponding to each ground control point , fit the objective function.

[0058] Establish the error model according to the error formation principle of the laser as follows. The relationship between the measured elevation error and the angle measurement error is: ; where: D is the vertical distance from the laser to the true ground; is the measured angle of the laser (symmetric left and right with 0 degrees vertically downward as the reference); is the angle measurement error function, which is an odd function and satisfies .

[0059] Assume that the angle measurement error is represented by an odd polynomial (symmetry): ; where a, b, c,... are the parameters to be fitted.

[0060] The simplified error model formula retains the main low-order terms (such as the first-order and third-order terms), and substituting them gives: ; Equivalently: ; According to the arranged ground control points, calculate the different angles θ of the control point positions, and the corresponding elevation errors .

[0061] By fitting parameters a and b through multiple linear regression, the objective function is as follows: ; where ε is the residual.

[0062] D. Based on the ranging system error of the laser sensor, calibrate the ranging error of the laser sensor; based on the angular regression fitting model, calibrate the angular error of the laser sensor.

[0063] Through the above process, the angular error and ranging error of the laser can be corrected. Substitute the corrected error model and the parameters of the regression fitting into the point cloud fusion, and then the corrected real ground data can be obtained, completing the high-precision calibration of the angle measurement and ranging of the laser sensor.

[0064] Embodiment 3: This embodiment is a device for mapping and calibrating the measurement error of a laser sensor, specifically including: A data acquisition module, which is used to acquire the three-dimensional coordinates of the unmanned aerial vehicle collected by the positioning device, the surface point cloud data collected by the laser sensor on the unmanned aerial vehicle, and the attitude information of the laser sensor collected by the inertial navigation device, where the surface point cloud data corresponds to the area where multiple ground control points are located; An actual elevation calculation module, which is used to determine the actual relative elevation of each ground control point relative to the unmanned aerial vehicle based on the three-dimensional coordinates of the unmanned aerial vehicle and the actual coordinate positions of each ground control point; An information extraction module, which is used to extract the data points corresponding to each ground control point in the surface point cloud data and obtain the attitude information of the laser sensor when collecting the data points corresponding to the ground control points; A measured elevation calculation module, which is used to determine the measured relative elevation of each ground control point relative to the unmanned aerial vehicle based on the three-dimensional coordinates of the unmanned aerial vehicle and the data points corresponding to each ground control point; An elevation error calculation module, which is used to determine the elevation error corresponding to each ground control point based on the actual relative elevation and the measured relative elevation of each ground control point; A data fitting module, which is used to obtain an angular regression fitting model based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point; An error calibration module, which is used to calibrate the angular error of the laser sensor based on the angular regression fitting model.

[0065] Embodiment 4: This embodiment is a storage medium, on which a calculator program executable by a processor is stored. When the computer program is executed, the steps of the method for mapping and calibrating the measurement error of the laser sensor described in Embodiment 1 or 2 are implemented.

[0066] Embodiment 5: This embodiment is a laser sensor measurement error mapping and calibration device, which has a memory and a processor. A calculator program that can be executed by the processor is stored on the memory. When the computer program is executed, the steps of the laser sensor measurement error mapping and calibration method described in Embodiment 1 or 2 are implemented.

[0067] Embodiment 6: This embodiment is a laser sensor measurement error mapping and calibration system, which specifically includes: a laser sensor calibration device, a data processing device, etc.

[0068] In this example, the laser sensor calibration device has a drone, a positioning device, an inertial navigation device, an on-board computer, etc. Among them, the drone is equipped with a laser sensor; the positioning device is installed on the drone and is used to collect the three-dimensional coordinates of the drone; the inertial navigation device is installed corresponding to the laser sensor and is used to collect the attitude information of the laser sensor; the on-board computer is used to achieve time synchronization of the positioning device, the inertial navigation device and the laser sensor, and record the data collected by the positioning device, the inertial navigation device and the laser sensor.

[0069] In this embodiment, the data processing device is configured to implement the steps of the laser sensor measurement error mapping and calibration method described in Embodiment 1 or 2.

