Scanning method and control system for improving distribution uniformity of laser radar point cloud

Abstract

The invention belongs to the technical field of laser radars, and provides a scanning method for improving the distribution uniformity of laser radar point clouds, which converts a constant-speed scanning mode of a prism in a period into a multi-section variable-speed scanning mode, and specifically comprises the following steps: S1, determining the variable-speed times of sectional scanning in a period; s2, determining the initial scanning speed of the prism; s3, determining the scanning speed of the prism after speed change; s4, determining speed change delay time; and S5, determining the angle position of each speed change in one period. According to the technical scheme of the invention, the problem of uneven distribution of laser radar imaging point clouds in an airborne platform can be effectively improved, the point cloud utilization rate is improved, point cloud resources are saved, the repetition frequency of a laser is reduced, and the cost is saved.

Description

Technical Field

[0001] The present invention belongs to the technical field of laser radar, and in particular relates to a scanning method and a control system for improving the uniformity of laser radar point cloud distribution. Background Art

[0002] LiDAR is an active detection imaging system that uses laser as a detection medium. Its principle is to radiate a laser beam to the target and detect the laser echo signal to resolve the target's distance information. Combined with a three-dimensional scanning system, the target's spatial coordinate information can be obtained. Among them, the scanning system not only determines the size of the LiDAR, but also determines the angular resolution of the LiDAR. For most existing mechanical and semi-solid LiDARs, there are movable mechanical scanning components inside, and they all rotate at a constant speed. The imaging range and angular resolution of this type of LiDAR are fixed. On some special platforms, LiDARs with uniform scanning imaging cannot meet the requirements of high resolution. For example, on a vehicle-mounted platform, the road surface needs to be uniformly imaged. Usually, local encrypted scanning is used to scan and control the region of interest (ROI) to improve the imaging capability of the target. On airborne platforms, prisms are usually used as scanning elements to achieve three-dimensional imaging as the aircraft moves forward. The prism scanning trajectory is usually arranged in a forward overlapping manner in the form of a circle. Due to the geometric shape, there is usually more overlap on both sides and no overlap in the middle, which causes the point cloud density on both sides to be much greater than the point cloud density in the middle. In order to achieve high-resolution requirements for ground targets, the point cloud density at the sparsest point in the middle is usually used as a feasibility criterion, which seriously wastes point cloud resources, limits imaging efficiency, increases laser power consumption, and increases development costs. Summary of the invention

[0003] 1. Technical issues to be resolved

[0004] The technical problem to be solved by the present invention is: the problem of uneven point cloud distribution caused by prism scanning in laser radar.

[0005] (II) Technical solution

[0006] In order to solve the above technical problems, the present invention provides a scanning method for improving the uniformity of laser radar point cloud distribution, which converts the uniform speed scanning mode within one cycle of the prism into a multi-segment variable speed scanning mode, and specifically includes the following steps:

[0007] S1, determine the number of speed changes of segmented scanning within one cycle;

[0008] S2, determine the initial scanning speed of the prism;

[0009] S3, determining the scanning speed of the prism after the speed change;

[0010] S4, determining the speed change delay time;

[0011] S5, determining the angular position of each speed change within a cycle;

[0012] S6, determining the acceleration value according to the initial scanning speed, the scanning speed after the speed change and the speed change delay time;

[0013] S7, determine the height of the equipment, laser repetition rate, lateral flight speed, beam deflection angle, and determine the point cloud density distribution curve after speed change;

[0014] S8, taking the variance value of the point cloud density distribution curve as a criterion, adjusting the magnitude of the acceleration value to optimize the uniformity of the point cloud density distribution;

[0015] S9. Calculate the optimal variable speed scanning conditions based on the actual achievable acceleration level and the initial state of the equipment.

[0016] Furthermore, the point cloud density distribution curve is a curve obtained by performing point counts in a certain size grid in the scanning trajectory diagram, obtaining the number of point clouds per unit area, and then performing point cloud density statistics in a certain direction.

[0017] The variance value of the point cloud density distribution curve is obtained by solving the variance of the data on the point cloud density distribution curve.

