-
棱镜出射光束方向矢量和目标处的空间坐标采用三维矢量光学方法求解,从棱镜出射光束角度和目标空间坐标两个方面去量化分析畸变影响。阵列光束通过菲涅耳棱镜光束扫描过程,如图3所示,以DOE位置为光束分束的起点M0,子光束方向向量S1,i-j与入射平面Σ1交点为M1,i-j,经棱镜表面折射,在棱镜内部的方向向量为S2,i-j,与出射平面Σ2交点为M2,i-j,经棱镜出射斜面折射后的出射方向为S3,i-j,经过空间传输后,光束与目标平面Σ3的交点为M3,i-j,其中,d0为M0到平面Σ1垂直距离,d1为棱镜中心位置与平面Σ1的垂直距离,与d2为目标平面Σ3与入射平面Σ1之间的距离。
基于三维矢量光学方法的阵列光束通过菲涅耳棱镜光束扫描三维成像数值仿真流程,如图4所示。三维矢量光线传输3个要素:起点位置Mi,光线方向向量Si,入射平面方程Σi;然后求解光线与平面Σi交点Mi+1,折射光束Si+1,结合下一个平面方程Σi+1,求解Mi+2和Si+2;串行求解到最后一个表面。
图 4 基于三维矢量光学方法光束传输流程图
Figure 4. Flow chart of beam transmission based on three-dimensional vector optics
入射光线起点Mi(xi, yi, zi)和方向向量Si(mi, ni, pi),则直线上任意一点M(x, y, z)满足直线参数(ti)方程:
$$ \frac{{{{x - }}{{{x}}_{{i}}}}}{{{{{m}}_{{i}}}}}{{ = }}\frac{{{{y - }}{{{y}}_{{i}}}}}{{{{{n}}_{{i}}}}}{{ = }}\frac{{{{ {\textit{z}} - }}{{{ {\textit{z}}}}_{{i}}}}}{{{{{p}}_{{i}}}}}{{ = }}{{{t}}_{{i}}} $$ (1) 光线入射平面Σi方程:
$$ {A_i}x + {B_i}y + {C_i}z + {D_i} = 0 $$ (2) 当直线与平面Σi相交,M(x,y,z)也满足Σi平面方程,联合公式(1)、(2)得到参数ti和Mi+1:
$$ {{{t}}_{i}}{\text{ = }}\frac{{}}{{}}\frac{{{A_i}{x_i} + {B_i}{y_i} + {C_i}{ {\textit{z}}_i} + {D_i}}}{{{A_i}{m_i} + {B_i}{n_i} + {C_i}{p_i}}} $$ (3) $$ {{{M}}_{{{i + 1}}}}{\text{ = [}}{m_i}{t_i} + {x_i},{n_i}{t_i} + {y_i},{p_i}{t_i} + { {\textit{z}}_i}] $$ (4) 基于Snell公式矢量形式[10],入射光束Si经过平面Σi后,折射光束方向Si+1:
$$ {{{S}}_{{{i + 1}}}}{\text{ = }}\mu {{{S}}_i}{\text{ + }}{{{N}}_{i}}\sqrt {1{{ - }}{\mu ^2}[1 - {{({{{N}}_{i}} \cdot {{{S}}_i})}^2}]} - \mu {{{N}}_{i}}({{{N}}_{i}} \cdot {{{S}}_i}) $$ (5) 式中:
$ \;\mu = {{{n_1}} \mathord{\left/ {\vphantom {{{n_1}} {{n_2}}}} \right. } {{n_2}}} $ ;平面Σi法向量$ {{{N}}_{i}} = [{A_{i}},{B_{i}},{C_{i}}] $ ;n1为入射光空间介质折射率;n2为折射光空间折射率;$ {{{N}}_{i}} \cdot {{{S}}_i} $ 为法向量Ni和入射光束向量Si的点积,即$ {{{N}}_{{i}}} \cdot {{{S}}_i} = {A_i}{m_i} + {B_i}{n_i} + {C_i}{p_i} $ 。其中,入射棱镜前表面Σ1:z=0,目标表面Σ3:z=d2,棱镜后表面Σ2的法向量为:
$$ N_{2}=[{\rm{sin}}( \phi ){\rm{cos}}( \varphi ),{\rm{sin}}( \phi ){\rm{sin}}( \varphi ),{\rm{cos}}( \phi )] $$ (6) 以及平面Σ2上棱镜中心坐标(0,0,d1),根据平面的点法式方程可得Σ2平面方程为:
$$ \begin{split} {\Sigma _{\text{2}}}{\text{: }}& \sin (\phi )\cos (\varphi )x + \sin (\phi )\sin (\varphi )y + \\& \cos (\phi )(z - {d_1}) = 0 \end{split}$$ (7) 式中:ϕ为棱镜顶角;φ为棱镜扫描角度。
依次类推,目标处激光脚点空间位置坐标为:
$$ {{{M}}_{{{i + 1}}}}{{ = }}{{{t}}_i} \cdot {S_i}{{ + }}{{{M}}_i} $$ (8) 根据阵列子光束入射方向向量S1,i−j(i, j=1,2,3
$ \cdots ) $ (含正射和斜射),分别求解棱镜出射光束方向矢量S3,i−j(i, j=1,2,3$\cdots ) $ 和目标处的空间坐标M3,i−j(i, j=1,2,3$\cdots ) $ 通过确定所有子光束的出射光束的角度和目标处的空间位置来综合表征阵列光束传输特征。
Analysis of point cloud accuracy and beam pointing of array beam through prism scanning
-
摘要: 在面阵扫描成像激光雷达中,阵列光束照明与棱镜扫描相结合实现了高能量利用率、高分辨率和宽探测视场,但阵列子光束倾斜入射棱镜,破坏了光束传输的旋转对称性,棱镜对子光束偏转能力存在差异,规则光束阵列产生了形状畸变,导致光束指向误差,影响点云位置精度。首先,将阵列光束与棱镜结合的圆锥扫描方式分解为多角度入射多波束并行扫描,通过所有子光束的传输特征来综合表征阵列光束传输特征;然后,采用三维矢量光学方法推导了阵列光束在棱镜中的传输过程,建立了子光束指向变化与棱镜扫描角度的关系;最后,通过对机载激光雷达棱镜扫描成像过程的数值仿真,建立了光束指向变化与点云数据质量的联系。仿真结果表明:阵列光束(3×3)棱镜扫描系统在航高0.5 km时,光束阵列畸变导致平面误差RMS约为5 cm,并随航高呈线性变化;斜率约为0.1 m/km,并随着阵列光束规模和子光束角间距增加点云平面精度随之下降。通过对棱镜扫描过程中光束阵列畸变规律掌握,为后续机载飞行试验数据的校正、阵列光束结合多棱镜扫描系统的设计提供了基础。Abstract:
Objective The prism scanning system is used to achieve optical imaging with both large field of view and high resolution by adjusting the beam direction or optic axis. It is widely used in optical reconnaissance, laser communication, lidar, etc. In airborne laser imaging lidar, the prism scanning system, as a transmission scanning structure, has high optical utilization, effectively reduces the volume of the system, and has the advantages of low power consumption, high precision and good stability. In array imaging lidar, high energy efficiency, high