Machine vision aided method for the autonomic diagnostic alignments
-
摘要: 激光惯性约束聚变实验需要使用数十台套诊断设备从不同方位对瞬态微尺度物理过程进行诊断表征。大部分诊断设备通常需要进入巨型靶室真空环境内，在厘米到米级的不同工作距离上，对聚变靶上面毫米到数十微米的靶标进行瞄准，大部分诊断设备的瞄准精度需要达到50 μm水平。双目瞄准方法是在真空环境下实现远距离高精度瞄准的一种重要方法，但目前主要依赖人工判读图像识别靶标和手动操作诊断搭载平台运动实现对靶瞄准，特别是靶室内照明条件或诊断设备瞄准视线存在夹角等条件会严重影响靶标识别效果，对诊断设备瞄准精度造成较大影响。发展了一种基于机器视觉的诊断自动瞄准方法，采用Mask R-CNN算法并以大量模拟瞄准图进行靶标识别训练，有效解决了靶标自动判读问题，对靶标识别误差控制在8个像素点以内；同时基于实验室瞄准测试平台开展了靶标像素偏差与瞄准坐标偏离关系的离线标定，开展了算法引导下的瞄准精度测试，根据测试结果预估指向瞄准精度优于30 μm、径向瞄准精度优于50 μm，对实现诊断设备的高精度自动瞄准有一定的基础参考价值。
-
关键词:
- 激光惯性约束聚变 /
- 诊断自动瞄准 /
- Mask R-CNN算法 /
- 靶标识别 /
- 瞄准反馈控制
Abstract: In the laser driven inertial confinement fusion experiments, several dozens of diagnostic instruments are needed, located in different sightlines of view. Most of the instruments are required to work inside the giant vacuum target chamber, in distances of centimeters to meters. They are always moved from several meters away, by the general diagnostic instrument manipulator and point the micro target in a precision of about 50 μm. The binocular pointing method is one of the continual alignment methods, which are able to work in the vacuum and in a faraway distance. However, at present this method needs target recognizing by eyes and pointing in a manual operation. Under some situations, such as under low illumination or pointing sightlines with an angle, the target marker may not be accurately recognized, as a result, the precision of the diagnostic alignment degenerates rapidly. In this work, a machine vision aided autonomic alignment method is proposed. The Mask R-CNN algorithm is used to recognize the target marker. Many simulated visual target graphs are created to train the algorithm. The accuracy of the target marker recognizing increases obviously. The error in the test is less than 8 pixels. Furthermore, the relation between the pixel shift in the visual graph and the pointing shift in the coordinate is calibrated in the laboratory. According to the machine vision aided alignment method, the tested alignment precision in pointing direction is estimated to be less than 30 μm, and in axial direction less than 50 μm. With the feedback of the alignment shift, the diagnostic alignment is able to be in an autonomic operation. -
图 2 双目视觉瞄准方法示意图及视觉图形与坐标偏移的坐标系
Figure 2. Sketch map of the binocular pointing alignment and the coordinate system between the vision graph and the shift in the coordinate
图 4 瞄准视线夹角为105°条件下的靶标识别图
Figure 4. Recognition of the center of the laser entrance hole with an angle of 105° between the sightline and the hohlraum
-
[1] 郑志坚, 丁永坤, 丁耀南, 等. 激光-惯性约束聚变综合诊断系统[J]. 强激光与粒子束, 2003, 15(11):1073-1078Zheng Zhijian, Ding Yongkun, Ding Yaonan, et al. Recent progress and application of diagnostic technique in laser fusion[J]. High Power Laser and Particle Beams, 2003, 15(11): 1073-1078 [2] Jiang Shaoen, Wang Feng, Ding Yongkun, et al. Experimental progress of inertial confinement fusion based at the ShenGuang-III laser facility in China[J]. Nuclear Fusion, 2019, 59: 032006. doi: 10.1088/1741-4326/aabdb6 [3] Kilkenny J D, Bell P M, Bradley D K, et al. The National Ignition Facility diagnostic set at the completion of the National Ignition Campaign, September 2012[J]. Fusion Science and Technology, 2016, 69: 420-451. doi: 10.13182/FST15-173 [4] Kilkenny J D, Hsing W W, Batha S H, et al. National Diagnostic Working Group (NDWG) for inertial confinement fusion (ICF)/high-energy density (HED) science: the whole exceeds the sum of its parts[J]. Review of Scientific Instruments, 2023, 94: 081101. doi: 10.1063/5.0128650 [5] 汪威, 史铁林, 来五星, 等. ICF诊断设备搭载平台结构设计与误差分析[J]. 