Macroscopic Fourier ptychography imaging technology reconstructs high-resolution complex images by stitching and integrating low-resolution images in the frequency domain. However， positional misalignment， which decreases the quality of reconstructed images， commonly occurs when the camera is moved to capture low-resolution images. Therefore， a correction method based on particle swarm optimization based on a point-by-point calibration strategy is proposed. First， low-frequency images are calibrated in the frequency domain， and the frequency spectrum is updated. After all low-resolution images are calibrated， the exact camera positions are determined， and high-resolution images are reconstructed using a phase recovery algorithm. In real-world settings， the reconstructed image resolutions of the traditional Fourier ptychography imaging algorithm and proposed algorithm after calibration are 4.00 lp/mm and 5.04 lp/mm， respectively. The reconstructed image quality of the proposed algorithm is significantly improved， and the correction effect is better than that of similar existing algorithms. Furthermore， the execution time is reduced by more than 10.9% when compared with that of similar calibration algorithms. Our algorithm effectively releases the severe requirements of macroscopic Fourier ptychography imaging technology for camera position accuracy， thereby improving the quality of reconstructed images and reducing the time expense.

Considering the development trends and demands of large-aperture prime focus optical telescopes， design and alignment methods for a large-aperture prime focus optical system were analyzed. A prime focus optical system with a large aperture was presented. This system was composed of an aspherical primary mirror and six spherical lenses with smaller apertures. The corrective lens could be used to enlarge the field of view， correct image aberrations， and ensure the image quality of the system. This system has a focal length of 1 300 mm， working waveband of 0.4-0.8 μm， clear aperture of 1 000 mm， and a 2.7°×2.7° field of view. In addition， this system was compensated by moving the detection image plane in the range of -40-+50 ℃. The modulation transfer function （MTF） of the system was above 0.45 at the Nyquist frequency， and the RMS speckle radius was less than 10 μm in the full field of view. A general method including the adjustment of the primary mirror， corrective lens， and system was proposed for this system， the results of which indicated that the star point 80% energy concentration degree in the full field of view of the system was in a 3×3 pixels region of the target area.

In order to meet the high-resolution imaging requirements of large ground-based telescopes， adaptive optics technology needs to be used to correct the wavefront error caused by atmospheric turbulence. The piezoelectric deformable mirror is the key component of an adaptive optics system. Aiming at the high-voltage and high-speed driving of the thousand-element piezoelectric deformable mirror in a large ground-based high-resolution imaging telescope， a high-performance piezoelectric ceramic high-voltage amplifier was designed. The high-voltage amplifier adopts a two-stage amplification structure based on high-speed low-voltage operational amplifiers and discrete power devices to realize high-precision and high-power driving of piezoelectric ceramics. Compared with a driving system composed of integrated operational amplifier devices， the high-voltage amplifier has the advantages of a stronger driving capability and lower cost， and it is more suitable for piezoelectric deformable mirror systems with many driving channels. The experimental results show that the high-voltage amplifier designed in this paper can achieve a 120 V output， the -3 dB bandwidth can reach 5 000 Hz when driving a 0.33 μF capacitive load， and the output response rise time is less than or equal to 100 μs. The high-voltage amplifier can meet the application requirements of a thousand-element piezoelectric deformable mirror.

In order to meet the wavefront processing scale， speed， and control bandwidth requirements of the thousand-element adaptive optics system of the 4-m-level ground-based large-aperture optical telescope， a GPU-based high-speed wavefront processor for the adaptive optics system is examined. The wavefront processing method of the adaptive optics system is introduced， and the realization and optimization methods of the GPU-based wavefront processing architecture are discussed. A testbed simulation experiment is performed using a turbulence simulator， in order to analyze the dynamic performance of the 961-unit adaptive optics system. The experimental results show that when the sampling frequency of the 961-unit adaptive optics system is 1 500 Hz， the 0 dB residual error rejection bandwidth of the system can reach 100 Hz. This rejection bandwidth value meets the application requirements of the 4-m optical telescope adaptive optics system.

