ESO ELT M1 local control system software design and development status (Conference Presentation)

WEB, the Wind Evaluation Breadboard, is an Extremely Large Telescope Primary Mirror simulator, developed with the aim of quantifying the ability of a segmented primary mirror to cope with wind disturbances. This instrument supported by the European Community (Framework Programme 6, ELT Design Study), is developed by ESO, IAC, MEDIA-ALTRAN, JUPASA and FOGALE. The WEB is a bench of about 20 tons and 7 meter diameter emulating a segmented primary mirror and its cell, with 7 hexagonal segments simulators, including electromechanical support systems. In this paper we present the WEB central control electronics and the software development which has to interface with: position actuators, auxiliary slave actuators, edge sensors, azimuth ring, elevation actuator, meteorological station and air balloons enclosure. The set of subsystems to control is a reduced version of a real telescope segmented primary mirror control system with high real time performance but emphasizing on development time efficiency and flexibility, because WEB is a test bench. The paper includes a detailed description of hardware and software, paying special attention to real time performance. The Hardware is composed of three computers and the Software architecture has been divided in three intercommunicated applications and they have been implemented using Labview over Windows XP and Pharlap ETS real time operating system. The edge sensors and position actuators close loop has a sampling and commanding frequency of 1KHz.

We propose an adaptive run-time failure recovery control system for quadcopter drones, based on remote real-time processing of measurement data streams. Particularly, the measured RPM values of the quadcopter motors are transmitted to a remote machine which hosts failure detection algorithms and performs recovery procedure. The proposed control system consists of three distinct parts: (1) A set of computationally simple PID controllers locally onboard the drone, (2) a set of computationally more demanding remotely hosted algorithms for real-time drone state detection, and (3) a digital twin co-execution software platform — the ModelConductor-eXtended — for two-way signal data exchange between the former two. The local on-board control system is responsible for maneuvering the drone in all conditions: path tracking under normal operation and safe landing in a failure state. The remote control system, on the other hand, is responsible for detecting the state of the drone and communicating the corresponding control commands and controller parameters to the drone in real time. The proposed control system concept is demonstrated via simulations in which a drone is represented by the widely studied Quad-Sim six degrees-of-freedom Simulink model. Results show that the trained failure detection binary classifier achieves a high level of performance with F1-score of 96.03%. Additionally, time analysis shows that the proposed remote control system, with average execution time of 0.49 milliseconds and total latency of 6.92 milliseconds in two-way data communication link, meets the real-time constraints of the problem. The potential practical applications for the presented approach are in drone operation in complex environments such as factories (indoor) or forests (outdoor).

To meet ambitious science goals and leverage NASA investments for the James Webb Space Telescope, some proposed mission concepts include large aperture telescopes with segmented primary mirrors. Aberration control at the segment level becomes critical for these architectures because rigid body motion of the individual mirrors overtakes full aperture aberrations as the driver for wavefront stability. Perturbations at the segment level cannot be effectively sensed by existing full pupil low-order wavefront sensors because the errors are discontinuous and low-order techniques applied to individual segments would have insufficient photon flux. Thus, an additional “mid-order” control loop is required. We propose the use of capacitive edge sensors to locally sense the relative motion of segments in piston, tip and tilt, which then provide input to a compensation arm. Ball Aerospace has developed capacitive sensor technology with proven measurement precision of <12 pm RMS that can be adapted for measuring primary mirror segment motion. This performance approaches the 10 pm stability often stated as a requirement for direct imaging of earth-like exoplanets with a coronagraph. Using the geometry of the existing hardware as a baseline, the sensor gap and plate area can be scaled to accommodate mounting to mirror segments while maintaining or increasing sensitivity and multiple plates in different orientations can be used to sense individual degrees of freedom. This paper will present measured results from the Ball capacitive sensor and use those results to develop expected sensitives and a notional sensor head geometry for stabilizing a large, segmented primary mirror with edge sensors.