resolution, and broad field detection are all achieved by means of array beam illumination and Risley-prism scanning. However, when sub-beams are obliquely incident on the prism, the rotational symmetry of the traces of ray propagation is broken, the beam deflection of the sub-beams through the prism is different, and the regular beam array produces shape distortion, resulting in beam pointing error and affecting the position accuracy of the point cloud. Therefore, it is necessary to analyze the rules of beam array distortion to improve the accuracy of the point cloud. Methods The conical scanning mode that combines the array beam and prism is broken down into multi-beam parallel scanning with numerous incident angles, and the propagation characteristics of the array beam are thoroughly described by the propagation characteristics of all sub-beams (Fig.2). The three-dimensional vector optical approach is used to establish the laser transmission process of the array beam through a Risley-prism (Fig.3), and the relationship between the pointing variation of the sub-beam and the scanning angle of the prism is obtained (Fig.4). The association between beam pointing variability and point cloud data quality is demonstrated by the numerical simulation of imaging process with prism scanning by flight experiment of airborne lidar (Fig.6-7). Results and Discussions When the array beam is orthographically and obliquely incident into the prisms with different angles, the beam steering of the prism to each sub-beam is different at various scanning angles (Fig.5(a)). The spatial shape distortion analysis of the array beam is based on the spatial angle difference between the outgoing sub-beam and the central sub-beam. When the prism rotates one cycle, the spatial shape distortion of the array beam is shown (Fig.6(b)). The quality of point cloud data affected by the array beam distortion is evaluated by using plane fitting RMS value as the quantitative index of point cloud position accuracy (Fig.7). Simulation results of ground scanning imaging process of prism in airborne lidar indicate that the plane error RMS is approximately 5 cm at a navigation height of 0.5 km (Fig.8(a)), which varies linearly with navigation height, and slopes at a rate of around 0.1 m/km in prism scanning system with beam array (3×3) (Fig.8(b)) and the accuracy of point cloud plane decreases with the increase of array beam scale (Fig.9) and sub-beam angular separation (Fig.10). Conclusions The combination of array beam illumination and prism scanning improves the energy utilization, spatial resolution and detection field of view of airborne lidar system. However, the shape distortion of array beam leads to beam pointing error and affects the accuracy of point cloud position. The array beam incidence prism includes orthographic and oblique incidence. The oblique sub-beam destroys the rotational symmetry of the beam propagating in the prism, and the beam steering ability is different at different scanning positions. Furthermore, the larger the oblique angle is, the stronger the steering ability of the prism to the beam is. Given that the above two work together, the time-varying array beam is emitted when the regular array beam is incident. The relationship between beam pointing error and spatial position error is obtained by using the three-dimensional vector optics method. The increase of the incident angle of the sub-beam and the altitude will lead to the dispersion of the point cloud and the decrease of the data quality. The law of beam array distortion during prism scanning lays a foundation for the correction of subsequent airborne flight test data, especially for the improvement of position accuracy of medium and long-distance airborne lidar. In addition, it provides a reference for the design of array beam combined with multi-prism scanning system. -