[0018] Furthermore, by the height h of the equipment and the laser repetition rate f laser , aircraft lateral flight speed v x , beam deflection angle α and initial angular velocity w 0 , angle values ​​θ at different times, t is time, calculated according to the following formula:

[0019]

[0020] Calculate the scanning trajectory of the prism within the time period t;

[0021] By intercepting the scanning trajectory range of the uniformly intersecting part, obtaining the number of point clouds per unit area, and calculating the point cloud density, the distribution diagram of the point cloud density can be obtained, and the variance of the data points on the curve can be obtained.

[0022] The calculation method of the optimal variable speed scanning condition is as follows:

[0023] S901, according to the angular velocity variation range [ω 1， ω 2 ], the range of speed change delay time Δt [t 1 , t 2 ], initial speed change angle position θ 0 , determine a set of boundary conditions and calculate the corresponding point cloud density distribution variance;

[0024] S902, compare the variance value obtained in S901 with the variance value of the point cloud distribution density obtained by uniform scanning. If it is greater than the variance of the point cloud distribution density obtained by uniform scanning, the set of boundary conditions is discarded, and a new set of boundary conditions is determined and returned to S901 for recalculation; if the obtained variance value is less than the variance of the point cloud density distribution obtained by uniform scanning, proceed to S903;

[0025] S903, saving the variance that meets the conditions and the boundary conditions of the group of inputs, and determining new boundary conditions again to return to S901 for loop calculation until all boundary conditions are calculated;

[0026] S904, compare all the variance values ​​that meet the conditions, and select the input boundary condition corresponding to the minimum variance value as the optimal solution, that is, the optimal variable speed scanning condition.

[0027] The present invention also provides a control system for improving the uniformity of laser radar point cloud distribution, including a scanning prism module, a scanning prism driving module, and a scanning control device; the scanning prism module is used to realize deflection scanning of the laser beam to form a continuous scanning trajectory; the scanning prism driving module is used to drive the prism to rotate in a predetermined manner; the scanning control device is used to download the optimal variable speed scanning conditions to the scanning control device in the form of a programming language to realize variable speed control of the scanning prism.

[0028] Furthermore, the scanning prism module may be a scanning rotating mirror or a scanning galvanometer mirror.

[0029] Furthermore, the scanning prism driving module is a driving electronic device for other scanning components.

[0030] Furthermore, the scanning prism control module is any electronic device with programming, storage and communication functions.

[0031] Furthermore, the scanning prism module and the scanning control device communicate with the host computer module, and the host computer module is any computer reading and writing device with communication function.

[0032] (III) Beneficial effects

[0033] Compared with the prior art, the present invention has the following beneficial effects: it can effectively improve the problem of uneven distribution of laser radar imaging point clouds in airborne platforms, improve point cloud utilization, save point cloud resources, reduce laser repetition rate, and save costs. In particular, it can be clearly found by those skilled in the art that this method is not only applicable to prism scanning methods, but also can be used for reference in scanning imaging methods such as galvanometers and rotating mirrors to perform point cloud homogenization adjustment. Therefore, this method has good reference significance for mechanical and semi-solid laser radar homogenization imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the prism multi-stage variable speed scanning method;

[0035] Figure 2 It is a flow chart of point cloud density distribution variance calculation;

[0036] Figure 3 It is a schematic diagram of uniform scanning trajectory and point cloud density determination;

[0037] Figure 4 is a flow chart of the calculation of the optimal variable speed scanning condition;

[0038] Figure 5 It is a schematic diagram of determining the initial speed change angle;

[0039] Figure 6 It is a schematic diagram of the variance statistics of point cloud density distribution under multi-scan boundary conditions;

[0040] Figure 7 It is a schematic diagram of the point cloud density distribution statistics obtained under a set of optimal variable speed scanning conditions;

[0041] Figure 8 It is a block diagram of a prism variable speed scanning control system. DETAILED DESCRIPTION

[0042] In order to make the purpose, content, and advantages of the present invention more clear, the specific implementation methods of the present invention are further described in detail below in conjunction with the accompanying drawings and examples.

[0043] The first purpose of the present invention is to propose a prism variable speed scanning method. This method changes the prism from uniform rotation to controllable multi-stage variable speed rotation. Through computational simulation, the optimal variable speed scanning method can be obtained, thereby achieving uniform imaging in the entire area. This can effectively improve the imaging efficiency of the lidar, save point cloud resources, and reduce costs.