强激光与粒子束, 2010, 22(6):1259-1264 doi: 10.3788/HPLPB20102206.1259Wang Wei, Shi Tielin, Lai Wuxing, et al. Structure design and error analysis for ICF diagnostic instrument manipulator[J]. High Power Laser and Particle Beams, 2010, 22(6): 1259-1264 doi: 10.3788/HPLPB20102206.1259 [6] Manuel A M, McCarville T J, Seppala L G, et al. Opposed port alignment system (OPAS): a commercial astronomical telescope modified for viewing the interior of the NIF target chamber[C]//Proceedings of SPIE 8491, Optical System Alignment, Tolerancing, and Verification VI. 2012: 84910E. [7] Ayers S. Target & diagnostic alignment basics[R]. Livermore: Lawrence Livermore National Security, 2016. [8] Brunton G, Abed Y, Fedorov M, et al. Status of the National Ignition Facility (NIF) integrated computer control and information systems[C]//Proceedings of the 16th International Conference on Accelerator and Large Experimental Control Systems. 2017: 14-18. [9] Shingleton N, Kalantar D, Wood R, et al. Alignment of an X-ray imager line of sight in the National Ignition Facility (NIF) target chamber using a diagnostic instrument manipulator (DIM) and opposed port alignment system (OPAS)[C]//Proceedings of SPIE8505, Target Diagnostics Physics and Engineering for Inertial Confinement Fusion. 2012: 72-81. [10] Le Breton J P, Alozy E, Boutin J Y. Laser integration line target diagnostics first results[J]. Review of Scientific Instrument, 2006, 77: 10F530. doi: 10.1063/1.2349746 [11] Cavailler C. Inertial fusion with the LMJ[J]. Plasma Physics and Controlled Fusion, 2005, 47: B389. doi: 10.1088/0741-3335/47/12B/S28 [12] 杨正华, 刘慎业, 曹柱荣, 等. 神光Ⅲ主机通用诊断搭载平台的研制[J]. 强激光与粒子束, 2011, 23(1):115-120 doi: 10.3788/HPLPB20112301.0115Yang Zhenghua, Liu Shenye, Cao Zhurong, et al. Development of Shenguang Ⅲ facility general diagnostic instrument manipulator[J]. High Power Laser and Particle Beams, 2011, 23(1): 115-120 doi: 10.3788/HPLPB20112301.0115 [13] 齐文博, 张琦, 何俊华, 等. 激光核聚变诊断设备双光束交汇瞄准技术研究[J]. 应用光学, 2009, 30(6):1032-1035Qi Wenbo, Zhang Qi, He Junhua, et al. Alignment technique for two-beam focus of laser-fusion diagnostic instrument[J]. Journal of Applied Optics, 2009, 30(6): 1032-1035 [14] 李颖洁, 陈铭, 陈黎, 等. 激光聚变实验中诊断设备反馈控制瞄准技术[J]. 控制理论与应用, 2020, 37(10):2147-2154Li Yingjie, Chen Ming, Chen Li, et al. A feedback control aiming technique of the diagnostic equipments in the inertial confinement fusion experiments[J]. Control Theory & Applications, 2020, 37(10): 2147-2154 [15] 夏立琼, 陈伯伦, 王鹏, 等. 深度神经网络在自动瞄靶技术中的应用研究[J]. 核聚变与等离子体物理, 2022, 42(S1):125-130Xia Liqiong, Chen Bolun, Wang Peng, et al. Research on the application of deep neural network in automatic targeting technology[J]. Nuclear Fusion and Plasma Physics, 2022, 42(S1): 125-130 [16] 卢卫涛, 闫亚东, 张敏, 等. 公共诊断搭载平台双CCD交汇瞄准系统[J]. 应用光学, 2011, 32(4):630-635Lu Weitao, Yan Yadong, Zhang Min, et al. Double CCD intersection positioning system for public diagnosis carrying platform[J]. Journal of Applied Optics, 2011, 32(4): 630-635 [17] Ren Shaoqing, He Kaiming, Girshick R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. doi: 10.1109/TPAMI.2016.2577031 [18] He Kaiming, Gkioxari G, Dollár P, Girshick R. Mask R-CNN[C]//Proceedings of2017 IEEE International Conference on Computer Vision (ICCV). 2017: 2980-2988. [19] Girshick R. Fast R-CNN[C]//Proceedings of 2015IEEE International Conference on Computer Vision (ICCV). 2015: 1440-1448. -
下载:
点击查看大图
图(9)
计量
- 文章访问数: 193
- HTML全文浏览量: 67
- PDF下载量: 37
- 被引次数: 0