Within a difficult and uncertain environment resulting from the COVID-19 pandemic as well as the global geopolitical and economic situation, the ESO’s Extremely Large Telescope (ELT) is progressing on all fronts of design, product and process qualification, manufacturing, subsystem assembly and the two M1 segment coating plants have just been commissioned in the dedicated ELT Technical Facility building at Paranal. The scientific Instruments are now undergoing their final design reviews while series production for the various components constituting the 39 m diameter segmented primary mirror (M1) is well underway. On Cerro Armazones the raft foundations for the dome and for telescope structure are being poured and the piers start to rise above ground. On the programmatic level, all deferred subsystems (so-called phase 2 items) are now funded, and the budget has been moderately increased to account also for infrastructure upgrade needed to fully and optimally integrate the ELT into the Paranal (VLT) Observatory operation scheme. After a year of complete closure of the construction site and other delays in several contracts due to the pandemic and technical difficulties, the programme schedule has been adjusted and Scientific First Light is planned by the end of 2027.

The DSP based longitudinal multi-bunch feedback system is being developed at SRRC. Various operation modes are supported for feedback and diagnostic. The feedback system consists of a digital signal processor array to execute feedback algorithm and to provide raw data for diagnostic purposes. Simulation tools are used to aid control rule development. Software components of feedback system include feedback control rules and data acquisition in DSPs, system resources management at VME host, signal analysis and simulation in console. Hardware structure includes of front end and back end electronics, timing circuitry. In order to develop, commissioning and operate longitudinal feedback system efficiently, an integrated software environment has been developed. The commercial off-the-shelf product solution was selected in DSP environment. Network based DSP development environment provide a convenient way to code debug, beam test, and operate the feedback system. System simulate tools on console computer is integrated with SRRC control system which offer a simple way to interface the feedback system. System structure and development status will be summaries in following paragraph.

The largest telescopes (ELT) involve a highly segmented primary mirror. The monitoring of the mirror shape use thousands of sensors, located on segment edges, which measure the relative piston, tip and tilt of all segments. Today, telescopes with segmented primary mirrors use capacitance-based edge sensors. Although this technology offers excellent metrological performances, its practical use is limited by the intrinsic sensitivity to humidity, dust and condensation, whose effect exceeds the requirements for future ELTs. Being specialized in both capacitive and inductive metrology, Fogale nanotech has developed a novel concept of edge sensors using the inductive technology, which does not suffer from humidity, condensation, and dust effects. Cost effective sensor with specific layout, associated electronics, demonstrated metrological performance (sub-nanometer resolution and nanometer stability) that outperforms the capacitive concept.

The primary mirror of ESO’s Extremely Large Telescope contains 798 hexagonal segments, which are equipped with edge sensors (ES) to measure the relative segment displacement. These ES are used for M1 figure control, i.e. for maintaining the reference shape of the primary mirror. Measuring with nanometric precision in three axes with a large dynamic range of hundreds of microns, the figure loop needs to be frequently calibrated. In this paper we present a combination of simulated case studies of operational scenarios and test results gained with a test setup (M1 Test Facility) comprising seven M1 segments, 21 position actuators, and 24 ES (12 around the inner segment, 12 between the outer segments), running a scaled down version of the figure control loop. The impact of ES uncertainties originating from calibration and linearization errors on the figure control of ELT‘s primary mirror is investigated with the purpose of improving the understanding of the ES performance in the ELT covering their complete lifecycle. The paper starts with a description of the relevant subsystems before comparing performance data of the ES validation models against specification values. Then the figure loop that controls the M1 shape is described as well as the simulation environment, including a FE model of the M1 and an ES model. A brief description of the M1 Test Facility is followed by simulations of operational scenarios to study the interaction of ES uncertainties with M1 figure control. Finally, some test results measured with the figure control of the M1 Test Facility are presented.