Claims

1. A calibration method for mapping measurement errors of a laser sensor, characterized in that, Including: Obtain the three-dimensional coordinates of the drone collected by the positioning device, the ground point cloud data collected by the laser sensor on the drone, and the attitude information of the laser sensor collected by the inertial navigation device, where the ground point cloud data corresponds to the area where multiple ground control points are located; Based on the three-dimensional coordinates of the drone and the actual coordinate positions of each ground control point, determine the actual relative elevation of each ground control point relative to the drone; Extract the data points corresponding to each ground control point in the ground point cloud data, and obtain the attitude information of the laser sensor when the laser sensor collects the data points corresponding to the ground control points; Based on the three-dimensional coordinates of the drone and the data points corresponding to each ground control point, determine the measured relative elevation of each ground control point relative to the drone; Based on the actual relative elevation and the measured relative elevation of each ground control point, determine the elevation error corresponding to each ground control point; Based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point, obtain an angular regression fitting model; Based on the angular regression fitting model, perform calibration of the angular error of the laser sensor.

2. The laser sensor measurement error mapping and calibration method according to claim 1, wherein, The obtaining of the angular regression fitting model based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point includes: Measuring the angles based on the measurement angles of the laser sensors corresponding to each ground control point , and the elevation errors corresponding to each ground control point , and fitting the objective function; The objective function is as follows: ; Where D is the vertical distance from the laser sensor to the real ground; a and b are parameters to be fitted; ε is the residual.

3. The laser sensor measurement error mapping and calibration method according to claim 1, characterized in that Also including: Based on the three-dimensional coordinates of the drone, determine the actual relative elevation of the drone relative to the ground below; Based on the distance collected when the laser sensor is perpendicular to the ground and combined with the actual relative elevation of the drone relative to the ground below, the elevation difference is determined , and according to the angle between the 0 axis of the laser sensor and the ground perpendicular line at this time the ranging system error of the laser sensor is determined; Based on the ranging systematic error of the laser sensor, perform calibration of the ranging error of the laser sensor.

4. The method for mapping and calibrating the measurement error of a laser sensor according to claim 3, characterized in that, The ranging systematic error of the laser sensor is: ； Among them, is the ranging system error.

5. A calibration device for mapping measurement errors of a laser sensor, characterized in that, Including: A data acquisition module for obtaining the three-dimensional coordinates of the drone collected by the positioning device, the ground point cloud data collected by the laser sensor on the drone, and the attitude information of the laser sensor collected by the inertial navigation device, where the ground point cloud data corresponds to the area where multiple ground control points are located; An actual elevation calculation module for determining the actual relative elevation of each ground control point relative to the drone based on the three-dimensional coordinates of the drone and the actual coordinate positions of each ground control point; An information extraction module for extracting the data points corresponding to each ground control point in the ground point cloud data and obtaining the attitude information of the laser sensor when the laser sensor collects the data points corresponding to the ground control points; A measured elevation calculation module for determining the measured relative elevation of each ground control point relative to the drone based on the three-dimensional coordinates of the drone and the data points corresponding to each ground control point; An elevation error calculation module for determining the elevation error corresponding to each ground control point based on the actual relative elevation and the measured relative elevation of each ground control point; A data fitting module for obtaining an angular regression fitting model based on the attitude information of the laser sensor and the elevation error corresponding to each ground control point; An error calibration module for performing calibration of the angular error of the laser sensor based on the angular regression fitting model.

6. A storage medium, on which a calculator program executable by a processor is stored, characterized in that When the computer program is executed, it implements the steps of the method for calibrating the measurement error mapping of the laser sensor according to any one of claims 1 to 4.

7. A calibration device for mapping measurement errors of a laser sensor, having a memory and a processor, with a calculator program stored on the memory that can be executed by the processor, characterized in that, When the computer program is executed, it implements the steps of the method for calibrating the measurement error mapping of the laser sensor according to any one of claims 1 to 4.

8. A laser sensor measurement error mapping and calibration system, characterized in that, Including: A drone equipped with a laser sensor on it; A positioning device, installed on a drone, for collecting the three-dimensional coordinates of the drone; An inertial navigation device, installed corresponding to the laser sensor, for collecting the attitude information of the laser sensor; An on-board computer, for achieving time synchronization of the positioning device, the inertial navigation device and the laser sensor, and recording the data collected by the positioning device, the inertial navigation device and the laser sensor; A data processing device, configured to be able to implement the steps of the laser sensor measurement error mapping and calibration method described in any one of claims 1 to 4.

Images