With continuous increases in the apertures of ground-based telescopes， the size and weight of the lens barrel structure is becoming greater and greater， and the spatial position change of the primary and secondary mirrors caused by the finite rigidity of the structure and the change of the gravity field is becoming increasingly obvious. The accurate measurement and correction of the deflection of the primary and secondary mirrors of a large-aperture telescope is a necessary prerequisite to ensure its imaging quality and pointing accuracy. Based on the structural characteristics of a large-aperture telescope， the factors affecting the position and attitude of the primary and secondary mirrors are analyzed， and an optical measurement method based on the internal focusing collimator is proposed. A telescope with a distance of 5.5 m between the primary and secondary mirrors is measured as an example. The maximum radial offset in the horizontal direction is 196 μm， the maximum radial offset in the vertical direction is 16 μm； the maximum angular deflection in the horizontal direction is 2.6″， and the maximum angular deflection in the vertical direction is 12.8″. Finally， the measurement uncertainty involved in this method is analyzed. The combined standard uncertainty of measuring radial offset is 9 μm， and the combined standard uncertainty of measuring angular deflection is 0.65″， meeting the alignment accuracy of large-aperpure telescopes.

As the diameter of ground-based telescopes increases， the structural size of the telescope also increases accordingly， and the effect of the telescope's optical axis pointing caused by gravity becomes increasingly pronounced. The measurement and calibration of optical axis changes for ground-based large-aperture telescopes is a necessary prerequisite to obtaining high-precision orbital target data. To realize the estimation， measurement， and active correction of the telescope pointing error， the optical axis changing rules of the telescope system are analyzed from the structure of the telescope's opt-mechanical system， and an optical axis change measurement and calibration method of the is proposed. The experimental results show that the maximum relative change of the optical axis is 126″ when the telescope is at low elevation. The main factor causing the pointing change is the effect of gravity on the telescope truss structure. Finally， a calibration model based on the secondary mirror adjustment structure and the coma-free theory of the Cassegrain system is proposed to improve the tracking performance of the 4-meter aperture telescope. The maximum relative change in the optical axis of the telescope after calibration is within 3″. According to the test results， the ground-based large telescope pointing change measurement and calibration method can be applied to the calibration and mounting process of the telescope.

The dome seeing is an important influencing factor for imaging quality of ground-based large-aperture telescopes. Seeing is mainly due to the uneven change in the refractive index of air. First， theoretical analysis shows that the main factors affecting the refractive index of air are the temperature， velocity， and pressure of the air. Then， with the help of CFD fluid simulation software， the temperature field， velocity field， and pressure field changes of the air near the telescope dome are obtained， and the change of the refractive index of the air around the dome and telescope is calculated. Thus， the effect of dome seeing produced by domes of different structural forms on telescopic imaging can be evaluated. The simulation analysis shows that the lifting dome is more conducive to the rapid change and stability of the temperature， speed， and pressure of the air around the telescope to reduce the impact of dome seeing. From the perspective of dome seeing， this paper provides a guide for the scheme selection and design of a dome at the 4-meter level and larger telescopes.

To satisfy the requirements of high stiffness and low inertia of the center section of a large aperture telescope， a lightweight center section is designed. Further， because the traditional method cannot detect the geometric tolerance of the center section， an optical detection method based on an autocollimator and laser tracker is proposed. According to the requirements of the overall structure of the telescope， the size range of the center section is determined， and the transmission path of the force is analyzed. The topology optimization analysis and structure design are carried out using the variable density method to minimize strain energy. Based on the angle measurement principle of the autocollimator， the reflecting mirror is placed on the face of the center section hole to measure the parallelism error of the face on both sides. Based on the position measurement principle of the laser tracker， a high-precision turntable is used to find the central point of the center section hole. Then， the central point coordinates of holes on both sides are measured， and the coaxiality error is measured. Compared with the center section designed by traditional means， the mass of the center section designed under the guidance of topology optimization is reduced by 21.5%， and the static stiffness is increased by 14.3%. The parallelism error of the faces of the holes on both sides of the center section detected by the optical method is 0.016 mm， and the concentric error of the holes on both sides is 0.03 mm. The center section designed by topology optimization has obvious advantages in improving stiffness and reducing inertia. The detection of geometric tolerance of the center section can be realized by using the proposed optical detection method， which solves the detection problems in processing.

To develop a reasonable transition process for physical signals with speed and acceleration limits， we propose a variable parameter linear tracking differentiator with speed and acceleration saturations. The parameter of the tracking differentiator is designed as a decreasing function of the tracking error， which avoids large acceleration shock in the initial stage when the tracking error is relatively large. Using the same set of parameters， the designed tracking differentiator can give reasonable transition signals that match the feedback capabilities of physical systems for a wide range of step signals. Simulation results show that the proposed tracking differentiator can produce nearly the same tracking effect as that of the classical nonlinear tracking differentiator and with reduced computational complexity. We applied the designed tracking differentiator to the rapid positioning of an altazimuth large area telescope. Compared with the commonly used position segmentation method， the response time was reduced by 15%–37%， thereby improving the search efficiency of the telescope. The proposed variable parameter tracking differentiator can not only play a transitional role but can also maximize the feedback capabilities of physical systems. The parameter setting of the proposed method is simple as well as easy to debug and implement in engineering practice.