We address the problem of unsupervised anomaly detection for multivariate data. Traditional machine learning based anomaly detection algorithms rely on specific assumptions of normal patterns and fail to model complex feature interactions and relations. Recently, existing deep learning based methods are promising for extracting representations from complex features. These methods train an auxiliary task, e.g., reconstruction and prediction, on normal samples. They further assume that anomalies fail to perform well on the auxiliary task since they are never trained during the model optimization. However, the assumption does not always hold in practice. Deep models may also perform the auxiliary task well on anomalous samples, leading to the failure detection of anomalies. To effectively detect anomalies for multivariate data, this paper introduces a teacher-student distillation based framework Distillated Teacher-Student Network Ensemble (DTSNE). The paradigm of the teacher-student distillation is able to deal with high-dimensional complex features. In addition, an ensemble of student networks provides a better capability to avoid generalizing the auxiliary task performance on anomalous samples. To validate the effectiveness of our model, we conduct extensive experiments on real-world datasets. Experimental results show superior performance of DTSNE over competing methods. Analysis and discussion towards the behavior of our model are also provided in the experiment section.

The features of two important application scenarios, supporting mobile switches with fixed end users and mobile switches with mobile users, are dramatically different from those of the traditional wired network structure. To exploit mobile switches, the location and configuration management of mobile switches is essential to handle the mobility and topology change of the wireless/mobile ATM network. In this paper we address the location management and configuration problems of mobile switches in an ATM network. We investigate several aspects of the location management problem including architecture to support switch mobility, mobile switch tracking, and mobile switch locating. We propose an approach that is based on the Private Network-Network Interface (PNNI) protocol. We extend the PNNI protocol to enable it to handle mobile switches. Moreover, we develop analytical models to determine the tracking and locating costs for mobile switches under the proposed scheme. The models illustrate the relation between total cost (tracking cost + locating cost) and peer group size. The models can be used to derive the optimal configuration for an ATM network with mobile switches.

Facebook Datacenter consists of a large number of servers that run diverse Facebook services aggregated to serve any given user request. To allow this aggregation, servers have to interact with each other via different traffic flows which are managed by networking fabric. The underlying connection powering this fabric consists of a large number of pluggable optical interconnects and On Board Optical (OBO) modules carrying production data. This connectivity at scale requires fast and reliable detection of the link failures to ensure resolution. In the first generation of the deployments, detection of the link failure was sequential and a slow process. The troubleshoot process was equally tedious as the available tools required characterizing one optical transceiver at a time. Further, the failure analysis also presented a majority of resolution with no failed optics as a root cause resulting in high No Trouble Found (NTF) rate. In this paper we introduce a novel link failure detection and resolution method that improves on the previous method across three dimensions: faster resolution, reliable troubleshooting and scalable implementation. We introduce BER Illusion Methodology (BIM) that is a highly scalable and resource efficient solution that significantly reduces the time taken to troubleshoot pluggable optical interconnects. This is also scalable to next-gen OBO modules at Facebook datacenters aiming to lower the NTF rate and optimally utilizing the available resources. BIM, which is based on Open Compute Platform (OCP) network switches, can be used to troubleshoot 128 QSFP28, 64 QSFP56 or 32 OBO modules simultaneously in under 30 minutes. The tool is easy to implement and capable of also reporting diagnostics on the transceiver such as Transmitter Power, Transmitter Bias Current, Receiver Power, Case Temperature, Bit Error Rate result per channel, Vendor information and Manufacturing part number. This additional test data report along with true failure indication helps optic suppliers gain confidence and build customer credibility. The open-source nature and the universal applicability of this tool offers possibility for other users to adopt and further customize it for their networking needs.

