The Narrow Field Infrared Adaptive Optics System (NFIRAOS) is the first light Adaptive Optics (AO) system for the Thirty Meter Telescope (TMT). A critical component of NFIRAOS is the Real-Time Controller (RTC) subsystem which provides real-time wavefront correction by processing wavefront information to compute Deformable Mirror (DM) and Tip/Tilt Stage (TTS) commands. The National Research Council of Canada - Herzberg (NRC-H), in conjunction with TMT, has developed a preliminary design for the NFIRAOS RTC. The preliminary architecture for the RTC is comprised of several Linux-based servers. These servers are assigned various roles including: the High-Order Processing (HOP) servers, the Wavefront Corrector Controller (WCC) server, the Telemetry Engineering Display (TED) server, the Persistent Telemetry Storage (PTS) server, and additional testing and spare servers. There are up to six HOP servers that accept high-order wavefront pixels, and perform parallelized pixel processing and wavefront reconstruction to produce wavefront corrector error vectors. The WCC server performs low-order mode processing, and synchronizes and aggregates the high-order wavefront corrector error vectors from the HOP servers to generate wavefront corrector commands. The Telemetry Engineering Display (TED) server is the RTC interface to TMT and other subsystems. The TED server receives all external commands and dispatches them to the rest of the RTC servers and is responsible for aggregating several offloading and telemetry values that are reported to other subsystems within NFIRAOS and TMT. The TED server also provides the engineering GUIs and real-time displays. The Persistent Telemetry Storage (PTS) server contains fault tolerant data storage that receives and stores telemetry data, including data for Point-Spread Function Reconstruction (PSFR).
Prototyping and benchmarking was performed for the Real-Time Controller (RTC) of the Narrow Field InfraRed Adaptive Optics System (NFIRAOS). To perform wavefront correction, NFIRAOS utilizes two deformable mirrors (DM) and one tip/tilt stage (TTS). The RTC receives wavefront information from six Laser Guide Star (LGS) Shack- Hartmann WaveFront Sensors (WFS), one high-order Natural Guide Star Pyramid WaveFront Sensor (PWFS) and multiple low-order instrument detectors. The RTC uses this information to determine the commands to send to the wavefront correctors. NFIRAOS is the first light AO system for the Thirty Meter Telescope (TMT).
The prototyping was performed using dual-socket high performance Linux servers with the real-time (PREEMPT_RT) patch and demonstrated the viability of a commercial off-the-shelf (COTS) hardware approach to large scale AO reconstruction. In particular, a large custom matrix vector multiplication (MVM) was benchmarked which met the required latency requirements. In addition all major inter-machine communication was verified to be adequate using 10Gb and 40Gb Ethernet. The results of this prototyping has enabled a CPU-based NFIRAOS RTC design to proceed with confidence and that COTS hardware can be used to meet the demanding performance requirements.
Raven is a multi-object adaptive optics (MOAO) demonstrator that will be mounted on the NIR Nasmyth platform of the Subaru telescope in May, 2014. Raven can use three open-loop NGS WFSs and an on-axis LGS WFS to control DMs in two separate science pick-off arms. Centroiding in open loop AO systems like Raven is more difficult than in closed loop AO systems because the Shack-Hartmann spots will not be driven to the same spot on a detector. Rather the spots can fall on any combination of pixels because the WFSs need to have sufficient dynamic range to measure the full turbulence. In this paper, we compare correlation and thresholded center of gravity (tCOG) centroiding methods in simulation, with Raven using its calibration unit, and on-sky. Each method has its own advantages. Correlation centroiding is superior to tCOG centroiding for faint NGSs and for extended sources (Raven open loop WFSs do not contain ADCs so spots will become elongated). We expect that correlation centroiding will push the limiting magnitude of Raven NGSs fainter by roughly one magnitude. Correlation centroiding is computationally more intensive, however, and actually will limit Raven’s sampling rate for shorter integrations. Therefore, for bright stars with sufficiently high signal-to-noise, Raven can be run significantly faster and with superior performance using the tCOG method. Here we quantify both the performance and timing differences of these two centroiding methods in simulation, in the lab and on sky using Raven.
