scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Predictor-corrector framework for the sequential assembly of optical systems based on wavefront sensing.

16 Apr 2018-Optics Express (Opt Express)-Vol. 26, Iss: 8, pp 10669-10681
TL;DR: This paper uses a hybrid optical simulation model that comprises virtual and identified component positions that enables prediction of the future wavefront at the detector plane and therefore allows for taking corrective measures accordingly during the assembly process if a user-defined tolerance on the wavefront error is violated.
Abstract: Alignment of optical components is crucial for the assembly of optical systems to ensure their full functionality. In this paper we present a novel predictor-corrector framework for the sequential assembly of serial optical systems. Therein, we use a hybrid optical simulation model that comprises virtual and identified component positions. The hybrid model is constantly adapted throughout the assembly process with the help of nonlinear identification techniques and wavefront measurements. This enables prediction of the future wavefront at the detector plane and therefore allows for taking corrective measures accordingly during the assembly process if a user-defined tolerance on the wavefront error is violated. We present a novel notation for the so-called hybrid model and outline the work flow of the presented predictor-corrector framework. A beam expander is assembled as demonstrator for experimental verification of the framework. The optical setup consists of a laser, two bi-convex spherical lenses each mounted to a five degree-of-freedom stage to misalign and correct components, and a Shack-Hartmann sensor for wavefront measurements.
Citations
More filters
Journal ArticleDOI
TL;DR: Different manufacturing defect simulation methods are classified as random noise methods, mesh morphing methods, and mode-based methods and the theoretical backgrounds of these methods are introduced, and the simulation examples are conducted on a consistency model to show their differences.
Abstract: Computer-aided tolerancing (CAT) aims to predict and control geometrical and dimensional deviations in the early design stage. Former simulation models based on the translation and rotation of nominal features cannot fulfill engineering demands or cover the product lifecycle. Nonideal feature-based simulation methods are, therefore, drawing a great deal of research attention. Two general problems for non-ideal feature-based methods are how to simulate manufacturing defects and how to integrate these defects into tolerance analysis. In this paper, we focus on the first problem. There are already many manufacturing defect simulation methods. Although they are derived from different fields and have different names, they share common characteristics in application. In this study, we collected different simulation methods and classified them as random noise methods, mesh morphing methods, and mode-based methods. The theoretical backgrounds of these methods are introduced, and the simulation examples are conducted on a consistency model to show their differences. Criteria such as multiscale, surface complexity, measurement data integration, parametric control, and calculation complexity are proposed to compare these methods. Based on these analyses, the advantages and drawbacks of each method are pointed out, which may help researchers and engineers to choose suitable methods for their work.

23 citations

Journal ArticleDOI
TL;DR: Wang et al. as mentioned in this paper proposed an underwater ghost imaging method based on the generative adversarial networks, which adopts U-Net with the double skip connections and the attention module to improve the reconstruction quality.
Abstract: Ghost imaging is widely used in underwater active optical imaging because of its simple structure, long distance, and non-local imaging. However, the complexity of the underwater environment will greatly reduce the imaging quality of ghost imaging. To solve this problem, an underwater ghost imaging method based on the generative adversarial networks is proposed in the study. The generator of the proposed network adopts U-Net with the double skip connections and the attention module to improve the reconstruction quality. In the network training process, the total loss function is the sum of the weighted adversarial loss, perceptual loss, and pixel loss. The experiment and simulation results show that the proposed method effectively improves the target reconstruction performance of underwater ghost imaging. The proposed method promotes the further development of active optical imaging of underwater targets based on ghost imaging technology.

14 citations

Journal ArticleDOI
23 Jul 2019
TL;DR: This letter employs a macro–micro manipulator for moving optical components and utilize filtering methods for realizing an in-process state estimation and compares these methods in simulation with current nonlinear approaches from the literature with respect to the estimation error.
Abstract: To date, the assembly of optical systems is still not fully automated. Automated assembly can be facilitated by having an optical simulation at hand, which, in turn, requires knowledge about the current state of the optical system. Due to the strict demands on the positioning tolerances, the uncertainties of the positioning system play an important role and lead to non-negligible deviations from the nominal poses. Therefore, the actual poses of the optical components need to be estimated in order to correct misaligned components. Furthermore, it is beneficial to develop methods that utilize the dedicated primary sensor and to avoid additional external sensors. In this letter, we employ a macro-micro manipulator for moving optical components and utilize filtering methods for realizing an in-process state estimation. For this, the uncertainty of the positioning system, as well as sensor noise, need to be identified, which lay the groundwork for methods such as the extended Kalman filter, iterated extended Kalman filter, unscented Kalman filter, or particle filter. In this letter, we compare these methods in simulation with current nonlinear approaches from the literature with respect to the estimation error. Experimental verification is carried out by a macro-micro manipulator comprised of a Cartesian piezo-driven 3-DOF positioning system attached to a 6-DOF industrial robot. With the proposed filtering approach and macro-micro manipulator, the pose of a bi-convex lens is estimated via a wavefront sensor.

