arXiv subjects

Andrew Gardner

Publications and source records attributed to Andrew Gardner.

8 recordsLinked to original sources

arXiv2024

Tuning the MAPS Adaptive Secondary Mirror: Actuator Control, PID Tuning, Power Spectra and Failure Diagnosis

The MMT Adaptive optics exoPlanet characterization System (MAPS) is currently in its engineering phase, operating on-sky at the MMT Telescope on Mt. Hopkins in southern Arizona. The MAPS Adaptive Secondary Mirror's actuators are controlled by a closed loop modified PID control law and an open loop feed-forward law, which in combination allows for faster actuator response time. An essential element of achieving the secondary's performance goals involves the process of PID gain tuning. To start, we briefly discuss the design of the MAPS ASM and its actuators. We then describe the actuator positional control system and control law. Next, we discuss a few of the issues that make ASM tuning difficult. We then outline our initial attempts at tuning the actuator controllers and discuss the use of actuator positional power spectra for both tuning and determining the health and failure states of individual actuators. We conclude by presenting the results of our latest round of tuning configuration trials, which have been successful at decreasing mirror latency, increasing operational mirror modes and improving image PSF.

astro-ph.IM↗

arXiv2024

ESCAPE: Efficient Synthesis of Calibrations for Adaptive optics through Pseudo-synthetic and Empirical methods

With the commissioning of the refurbished adaptive secondary mirror (ASM) for the 6.5-meter MMT Observatory under way, special consideration had to be made to properly calibrate the mirror response functions to generate an interaction matrix (IM). The commissioning of the ASM is part of the MMT Adaptive optics exoPlanet characterization System (MAPS) upgrade the observatory's legacy adaptive optics (AO) system. Unlike most AO systems, MAPS employs a convex ASM which prevents the introduction of a calibration source capable of simultaneously illuminating its ASM and wavefront sensor (WFS). This makes calibration of the AO system a significant hurdle in commissioning. To address this, we have employed a hybrid calibration strategy we call the Efficient Synthesis of Calibrations for Adaptive Optics through Pseudo-synthetic and Empirical methods (ESCAPE). ESCAPE combines the DO-CRIME on-sky calibration method with the SPRINT method for computing pseudo-synthetic calibration matrices. To monitor quasi-static system change, the ESCAPE methodology rapidly and continuously generates pseudo-synthetic calibration matrices using continual empirical feedback in either open or closed-loop. In addition, by measuring the current IM in the background while in close-loop, we are also able to measure the optical gains for pyramid wavefront sensor (PyWFS) systems. In this paper, we will provide the mathematical foundation of the ESCAPE calibration strategy and on-sky results from its application in calibrating the MMT Observatory's ASM. Additionally, we will showcase the validation of our approach from our AO testbed and share preliminary on-sky results from MMT.

astro-ph.IM↗

arXiv2023

The MAPS Adaptive Secondary Mirror: First Light, Laboratory Work, and Achievements

The MMT Adaptive Optics exoPlanet Characterization System (MAPS) is a comprehensive update to the first generation MMT adaptive optics system (MMTAO), designed to produce a facility class suite of instruments whose purpose is to image nearby exoplanets. The system's adaptive secondary mirror (ASM), although comprised in part of legacy components from the MMTAO ASM, represents a major leap forward in engineering, structure and function. The subject of this paper is the design, operation, achievements and technical issues of the MAPS adaptive secondary mirror. We discuss laboratory preparation for on-sky engineering runs, the results of those runs and the issues we discovered, what we learned about those issues in a follow-up period of laboratory work, and the steps we are taking to mitigate them.

astro-ph.IM↗

arXiv2020

Design and development of a high-speed Visible Pyramid Wavefront Sensor for the MMT AO system

MAPS, MMT Adaptive optics exoPlanet characterization System, is the upgrade of legacy 6.5m MMT adaptive optics system. It is an NSF MSIP-funded project that includes (i) refurbishing of the MMT Adaptive Secondary Mirror (ASM), (ii) new high sensitive and high spatial order visible and near-infrared pyramid wavefront sensors, and (iii) the upgrade of Arizona Infrared Imager and Echelle Spectrograph (ARIES) and MMT high Precision Imaging Polarimeter (MMTPol) science cameras. This paper will present the design and development of the visible pyramid wavefront sensor. This system consists of an acquisition camera, a fast-steering tip-tilt modulation mirror, a double pyramid, a pupil imaging triplet lens, and a low noise and high-speed frame rate based CCID75 camera. We will report on hardware and software and present the laboratory characterization results of the individual subsystems, and outline the on-sky commissioning plan.