The Gemini Planet Imager (GPI) is a facility extreme-AO high-contrast instrument – optimized solely for study of faint companions – on the Gemini telescope. It combines a high-order MEMS AO system (1493 active actuators), an apodized pupil Lyot coronagraph, a high-accuracy IR post-coronagraph wavefront sensor, and a near-infrared integral field spectrograph. GPI incorporates several other novel features such as ultra-high quality optics, a spatially-filtered wavefront sensor, and new calibration techniques. GPI had first light in November 2013. This paper presnets results of first-light and performance verification and optimization and shows early science results including extrasolar planet spectra and polarimetric detection of the HR4696A disk. GPI is now achieving contrasts approaching 10-6 at 0.5” in 30 minute exposures.
The Gemini Planet Imager (GPI) is an “extreme” adaptive optics coronagraph system that is now on the
Gemini South telescope in Chile. This instrument is composed of three different systems that historically have
been separate instruments. These systems are the extreme Adaptive Optics system, with deformable mirrors,
including a high-order 64x64 element MEMS system; the Science Instrument, which is a near-infrared
integral field spectrograph; and the Calibration system, a precision IR wavefront sensor that also holds
key coronagraph components. Each system coordinates actions that require precise timing. The
observatory is responsible for starting these actions and has typically done this asynchronously across
independent systems. Despite this complexity we strived to provide an interface that is as close to a onebutton
approach as possible. This paper will describe the sequencing of these systems both internally and
externally through the observatory.
As a part of the trade study for the Narrow Field Infrared Adaptive Optics System, the adaptive optics system for the Thirty Meter Telescope, we investigated the feasibility of performing real-time control computation using a Linux operating system and Intel Xeon E5 CPUs. We also investigated a Xeon Phi based architecture which allows higher levels of parallelism. This paper summarizes both the CPU based real-time controller architecture and the Xeon Phi based RTC. The Intel Xeon E5 CPU solution meets the requirements and performs the computation for one AO cycle in an average of 767 microseconds. The Xeon Phi solution did not meet the 1200 microsecond time requirement and also suffered from unpredictable execution times. More detailed benchmark results are reported for both architectures.
With two large deformable mirrors with a total of more than 7000 actuators that need to be driven from the measurements of six 60x60 LGS WFSs (total 1.23Mpixels) at 800Hz with a latency of less than one frame, NFIRAOS presents an interesting real-time computing challenge. This paper reports on a recent trade study to evaluate which current technology could meet this challenge, with the plan to select a baseline architecture by the beginning of NFIRAOS construction in 2014. We have evaluated a number of architectures, ranging from very specialized layouts with custom boards to more generic architectures made from commercial off-the-shelf units (CPUs with or without accelerator boards). For each architecture, we have found the most suitable algorithm, mapped it onto the hardware and evaluated the performance through benchmarking whenever possible. We have evaluated a large number of criteria, including cost, power consumption, reliability and flexibility, and proceeded with scoring each architecture based on these criteria. We have found that, with today’s technology, the NFIRAOS requirements are well within reach of off-the-shelf commercial hardware running a parallel implementation of the straightforward matrix-vector multiply (MVM) algorithm for wave-front reconstruction. Even accelerators such as GPUs and Xeon Phis are no longer necessary. Indeed, we have found that the entire NFIRAOS RTC can be handled by seven 2U high-end PC-servers using 10GbE connectivity. Accelerators are only required for the off-line process of updating the matrix control matrix every ~10s, as observing conditions change.
An Atmospheric Dispersion Corrector (ADC) uses a double-prism arrangement to nullify the vertical chromatic
dispersion introduced by the atmosphere at non-zero zenith distances.