6 citations


Cites background from "Predictor-corrector framework for t..."

  • ...However, the identification process can be repeated in a straightforward manner for each (additional) optical component to be assembled which allows the integration into more complex assembly procedures [12]....

    [...]

  • ...III-B with common nonlinear approaches based on a quadratic cost function [10], [12]....

    [...]

  • ...and experiment, the directions with low sensitivity are z, θx, and θy and directions with high sensitivity are x and y [12]....

    [...]

  • ...The aforementioned methods do not account for sensor noise and positioning uncertainty and the identification process might result in local minima due to the optical sensitivity [12]....

    [...]

Journal ArticleDOI
TL;DR: In this paper , a framework for data-driven geometric deviation generation is proposed for non-ideal surface modeling in tolerancing, where deviations that are classified as systematic and random types are represented and stored as samples in the dataset.
Abstract: The digitalization techniques enable quality control as well as performance simulation of mechanical products in the Digital Twin era. In the product design process, realistic models regarding surface geometrical deviations are essential for further functional analysis. In the context of ISO standards on geometrical product specifications and verification (GPS), the Skin Model Shapes (SMSs) concept stemmed from the Skin Models paradigm is put forward to represent deviations of mechanical parts in accordance with the nature of geometric deviations on the surfaces. Contributions have been made to generate SMSs by combining both systematic deviations and random deviations on the nominal model. However, the non-ideal surface generation process is limited by the reuse of geometric deviation’s knowledge. Existing geometric deviations that are predicted or observed on the parts cannot be applied for other SMSs generation without new parameter settings using different methods. In this paper, a framework for data-driven geometric deviation generation is proposed for non-ideal surface modeling in tolerancing. A database that contains a variety of deviations generated by different methods is constructed. Deviations that are classified as systematic and random types are represented and stored as samples in the dataset. Deviation pattern identification is addressed by transfer learning using AlexNet and it is used for different types of non-ideal surface generation. Different approaches regarding both prediction and observation stage within SMS schema are included for constructing the deviation samples. The mapping process for both systematic and random deviation samples are considered. Therefore, there is no need to re-design deviations with parameters for the surfaces. A case study is used to illustrate the non-ideal surface generation process by the propose method.

4 citations

Proceedings ArticleDOI
05 Nov 2018
TL;DR: Wavefront measurements along with position identification methods can be utilized to continuously update a simulation model which allows for predictions on future wavefront errors and enables to take according correction measures during the assembly process if a certain wavefront tolerance specification is not met.
Abstract: Industrial assembly of optical systems is still a tedious and cost-intensive task that is mostly dominated by manual labor. Positional fine-adjustment of optical components is pivotal to ensure a desired performance of the optical device at hand. In this paper, we use wavefront predictions to aim for fully automated assembly procedures. Wavefront measurements along with position identification methods can be utilized to continuously update a simulation model which in turn allows for predictions on future wavefront errors. This enables to take according correction measures during the assembly process if a certain wavefront tolerance specification is not met. In order to demonstrate the efficacy of the proposed approach and methods, a beam expander is sequentially assembled. The setup consists of a laser, two bi-convex lenses, and a Shack-Hartmann wavefront sensor and has to satisfy a certain wavefront tolerance specification after its assembly.

