Automatic Alignment of Multiple Optical Components Using Genetic Algorithm

Markov random field based automatic image alignment for electron tomography

An industrial quality automatic resonator alignment system for a 20 kW CO 2 laser

We propose a self-learning method for automatic wafer alignment in the semiconductor manufacturing process. A feed forward neural network is trained by and used for wafer alignment. The network determines the movement of kinematic parts from the misalignment inspected by machine vision. However, it is time-consuming and inconvenient to obtain training data in this way. So, we built an automatic learning rule to gather the data and train the network. The network may determine wrong outputs and cause other misalignments at first, but the error can decrease as the training proceeds. The training sets consisted of a variation of misalignment data and the movement of an alignment stage. Five recent sets are used for training and others are dismissed or forgotten. This retrained network tried aligning, measured misalignment, and made new training sets. This sequence makes it possible to acquire alignment skill and automate the process. After learning, automatic alignment accomplished sub-pixel accuracy for several cases of misalignment. The result showed that the proposed method could be applied to the semiconductor manufacturing process. Its performance improved about 6% compared with conventional algorithms.

It is well known that the pointing stability of a laser beam is decisive in determining the quality and efficiency of laser cutting, welding and drilling. Due to the instability of pointing for a long beam path which is increasingly used in material processing, there is a need for an automatic laser beam alignment system which is designed to accurately position laser beams to the working point of workstations. Such a system would require a device to automatically identify where the beam is. An instrument for doing this is described. The study presented here combines a rotating disk, having one slit, with a high speed optical sensor and a convergent lens to form a sensory unit to detect the laser beam position.

We first studied the characteristics of alignment performances of two computer-aided alignment algorithms i.e. merit function regression (MFR) and differential wavefront sampling (DWS). The initial study shows i) that, utilizing damped least square algorithm, MFR offers accurate alignment estimation to the optical systems with non-linear wavefront sensitivity to changes in alignment parameters, but at the expense of neglecting the coupling effects among multiple optical components, and ii) that DWS can estimate the alignment state while taking the inter-element coupling effects into consideration, but at the expense of increased sensitivity to measurement error associated with experiment apparatus. Following the aforementioned study, we report a new improved alignment computation technique benefitted from modified MFR computation incorporating the concept of standard DWS method. The optical system used in this study is a three-mirror anastignmat (TMA) based optical design for the next generation geostationary ocean color instrument (GOCI-II). Using an aspheric primary mirror of 210 mm in diameter, the F/7.3 TMA design offers good imaging performance such as 80% in 4 um in GEE, MTF of 0.65 at 65.02 in Nyquist frequency. The optical system is designed to be packaged into a compact dimension of 0.25m × 0.55m × 1.050m. The trial simulation runs demonstrate that this integrated alignment method show much better alignment estimation accuracies than those of standard MFR and DWS methods, especially when in presence of measurement errors. The underlying concept, computational details and trial simulation results are presented together with implications to potential applications.

Large, high-power laser amplifiers use the imaging properties of multiple spatial filters. Spatial filtering has been shown to control instabilities; spatial filters preserve the transverse intensity profile of a high power beam as it propagates long distances through nonlinear elements as well. In this paper, image relaying is presented as a technique for aligning beams onto mm-sized target. On base of summarizing the preceding work on the near-field image relaying of multiple spatial filters, the far-field image relaying is suggested firstly. The application of near-field and far-field image relaying properties of multiple spatial filters in laser beams automatic alignment system is analyzed. A geometrical optics approach and an ABCD ray matrix theory are used throughout. A reasonable and optimized scheme for automatic aligning multipass beam paths is presented. It is demonstrated on the multipass amplifier experimental system of the SG-III prototype.

The automatic alignment system of the Virgo interferometer differs substantially from those used in similar experiments, since it uses a variant of the Anderson technique. This implies a completely different control topology with respect to other detectors, and the main feature is a strong coupling of different degrees of freedom in the optical signals. It also provides two extra output ports in which differential wave-front sensors can be placed, namely the light transmitted by the Fabry–Perot arm cavities. We report on the first experimental demonstration of this technique on a large scale recycled interferometer, and on the present status of the automatic alignment system.

Surface plasmon polaritons (SPPs) have shown high potential for various applications in various fields, ranging from physics, chemistry, and biology to integrated photonic circuits due to their the strong confinement of light to the metal surface. Exciting an SPP from a free-space photon in a controllable manner is an essential step toward more complex and integrated applications. Methods for coupling photons to SPPs are numerous, but in order to control the amplitude and phase of an SPP, most of these methods require bulky or multiple optical components or sensitive adjustments that are difficult to control. Here we present a novel approach for an independent control of the amplitude and phase of an SPP excited by a normally incident beam using a metasurface. The full control in amplitude and phase is achieved via the polarization state and polarization orientation angle of the electrical field of the incoming light. We experimentally demonstrate the functionality of such a metasurface consisting of periodic nanoantennas for the excitation of SPPs at a metal–dielectric interface. Our approach opens up new ways for coherently controllable integrated plasmonic circuits that can be used in conjunction with fast dynamic polarization modulation techniques.