The ADC installed in the Gemini Planet Imager (GPI) was first tested in August 2012 while the instrument was
in the laboratory. GPI was installed at the Gemini South telescope in August 2013 and first light occurred later
that year on November 11th.
In this paper, we give an overview of the characterizations and performance of this ADC unit obtained in the
laboratory and on sky, as well as the structure of its control software.
The Gemini Planet Imager is a next-generation instrument for the direct detection and characterization of young warm exoplanets, designed to be an order of magnitude more sensitive than existing facilities. It combines a 1700-actuator adaptive optics system, an apodized-pupil Lyot coronagraph, a precision interferometric infrared wavefront sensor, and a integral field spectrograph. All hardware and software subsystems are now complete and undergoing integration and test at UC Santa Cruz. We will present test results on each subsystem and the results of end-to-end testing. In laboratory testing, GPI has achieved a raw contrast (without post-processing) of 10-6 5σ at 0.4”, and with multiwavelength speckle suppression, 2x10-7 at the same separation.
NFIRAOS is the first-light adaptive optics system planned for the Thirty Meter Telescope, and is being designed at the
National Research Council of Canada's Herzberg Institute of Astrophysics. NFIRAOS is a laser guide star multiconjugate
adaptive optics system - a practical approach to providing diffraction limited image quality in the NIR over a
30" field of view, with high sky coverage. This will enable a wide range of TMT science that depends upon the large
corrected field of view and high precision astrometry and photometry. We review recent progress developing the design
and conducting performance estimates for NFIRAOS.
The instrument group of the Herzberg Institute of Astrophysics has been commissioned by the Gemini Observatory to
develop and implement a new focal plane assembly with an array of three Hamamatsu CCDs for the Gemini Multi-
Object Spectrographs[1,2]. This paper describes the overall design of the new focal plane system with respect to the
existing interface and requirements and outlines the test methodology to validate the new system against its performance
requirements. The characterization and performance optimization processes of the Hamamatus CCDs are also described.
The next generation of ground-based optical telescopes will employ increasingly large primary mirrors to achieve
superior resolution and light collecting abilities. Many of these large mirror surfaces will be segmented into an array of
hundreds of smaller mirror segments. The corresponding number of required sensors and actuators will be in the order of
thousands, which creates a challenging control problem to stabilize and align each segment from external disturbances - wind shake, gravity forces, thermal effects, seismic effects and induced vibrations from surrounding equipment and
telescope motion - so that the telescope's image quality requirements can be met. The use of a centralized control
scheme may be infeasible due to the large number of inputs and outputs of the resulting control system, while a
decentralize control scheme would lack global performance. An attractive alternative approach is an interconnected
network of distributed controllers that provide global control with a highly scalable design and implementation. A
segmented mirror can be considered as an interconnected system comprised of many similar discrete subsystems, where
each subsystem represents an individual mirror segments and its dynamics are coupled directly to its neighboring
segments. The resulting distributed controller network of controller subsystems are similarly coupled and working
cooperatively to achieve the desired global performance.
The Gemini Planet Imager (GPI) is an "extreme" adaptive optics coronagraph system that will have the ability to directly
detect and characterize young Jovian-mass exoplanets. The design of this instrument involves eight principal institutions
geographically spread across North America, with four of those sites writing software that must run seamlessly together
while maintaining autonomous behaviour. The objective of the software teams is to provide Gemini with a unified
software system that not only performs well but also is easy to maintain. Issues such as autonomous behaviour in a
unified environment, common memory to share status and information, examples of how this is being implemented,
plans for early software integration and testing, command hierarchy, plans for common documentation and updates are
explored in this paper. The project completed its preliminary design phase in 2007, and has just recently completed its
critical design phase.