To meet the requirements of fast search and high precision tracking at ultra-low speed for a 2.5 m wide field telescope， the design method of the three closed-loop control system of position， velocity， and current of telescope main axes AC servo is introduced. Firstly， a sinusoidal signal is injected into the set current closed-loop reference input end， and the speed open-loop frequency characteristic curve of the main axes servo system for the telescope is measured using a frequency sweep. Then， the structural filter is designed according to the resonant frequency to suppress the mechanical resonance so as to achieve a higher closed-loop bandwidth. Finally， according to the identified control model， the speed loop linear active disturbance rejection （LADRC） controller and position loop proportional controller are designed. The velocity loop estimates the disturbance and compensates for its effects based on the linear extended state observer， which achieves higher low-speed tracking accuracy. To solve the fast search problem， the position loop arranges the transition process according to the maximum allowable velocity and acceleration of the device and based on the discrete maximum velocity tracking differentiator. The experimental results show that， compared with the PI control， LADRC reduces the fast step search time of 1.24° FOV for a 2.5 m wide field telescope from 1.6 s to 1.0 s. The equivalent sinusoidal guidance error RMS value decreases from 1.08″ to 0.60″ when tracking the sine curve with a velocity of 2 （°）/s and acceleration of 1 （°）/s2， The steady-state error RMS of position decreases from 0.015 8″ to 0.010 6" when tracking the slope of position at 0.000 1 （°）/s. The experimental results indicate that LADRC can meet the requirements of high efficiency， fast search， and precise tracking at low speed for large field angle telescopes.

In order to achieve fast and high-precision position switching control of a ground-based telescope， without overshoot and while improving the positioning performance， a trajectory-planning strategy based on an approximate optimal-command-shaping algorithm is proposed. The approximate optimal-command-shaping algorithm is utilized to design the command shaper， which is located before the position controller. The designed command shaper is able to perform trajectory planning， using the reference command， and the speed- and acceleration-limit information of the system. As a result of the algorithm-designed command shaper， the system can be guided to its destination quickly and smoothly. Compared to the traditional trapezoidal-command-shaping algorithm， the proposed approximate optimal-command-shaping algorithm overcomes the problem of chattering， thus achieving better position-control performance. The experimental results demonstrate that， compared to the positioning strategy without trajectory planning， the system using the proposed positioning strategy has a reduced adjustment time. In particular， the adjustment time for the system to enter the 2″ error band in response to a 2.5° and 30° step reference is reduced by 1.14 s and 1.57 s， respectively. In addition， the overshoot of the position response is significantly reduced. The simulation and experimental results show that the positioning performance of the telescope can be improved effectively by the trajectory-planning positioning strategy， based on the proposed approximate optimal-command-shaping method.

To improve the dynamic target tracking performance of optical systems based on piezoelectric fast mirrors， the effect of hysteresis nonlinearity and the corresponding compensation methods are investigated in this study. First， based on the Prandtl–Ishlinskii （P-I） model， a rate dependent hysteresis model in series with the P-I model and the auto regressive empirical model is established. Then， a rate dependent model compound control compensation algorithm based on P-I is proposed. Finally， an experimental platform using a piezoelectric fast mirror is built to identify the hysteresis effect of piezoelectric ceramics， and a feedforward controller is designed based on its inverse model. Simultaneously， the optimization effect of the hysteresis compensation method on the dynamic target tracking system is verified. The experimental results show that the rate dependent model based on P-I is suitable for moving target tracking systems. When the compound control compensation algorithm is applied， the error suppression bandwidth of the dynamic target tracking system is increased by 26 Hz. Thus， the error suppression capability of the dynamic target tracking system is effectively improved.

A secondary mirror mount in a large telescope introduces different attitude errors at different elevations， reducing the accuracy of the imaging device. In particular， large SiC telescopes suffer from this problem. Without correction of the secondary mirror attitude, a large error of imaging point position was introduced that result in an overload of the precision tracking system. By considering the optical design parameters of the telescope， a secondary mirror correction method based on the coma-free point （CFP） and center of curvature was developed. Stewart platform is used to correct the attitude of secondary mirror based on the telescope elevation. The alignment error before the precision tracking system is reduced from 12.85″ to 1.80″，as a result of the attitude correction of the secondary mirror. The correction method is simple and effective， allowing for coarse alignment of the precision tracking system and preserving the accuracy of the imaging system.