LAMOST Fiber Position Detection And Observation Planning Issues Related To The Study

Large Sky Area Multi-object Optical Spectroscopic Telescope - LAMOST, is a reflective Schmidt optical telescope with large diameter and wide field of view designed by Chinese astronomers. It is one of the ongoing great national scientific projects in China. According to the holistic design of LAMOST, about 4,000 fibers are needed to be positioned precisely on a focal plate of 1.75 m in diameter.Professor Xing Xiaozheng, from Precision Machinery and Precision Instrumentation of USTC, brought forward a novel parallel controllable dual rotational fiber positioning method based on Regional Division. This program has lots of advantages in technology: the system can position quickly; the system can compensate error real-time caused by temperature or atmospheric refraction; the fibers can point to stars directly; no blind area; 4,000 positioning units use same components, so they have high processing reliability. To achieve this system lies in two points: positioning with high precision and observation with high efficiency. The former ensures that LAMOST can observe stars and the latter ensures that LAMOST can observe stars with high efficiency.To verify the precision of fiber positioning, The geometrical coordinates of fibers need to be measured. The traditional position detection technology, such as three-dimensional Measuring Machine, Theodolite and three-dimensional Laser Tracker, is not suitable for un-touched and fast detection of so much objects. According to the characteristic of wide field of view and fibers distributing discretely for LAMOST fibers detection, photogrammetric measurement based on Regional Division is brought out. Aiming at the key techniques when building up the detection system, in-depth research have been done on theories, methods and experiments in the aspects of detection stability of CCD detection system, model establishment of CCD detection system, the local and global calibration method based on CCD measurement system, coverage method for field of view and space coordinate algorithm. The research has the significance of basic research for detection stability of CCD detection system and provides a feasible means for the realization of fibers position detection of LAMOST.In the research of Stars Observation Plan, stars allocation is mapping to the network and the method of stars allocation using network algorithm is brought up. Detailed research have been done in the aspect of the realization of this algorithm and LAMOST sky area coverage. The research will make full preparations for the realization of LAMOST Observation Plan.This dissertation mainly completes the following work:1. Static and dynamic stability of CCD detection system is evaluated through experiments under different conditions. In addition, which factor influence detection stability and how to decrease or eliminate measurement error are discussed. Meanwhile, the system detection error is analyzed.2. Nonlinear imaging model which is more consistent with the actual situation is established from Pinhole model, and accuracy analysis is done based on this model. Meanwhile, the relationship between accuracy and structure parameters is discussed.3. Discussion is made on the camera calibration. According to the characteristic of internal parameters fixed and camera spatial posture changing constantly, internal parameters are obtained through a traditional local calibration method while camera spatial posture through a global calibration method. This will ensure the consistency between calibration and measurement status.4. Detailed discussion is made about corresponding image points matching and measurement error approach during the process of space coordinates computation. A novel matching method based on the comparison of space coordinates is proposed. And space coordinates of objects are obtained through light rays adjustment method, this will be helpful to eliminate measurement error.5. Based on the above researched theories and methods, position measurement system with science class CCD camera is set up and measurement experiments are conducted. The results indicate that the calibration precision of this system is about 15 micron and the fibers measurement precision in the area of 600mm×600mm is less than 30 micron.6. On the basis of existing LAMOST Observation Plan, the NetFlow Algorithm is applied in it. The realization of this algorithm is described and it increases observation efficiency and fiber used ratio. Finally, LAMOST sky area coverage is discussed based on Min-Cost Flow Algorithm. The concept of dynamic measurement stability is proposed and the factors which influences the system measurement stability is analyzed through experiments.This dissertation mainly inquires into the important problem of fiber po- sitioning system for LAMOST: position measurement with high precision for discrete objects and stars observation plan. The concept of dynamic measurement stability is proposed and the factors which influences the system measurement stability is analyzed through experiments. Photogrammetric measurement based on Regional Division is brought out and CCD coordinate measurement system is set up. The stars allocation method through Network Algorithm is proposed and realized. The research will definitely have the practical value and theoretical basis for the final LAMOST fiber positioning system development.