Progress in die separation technology or “dicing” has improved device yield and productivity over the years. Today, the basic dicing process is one in which a rotating, abrasive-edged blade is positioned to cut in the “streets” or the “kerf” spaces between functional integrated circuits or other devices. The thin, 25 µm cutting blade rotates at very high speeds (30,000 to 60,000 rpm with blade-edge linear speeds of 83 to 175 m/sec). Dicing equipment has evolved over the years from simple, manual systems to fully automatic systems with automatic alignment, auto-handlers, dual spindles and quality monitoring systems. As device technology has evolved, so has the requirement for dicing saw manufacturers to meet the ever-increasing needs for higher precision, greater throughput and reduced damage to the topside and bottom-side edges of the dice. In this chapter, the reader is introduced to the dicing process and made aware of the key issues which must be considered when setting up a die separation process [1, 2].

Simultaneous alignment and figure testing have been simulated by computer for components of a Cooke triplet and a two-. Petzval telescope using end-to-end wavefront aberration measurement and computer model reverse optimization. Experimental confirmation is discussed. 2.

We present a system for automatic real time alignment of an acoustic music performance with a digital representation of its score, a problem which is usually defined score following. The alignment is based on an application of hidden Markov models. A model is automatically built from a music score, while decoding is used to compute the most probable location of the performance along the score model. The effectiveness of the proposed approach has been tested with a collection of recordings of orchestral music. Even if the typical application of a score following system is automatic accompaniment, in this paper we propose a set of novel applications that are targeted also to non musicians, for educational use.

In this paper, we propose a new optical alignment system using Hadamard transformation that is designed for aligning multiple channel optical components. A conventional optical alignment system assumes that the whole channels of the components are perfectly aligned with two detectors, which are placed at the end of the outermost channels of the components, indicate the maximum optical power. The proposed system is able to give us the optical power information of each channel with using one detector. The proposed optical alignment system can make up for the drawback of the conventional optical alignment system.

The quadcopter is an unmanned aerial vehicle that has been used in civil and military applications because of its relatively low cost, simple design, and ease of maintenance in addition to its unique flying characteristics and vast potential. Six Proportional-Integral-Derivative (PID) controllers are usually built for the position and attitude control of the quadcopter. However, tuning the controllers’ parameters poses a real challenge, as the system is highly coupled and contains two cascaded control loops while being underactuated (only four control inputs for the six degrees of freedom). In this study, a quadcopter model has been implemented by using MATLAB / SIMULINK where the model has been constructed to be integrated with the evolutionary genetic algorithm in addition to a third-party toolbox that is used to simulate fractional order calculus. The gains of the traditional and the fractional-order proportional-integral-derivative controllers have been tuned by using the genetic algorithm against different path shapes. The performance of the manually and the genetic algorithm tuned proportional-integral-derivative controllers have been compared with the genetic algorithm tuned fractional-order proportional-integral-derivative controllers. The genetic algorithm tuned fractional-order proportional-integral-derivative controllers have been superior and more robust in all cases with a lower fitness function value, much smother responses, and higher trajectory tracking ability.

Automatic alignment (matching) of two-dimensional gel electrophoresis images is of primary interest in the evolving field of proteomics. In the present study, feature-based matching techniques, in their classical and robust versions, are described, and an automatic method of fuzzy alignment (FA) is introduced. This method allows automatic matching of two gel images with different numbers of features with unknown correspondence. Performance of FA is tested on simulated and real data sets.

Genetic algorithm (GA) has been extensively used in recent decades to solve many optimization problems in various fields of science and engineering. In most cases, the number of iterations is the only criterion which is used to stop the GA. In practice, this criterion will lead to prolong execution times to ensure proper solution. A novel approach is presented in this article as the approximate number of decisive iterations (ANDI ) which can be used to successfully terminate the GA optimization method with minimum execution time. Two simple correlations are presented which relate the new parameter (ANDI ) with approximate degrees of freedom (Adf ) of the merit function at hand. For complex merit functions, a linear smoother (such as Regularization network) can be used to estimate the required Adf. Four illustrative case studies are used to successfully validate the proposed approach by effectively finding the optimum point by using to the presented correlation. The linear correlation is more preferable because it is much simpler to use and the horizontal axis represents the approximate (not exact) degrees of freedom. It was also clearly shown that the Regularization Networks can successfully filter out the noise and mimic the true hyper-surface underlying a bunch of noisy data set.