[0044] A prism variable speed scanning method is provided, which converts the uniform speed scanning mode within a prism cycle into a multi-segment variable speed scanning mode, including: determining the number of speed changes of segmented scanning within a cycle; determining the initial scanning speed of the prism; determining the scanning speed of the prism after the speed change; determining the speed change delay time; determining the angular position of each speed change within a cycle; the acceleration value can be determined by the initial scanning speed, the scanning speed after the speed change and the speed change delay time. At the same time, the height of the equipment, the laser repetition rate, the lateral flight speed, and the beam deflection angle can be determined to determine the point cloud density distribution curve after the speed change, and the variance value of the point cloud density distribution curve is used as a criterion to adjust the magnitude of the acceleration value to optimize the uniformity of the point cloud density distribution. Therefore, the method can calculate the optimal variable speed scanning condition according to the actual achievable acceleration level and the initial state of the equipment, thereby effectively improving the imaging uniformity of the laser radar, improving the point cloud utilization rate, and saving point cloud resources.

[0045] The point cloud density distribution curve is a curve obtained by performing point counts in a certain size grid in the scanning trajectory diagram, obtaining the number of point clouds per unit area, and then performing point cloud density statistics in a certain direction.

[0046] The variance value of the point cloud density distribution curve is obtained by solving the variance of the data on the point cloud density distribution curve.

[0047] The second object of the present invention is to provide a prism variable speed scanning control system, which includes a scanning prism module, a scanning prism driving module, and a scanning control device. The scanning prism module is used to realize the deflection scanning of the laser beam to form a continuous scanning trajectory; the scanning prism driving module is used to drive the prism to rotate in a predetermined manner; the scanning control device is used to control programming, and the optimal variable speed scanning conditions described in the prism variable speed scanning method are downloaded to the scanning control device in the form of a programming language to realize variable speed control of the scanning prism.

[0048] The optimal variable speed scanning condition refers to the variable speed angle position, the initial scanning speed, the scanning speed after the speed change, and the variable speed delay time when the variance of the point cloud density distribution curve is the smallest, which are calculated by the prism variable speed scanning method.

[0049] Optionally, the register addresses of the four variables of speed change angle position, initial scanning speed, scanning speed after speed change, and speed change delay time can be opened during programming to facilitate reading, writing and modification. The register address is the memory of the scanning control device.

[0050] Embodiment 1

[0051] The segmented variable speed scanning method in this embodiment divides a uniform speed scanning cycle into N segments of variable speed scanning. Optionally, in this step, the number N of segments of the variable speed scanning can be determined according to actual needs.

[0052] like Figure 1 As shown, in this embodiment, a scanning cycle is divided into four variable speed scanning strokes, and the four variable speed scanning strokes correspond to two groups of scanning strokes with different angular velocities, wherein:

[0053] ω 1 is the initial scanning speed, ω 2 is the scanning speed after speed change, and ω 2 >ω 1 , if the positive direction of the X-axis is 0°, that is, in this embodiment, the scanning speed near 0° and 180° is smaller than the scanning speed near 90° and 270°, so as to meet the control of the number of scanning points in different areas. In addition, this embodiment also includes four acceleration and deceleration areas, where Δt is the speed change delay time, that is, the angular velocity is changed from ω 1 Increase to ω 2 , or by ω 2 Reduce to ω 1 The time required. 0 is the initial speed change angle position, that is, the prism moves at the initial speed ω 1 Rotate to θ 0 The speed change starts at the position. For the convenience of calculation, the other three speed change positions in this embodiment are 180°-θ 0 、180°+θ 0 and 360°-θ 0 Thus, variable scanning within one cycle can be achieved.

[0054] In this embodiment, in order to obtain the optimal variable speed scanning parameters, the initial scanning speed ω can be 1 , scanning speed after speed change ω 1 , initial speed change angle position θ 0 The speed change delay time Δt is used for regulation, and the evaluation basis is the variance value of the point cloud density distribution curve.