astro-ph.IM↗

arXiv2020

DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensembles

Recent research finds CNN models for image classification demonstrate overlapped adversarial vulnerabilities: adversarial attacks can mislead CNN models with small perturbations, which can effectively transfer between different models trained on the same dataset. Adversarial training, as a general robustness improvement technique, eliminates the vulnerability in a single model by forcing it to learn robust features. The process is hard, often requires models with large capacity, and suffers from significant loss on clean data accuracy. Alternatively, ensemble methods are proposed to induce sub-models with diverse outputs against a transfer adversarial example, making the ensemble robust against transfer attacks even if each sub-model is individually non-robust. Only small clean accuracy drop is observed in the process. However, previous ensemble training methods are not efficacious in inducing such diversity and thus ineffective on reaching robust ensemble. We propose DVERGE, which isolates the adversarial vulnerability in each sub-model by distilling non-robust features, and diversifies the adversarial vulnerability to induce diverse outputs against a transfer attack. The novel diversity metric and training procedure enables DVERGE to achieve higher robustness against transfer attacks comparing to previous ensemble methods, and enables the improved robustness when more sub-models are added to the ensemble. The code of this work is available at https://github.com/zjysteven/DVERGE

cs.LG↗

arXiv2019

Deep Detector Health Management under Adversarial Campaigns

Machine learning models are vulnerable to adversarial inputs that induce seemingly unjustifiable errors. As automated classifiers are increasingly used in industrial control systems and machinery, these adversarial errors could grow to be a serious problem. Despite numerous studies over the past few years, the field of adversarial ML is still considered alchemy, with no practical unbroken defenses demonstrated to date, leaving PHM practitioners with few meaningful ways of addressing the problem. We introduce turbidity detection as a practical superset of the adversarial input detection problem, coping with adversarial campaigns rather than statistically invisible one-offs. This perspective is coupled with ROC-theoretic design guidance that prescribes an inexpensive domain adaptation layer at the output of a deep learning model during an attack campaign. The result aims to approximate the Bayes optimal mitigation that ameliorates the detection model's degraded health. A proactively reactive type of prognostics is achieved via Monte Carlo simulation of various adversarial campaign scenarios, by sampling from the model's own turbidity distribution to quickly deploy the correct mitigation during a real-world campaign.

cs.LG↗

arXiv2018

On the Definiteness of Earth Mover's Distance and Its Relation to Set Intersection

Positive definite kernels are an important tool in machine learning that enable efficient solutions to otherwise difficult or intractable problems by implicitly linearizing the problem geometry. In this paper we develop a set-theoretic interpretation of the Earth Mover's Distance (EMD) and propose Earth Mover's Intersection (EMI), a positive definite analog to EMD for sets of different sizes. We provide conditions under which EMD or certain approximations to EMD are negative definite. We also present a positive-definite-preserving transformation that can be applied to any kernel and can also be used to derive positive definite EMD-based kernels and show that the Jaccard index is simply the result of this transformation. Finally, we evaluate kernels based on EMI and the proposed transformation versus EMD in various computer vision tasks and show that EMD is generally inferior even with indefinite kernel techniques.

cs.LG↗

arXiv2017

Classifying Unordered Feature Sets with Convolutional Deep Averaging Networks

Unordered feature sets are a nonstandard data structure that traditional neural networks are incapable of addressing in a principled manner. Providing a concatenation of features in an arbitrary order may lead to the learning of spurious patterns or biases that do not actually exist. Another complication is introduced if the number of features varies between each set. We propose convolutional deep averaging networks (CDANs) for classifying and learning representations of datasets whose instances comprise variable-size, unordered feature sets. CDANs are efficient, permutation-invariant, and capable of accepting sets of arbitrary size. We emphasize the importance of nonlinear feature embeddings for obtaining effective CDAN classifiers and illustrate their advantages in experiments versus linear embeddings and alternative permutation-invariant and -equivariant architectures.

cs.LG↗