We describe a tool to analyze the effects of gravity induced deflections on a telescope structure with segmented primary mirror optics. An objective of the telescope structural design process is to minimize image quality degradation due to uncorrectable static deflections of the optics under gravity, while ensuring that the overall system meets several requirements including limits of maximum primary mirror actuator stroke, segment rotation and decenter, and secondary mirror actuation. These design and performance criteria are not readily calculated within a finite element program. Our Merit Function routine, implemented in MATLAB and called by ANSYS, calculates these parameters and makes them available within ANSYS for evaluation and design optimization. In this analysis, ANSYS outputs key structural-model nodal displacements to a file, which are used to determine the 6 degree of freedom motion of the telescope's optical surfaces. MATLAB then utilizes these displacements, along with a database containing coordinate system transforms and a linear optics model derived from ZEMAX, to calculate various performance criteria. The values returned to ANSYS can be used to iteratively optimize performance over a set of structural design parameters. Optical parameters calculated by this routine include the optical path difference at the pupil, RMS wavefront, encircled energy and low order Zernike terms resulting from primary mirror segment rotation and decenter. Also reported are the maximum actuator strokes required to restore tip-tilt and piston of the primary mirror segments, and the deflection of the secondary mirror under gravitational load. The merit function routine is being used by the Thirty Meter Telescope (TMT) project to optimize and assess the performance of various telescope structural designs. This paper describes the mathematical basis of the calculations, their implementation and gives preliminary results of the TMT Telescope Structure Reference Design.
The Herzberg Institute of Astrophysics has developed Integrated Modeling tools for the Thirty Meter Telescope (TMT) project. This simulation software, implemented in MATLAB, models the telescope optical system, structural dynamics, and the segmented primary mirror and secondary mirror active optical control systems. For the TMT project, the integrated model was used to assess the effect of wind loading on the telescope in terms of delivered image quality. The simulation includes a state-space model of the telescope structural dynamics derived from an ANSYS finite element model, a linear optics model derived from a ZEMAX prescription, and wind loading forces derived from PowerFLOW computational dynamics software. The overall complexity of the model necessitates rigorous validation and verification procedures to ensure that the simulation data structures properly represent the original design, and that the calculations performed by the system are reliable. In this paper we discuss the validation and verification of the model data structures, structural configuration, optical configuration, coordinate system transforms, linear optics model, Zernike calculations, and wind loading model. As a case study, we present the verification, validation, and simulation results of the Thirty Meter Telescope Reference Design.
The National Research Council's Herzberg Institute of Astrophysics (NRC-HIA) has developed an opto-mechanical integrated modeling toolset called TM-IM. This time-domain state-space toolset has been implemented using Matlab/Simulink/C. The toolset was originally developed for the Very Large Optical Telescope (VLOT) design work, and continued when Canada joined in the Thirty Meter Telescope (TMT) project. The TM-IM toolset has been developed to accommodate different structural and optical designs and has been used to evaluate telescope performance to assist in making decisions for the TMT reference design expected fall 2004. Preliminary results include delivered image quality as a function of wind loading on the structure, primary and secondary mirror, and the simulation of an Adaptive Optics system which provides control feedback to the primary mirror.
The next generation of ground-based telescopes will have apertures of 20 meters or more and will be increasingly dependent on active and adaptive optics (AO) to deliver good image quality. A numerical model of the complete telescope system, including optical, mechanical, and atmospheric seeing effects, will be a vital tool during the design process. The Thirty Meter Telescope (TMT) / Very Large Optical Telescope (VLOT) Integrated Model (IM) is written in MATLAB and runs on a Windows PC. One goal of the IM is to study the interaction of various AO designs with several telescope configurations. This requires the inclusion of an AO simulation engine; the IDL-based CAOS code was chosen as a starting point. Socket based software was developed to allow MATLAB MEX functions called from the IM to control the CAOS code running on a Linux PC. Software was also developed to allow MATLAB MEX functions to interact with IDL on the same Windows computer using callable IDL.