[0055] In the above steps, the variance value of the point cloud density distribution curve is obtained mainly according to Figure 2 To proceed with the process shown in the figure, first determine the known conditions in S201, including the height h of the equipment and the laser repetition rate f. laser , aircraft lateral flight speed v x , beam deflection angle α and initial angular velocity w 0 , the angle value at different times is θ, t is time, and the calculation formula is as follows:

[0056]

[0057] The scanning trajectory within the time period t can be calculated from the above formula, such as Figure 3 The figure shows the scanning trajectory of the prism when it is scanned at a constant speed. Figure 2 In step S203, the scanning track range of the uniformly intersecting part is intercepted to obtain the number of point clouds per unit area, that is, using Figure 3 The black grid lines shown divide the scanning trajectory, count the number of point clouds in each grid, and calculate the point cloud density. Figure 3 Calculate the distribution of the point cloud density in the Y direction and find the variance of the data points on the curve.

[0058] In the above steps, the optimal variable speed scanning parameters can be obtained by pressing Figure 4 First, the known parameters in step S401 need to be determined, and then the boundary conditions in step S402 are input, mainly including: angular velocity variation range [ω 1， ω 2 ], the range of speed change delay time Δt [t 1 , t 2 ], initial speed change angle position θ 0 ; In S403, a set of boundary conditions is determined, and the corresponding point cloud density distribution variance is calculated respectively by steps S404 to S406, and the obtained variance value is compared with the point cloud distribution density variance value obtained by uniform speed scanning. If it is greater than the point cloud distribution density variance obtained by uniform speed scanning, the set of calculation conditions is discarded, and a new set of boundary conditions is re-determined for cyclic calculation. If the obtained variance value is less than the point cloud density distribution variance obtained by uniform speed scanning, the step S407 is entered, the obtained variance and the set of input boundary conditions are saved, and new boundary conditions are determined again for cyclic calculation until all boundary conditions are calculated, all variance values ​​that meet the conditions are compared, and the input boundary condition corresponding to the obtained minimum variance value is selected as the optimal solution, that is, the optimal variable speed scanning parameter.

[0059] In the above steps, the initial speed change angle position can be determined according to the calculation rules. Figure 5 For the same set of boundary conditions, different initial speed change angle positions θ 0 The calculated point cloud density distribution diagram shows that the size of the point cloud density at the middle position is related to the initial speed change angle position θ 0 Regardless, the distribution of the lowest point density at the edge position increases with θ 0 The increase of gradually moves to both sides, and the value of the lowest point of the point cloud density gradually increases. In this implementation case, θ 0 =60° is used as the initial speed change angle position. Under this condition, the lowest point at the edge is close to the lowest point at the center, which meets the purpose of uniform distribution of point cloud.

[0060] Figure 6This is a schematic diagram of the variance distribution calculated under multiple sets of boundary conditions. It can be seen from the figure that there are convergent boundary conditions in a certain area within the entire range, that is, there is a set of optimal scanning parameters.

[0061] The above steps can calculate a set of ideal variable speed scanning parameters. Figure 7 A set of point cloud density distribution diagrams are calculated. It can be seen from the figure that the segmented variable speed scanning method can reduce the point cloud density at the edge and increase the point cloud density in the middle. On the one hand, the uniformity of the point cloud distribution is improved, and on the other hand, the density of the overall point cloud is also improved, thereby improving the resolution of the imaging. Practitioners in this field can clearly recognize that the optimal scanning parameters obtained under ideal conditions may not completely match the actual scanning device, and the range of parameter variables needs to be further constrained to obtain compromise scanning parameters.

[0062] In order to realize the scanning method shown in this embodiment, a prism variable speed scanning system is also proposed.

[0063] Figure 8 The figure shows the structure of the prism variable speed scanning system of this embodiment. S801 is a power supply module, which is used to supply power to each module of the system. S802 is a scanning prism control module, which is used to program and implement the variable speed scanning control mode. It contains a storage module and can communicate with the S805 host computer module through an open register address to modify the scanning parameters. S804 is a scanning prism drive module, which is used to drive the scanning prism to rotate. S803 is a scanning prism module, which is used to deflect the outgoing light to realize circular trajectory scanning. This module has a built-in encoder to realize the output of angle values.

[0064] Optionally, the scanning prism module may be other mechanical scanning components, such as a scanning rotating mirror, a scanning galvanometer, etc.