3 citations

References
More filters
Journal ArticleDOI
TL;DR: It is found that the ability of all these solvers to obtain good solutions diminishes with increasing problem size, and TomLAB/MULTIMIN, TOMLAB/GLCCLUSTER, MCS and TOMLab/LGO are better, on average, than other derivative-free solvers in terms of solution quality within 2,500 function evaluations.
Abstract: This paper addresses the solution of bound-constrained optimization problems using algorithms that require only the availability of objective function values but no derivative information. We refer to these algorithms as derivative-free algorithms. Fueled by a growing number of applications in science and engineering, the development of derivative-free optimization algorithms has long been studied, and it has found renewed interest in recent time. Along with many derivative-free algorithms, many software implementations have also appeared. The paper presents a review of derivative-free algorithms, followed by a systematic comparison of 22 related implementations using a test set of 502 problems. The test bed includes convex and nonconvex problems, smooth as well as nonsmooth problems. The algorithms were tested under the same conditions and ranked under several criteria, including their ability to find near-global solutions for nonconvex problems, improve a given starting point, and refine a near-optimal solution. A total of 112,448 problem instances were solved. We find that the ability of all these solvers to obtain good solutions diminishes with increasing problem size. For the problems used in this study, TOMLAB/MULTIMIN, TOMLAB/GLCCLUSTER, MCS and TOMLAB/LGO are better, on average, than other derivative-free solvers in terms of solution quality within 2,500 function evaluations. These global solvers outperform local solvers even for convex problems. Finally, TOMLAB/OQNLP, NEWUOA, and TOMLAB/MULTIMIN show superior performance in terms of refining a near-optimal solution.

1,183 citations

Journal ArticleDOI
TL;DR: A system for automatic alignment of optical interferometers relies on using differential phase modulation to permit the detection of the phase difference between two fundamental-mode Gaussian beams at the output of an interferometer.
Abstract: We present a description of a system for automatic alignment of optical interferometers. The technique relies on using differential phase modulation to permit the detection of the phase difference between two fundamental-mode Gaussian beams at the output of an interferometer. Measurements of the spatially varying phase difference between the two beams by use of one or more multielement photodiodes permits information to be derived about the mismatch in overlap between the phase fronts at the output of the interferometer.

259 citations

01 Jan 2013
TL;DR: In IC manufacturing, lithographic scanners expose a circuit pattern onto a semiconductor wafer by means of an optical system as discussed by the authors, and the stage holding the wafer must have a scanning position accuracy of only a few nanometers to support imaging and overlay requirements.
Abstract: In IC manufacturing, lithographic scanners expose a circuit pattern onto a semiconductor wafer by means of an optical system. The stage holding the wafer must have a scanning position accuracy of only a few nanometers to support imaging and overlay requirements. In the past 10 years, stage acceleration, position error and settling time have all improved 5-8 times. Compared to lithographic steppers of 25 years ago, the number of controlled stage axes has grown from 3 (analog) to 50 (high-speed digital).

205 citations

Journal ArticleDOI
TL;DR: A new merit function regression method that has the potential to bring to such an optical alignment process higher efficiency and accuracy than the conventional sensitivity table method is reported.
Abstract: The precision alignment of high-performance, wide-field optical systems is generally a difficult and often laborious process. We report a new merit function regression method that has the potential to bring to such an optical alignment process higher efficiency and accuracy than the conventional sensitivity table method. The technique uses actively damped least square algorithm to minimize the Zernike coefficient-based merit function representing the difference between the designed and misaligned optical wave fronts. The application of this method for the alignment experiment of a Cassegrain type collimator of 900mm in diameter resulted in a reduction of the mean system rms wave-front error from 0.283 lambda to 0.194 lambda;, and in the field dependent wave-front error difference from +/-0.2 lambda to +/-0.014 lambda in just two alignment actions. These results demonstrate a much better performance than that of the conventional sensitivity table method simulated for the same steps of experimental alignment.

76 citations

Journal ArticleDOI
TL;DR: A generalized reverse-optimization alignment solution algorithm using Zernike sensitivity, and proposes the minimum number of fields to take interferograms that was successfully applied on a Cassegrain telescope design for Earth observation from space with arbitrary misalignments and a model including some primary mirror deformation.
Abstract: When aligning catoptric or catadioptric telescopes for space cameras, it is difficult to align precisely if the field of view is large or there are several reflective surfaces. The quantitative knowledge of mirror misalignments greatly helps align a misaligned telescope precisely, and also reduce the alignment time. This paper describes a generalized reverse-optimization alignment solution algorithm using Zernike sensitivity, and proposes the minimum number of fields to take interferograms. This method was successfully applied on a Cassegrain telescope design for Earth observation from space with arbitrary misalignments and a model including some primary mirror deformation.

42 citations