Key words:
- airborne imaging lidar /
- array beam /
- prism scanning /
- pointing error /
- point cloud accuracy
-
图 5 (a)出射子光束与中心光束空间角度差值ΔAi-j和棱镜旋转角度φ关系,虚线框中展示了曲线细节部分(φ=45º);(b)在平面Σ2上,入射角γ与棱镜旋转角度φ关系;(c)出射阵列光束空间分布情况(φ=45º)
Figure 5. (a) Relationship between angle difference ΔAi-j between the emergent subbeam and center beam and scanning angle φ, the details of the curve in the dotted box; (b) Relationship between the incidence and angle γ scanning angle φ on the plane Σ2; (c) Spatial distribution of outgoing array beams at φ=45º
图 8 (a)平面拟合平面RMS值,畸变未校正数据(红线),畸变校正数据(蓝线),两者之间的差值(灰线);(b)航高与平面拟合RMS值之间关系
Figure 8. (a) Planar fit values RMS for planes used before (red line) and after (blue line) adjustment, difference between the two planar fit values (gray line); (b) Relationship between navigation height and RMS planar fit values for planes
-
[1] Hong Huajie, Zhou Yuan, Tao Zhong, et al. Application of risley prism on optical reconnaissance [J]. Journal of Applied Optics, 2014, 35(2): 179-186. (in Chinese) [2] Fan Dapeng, Zhou Yuan, Lu Yafei, et al. Overview of beam steering technology based on rotational double prisms [J]. Chinese Optics, 2013, 6(2): 136-150. (in Chinese) [3] Zhu Jingguo, Li Feng, Huang Qitai, et al. Design and realization of an airborne LiDAR dual-wedge scanning system [J]. Infrared and Laser Engineering, 2016, 45(5): 0502001. (in Chinese) doi: 10.3788/irla201645.0502001 [4] Brown E, Kim H, Carr D, et al. Seahawk lidar [C]//Proc of SPIE, 2019, 11005: 1100506. [5] Fuchs E, Mathur A. Utilizing circular scanning in the CZMIL system [C]//Proc of SPIE, 2010, 7695: 76950W. [6] Brazeal R G, Wilkinson B E, Hochmair H H. A rigorous observation model for the risley prism-based livox mid-40 lidar sensor [J]. Sensors, 2021, 21(14): 4722. doi: 10.3390/s21144722 [7] Clifton W, Steele B, Nelson G, et al. Medium altitude airborne Geiger-mode mapping LIDAR system[C]//Proc of SPIE, 2015, 9465: 946506. [8] Degnan J J. Scanning, multibeam, single photon lidars for rapid, large scale, high resolution, topographic and bathymetric mapping [J]. Remote Sensing, 2016, 8(11): 958. doi: 10.3390/rs8110958 [9] Lavigne V, Ricard B. Fast Risley prisms camera steering system: calibration and image distortions correction through the use of a three-dimensional refraction model [J]. Optical Engineering, 2007, 46(4): 043201. doi: 10.1117/1.2722754 [10] Huang F, Ren H, Shen Y, et al. Error analysis and optimization for Risley-prism imaging distortion correction [J]. Appl Opt, 2021, 60(9): 2574-2582. doi: 10.1364/AO.418007 [11] Li A, Zuo Q, Sun W, et al. Beam distortion of rotation double prisms with an arbitrary incident angle [J]. Appl Opt, 2016, 55(19): 5164-5171. doi: 10.1364/AO.55.005164 [12] Pan Z, Hartzell P, Glennie C. Calibration of an airborne single-photon lidar system with a wedge scanner [J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(8): 1418-1422. doi: 10.1109/LGRS.2017.2716100 [13] Wang Qianghui. Method on quality assurance of airborne original lidar point cloud data [J]. Journal of Geomatics, 2018, 43(5): 35-37. (in Chinese) doi: 10.14188/j.2095-6045.2017352