[0065] Optionally, the scanning prism driving module may be a driving electronic device for other scanning components.

[0066] Optionally, the scanning prism control module can be any electronic device with programming, storage, and communication functions.

[0067] Optionally, the host computer module can be any computer read-write device with communication function.

[0068] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A scanning method for improving the uniformity of laser radar point cloud distribution, characterized in that: The uniform speed scanning mode within one cycle of the prism is converted into a multi-segment variable speed scanning mode, specifically including the following steps: S1, determine the number of speed changes of segmented scanning within one cycle; S2, determine the initial scanning speed of the prism; S3, determining the scanning speed of the prism after the speed change; S4, determining the speed change delay time; S5, determining the angular position of each speed change within a cycle; S6, determining the acceleration value according to the initial scanning speed, the scanning speed after the speed change and the speed change delay time; S7, determine the height of the equipment, laser repetition rate, lateral flight speed, beam deflection angle, and determine the point cloud density distribution curve after speed change; S8, taking the variance value of the point cloud density distribution curve as a criterion, adjusting the magnitude of the acceleration value to optimize the uniformity of the point cloud density distribution; S9. Calculate the optimal variable speed scanning conditions based on the actual achievable acceleration level and the initial state of the equipment.

2. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 1, characterized in that: The point cloud density distribution curve is a curve obtained by performing point counts in a certain size grid in the scanning trajectory diagram, obtaining the number of point clouds per unit area, and then performing point cloud density statistics in a certain direction.

3. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 2, characterized in that: The variance value of the point cloud density distribution curve is obtained by solving the variance of the data on the point cloud density distribution curve.

4. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 3, characterized in that: Through the equipment height h, laser repetition frequency f laser , aircraft lateral flight speed v x , beam deflection angle α and initial angular velocity w0, angle values ​​θ at different times, t is time, calculated according to the following formula: Calculate the scanning trajectory of the prism within the time period t; By intercepting the scanning trajectory range of the uniformly intersecting part, obtaining the number of point clouds per unit area, and calculating the point cloud density, the distribution diagram of the point cloud density can be obtained, and the variance of the data points on the curve can be obtained.

5. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 1, characterized in that: The optimal variable speed scanning condition calculation method is as follows: S901, according to the angular velocity variation range [ω 1， ω2], the range of the speed change delay time Δt [t1, t2], the initial speed change angle position θ0, determine a set of boundary conditions, and calculate the corresponding point cloud density distribution variance; S902, compare the variance value obtained in S901 with the variance value of the point cloud distribution density obtained by uniform scanning. If it is greater than the variance of the point cloud distribution density obtained by uniform scanning, the set of boundary conditions is discarded, and a new set of boundary conditions is determined and returned to S901 for recalculation; if the obtained variance value is less than the variance of the point cloud density distribution obtained by uniform scanning, proceed to S903; S903, saving the variance that meets the conditions and the boundary conditions of the group of inputs, and determining new boundary conditions again to return to S901 for loop calculation until all boundary conditions are calculated; S904, compare all the variance values ​​that meet the conditions, and select the input boundary condition corresponding to the minimum variance value as the optimal solution, that is, the optimal variable speed scanning condition.

6. A control system for improving the uniformity of laser radar point cloud distribution, characterized in that: It includes a scanning prism module, a scanning prism driving module, and a scanning control device; the scanning prism module is used to realize deflection scanning of the laser beam to form a continuous scanning trajectory; the scanning prism driving module is used to drive the prism to rotate in a predetermined manner; the scanning control device is used to download the optimal variable speed scanning conditions to the scanning control device in the form of a programming language to realize variable speed control of the scanning prism.

7. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 6, characterized in that: The scanning prism module can be a scanning rotating mirror or a scanning galvanometer.

8. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 6, characterized in that: The scanning prism drive module is the driving electronic device for other scanning components.

9. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 6, characterized in that: The scanning prism control module is any electronic device with programming, storage and communication functions.

10. The scanning method for improving the uniformity of laser radar point cloud distribution according to claim 6, characterized in that: The scanning prism module and the scanning control device both communicate with the host computer module, and the host computer module is any computer reading and writing device with communication function.

Images