The European Extremely Large Telescope (ELT) construction¹ is advancing and the M2, M3 and M5 cells are running the same path. The M2² and M3 mirrors are 4-meter class Zerodur® meniscus supported axially and laterally by mechanical whiffletrees. To allow low order corrections, axial support is provided with a warping harness system. Additionally, an electromechanical hexapod allows the movement in 5 degrees of freedom to optically align the telescope. The M2 Cell is ending the integration and test phase and approaching to the verification phase, where the highly demanding requirements will be checked. On the other hand, M3 cell is being assembled, taking advantage of the lessons learned during the M2 cell assembly. The M5 mirror is lightweighted silicon carbide elliptical flat mirror of 2.7 x 2.1 m², having the objective of folding the optical beams towards the Nasmyth platforms. Additionally, stabilizes the image movements induced by the telescope mechanisms and wind shaking vibrations. The M5 Cell is divided in two stages: the Tip-tilt and the Alignment Stage. The former allows to perform a fast-steering control while the latter provides active alignment capabilities to cope with the overall telescope structure misalignments The ELT M5 Cell is currently in the final design review, with qualification models for both stages successfully developed and tested. Preliminary results shown submicrometric resolutions for the Alignment Stage and around 5 nanometres for the Tip-Tilt Stage. This paper summarizes the design of the M2, M3 and M5 cells and presents its current development status.

In optical burst switching (OBS) networks, when contention occurs at an intermediate node, two or more bursts that are in contention can be dropped because a forwarding path reservation is not made for a burst. This is the reason why one of the critical design issues in OBS networks is finding ways to minimize burst dropping resulting from resource contention. In this paper, we propose a configuration management policy based on an operation cost model developed according to network status information which relates to guaranteeing quality of service (QoS). Also, we develop the criteria for the configuration management by providing the alternate path using bounded range of the sensitivity of this cost model. Finally, throughout the theoretical and experimental analysis, we can obtain that our proposed scheme has good performance in an aspect of blocking rate and outperforms the conventional scheme in complexity as results

This paper suggests advanced video processing methods for monitoring temporary structures to detect possible structural failures. The causes of failures are mainly related to human error, but the imposed loads of structural failures can also be caused by complicated events such as accumulated stress, deflection, wind, vibration, and lateral forces. However, there is a lack of interactive tools to detect complicated failures and thus address the safety of temporary structures promptly. This paper applies video analysis techniques to concrete shoring and outlines the primary characteristics of automated detection. Concrete shoring is an important temporary structure often used to support newly poured concrete and provide stability and safety for incomplete structures in multistory work, especially concrete slabs, roofs and decks. Shoring failure seriously damages materials and equipment, as well as causing injury or even loss of life among construction crews or members of the public. The suggested approach applies video analysis to characterize the deformation of temporary structures in various failure situations and detect possible failures in early stages of deformation. The method proceeds via two steps: 1) learning and 2) detection of the failure. The first step examines deformation characteristics extracted from video sequences of simulated failure situations. Here, a Hidden Markov Model (HMM) is used to learn and to draw an inference for a possible failure and its cause. The suggested method then incorporates this into comprehensive site inspections and supervision.

The Uniform Development Environment (UDE) is being investigated as a means of enforcing software engineering standards. For the programmer, it provides an environment containing the tools and utilities necessary for orderly and controlled development and maintenance of code according to requirements (DoD-STD-2167A). In addition, it provides DoD management and developer management the tools needed for all phases of software life cycle management and control, from project planning and management, to code development, configuration management, version control, and change control. This paper reports the status of UDE development and field testing. >

In a central receiver solar power plant, heliostats are arranged with respect to the central receiver so as to reflect the rays from the sun onto the power tower with high precision by tracking the sun in both the azimuth and elevation directions. The master control system of a solar power plant consists of different levels. The first level is local control; it takes care of the positioning of the heliostats when the aiming point and the time are given to the system, and informs upper level about the status of the heliostats field. The second logic level makes some important dispatch calculations of heliostats field. The most popular linear two-axis local driving system of heliostat consists of two linear driving actuators, the driving mechanism with rotary joints, and the controller. Traditional methods for heliostat design are often based on a sequential approach in which the mechanical structure is designed first and then the control system is advised. In order to reach the optimal design of heliostats, an integrated design approach that concurrently considers the interactions between the mechanical and control subsystems is necessary. In this article, an integrated design methodology of heliostat drive system is presented. The methodology is based on modeling and simulation. The dynamic models that describe the behavior of the mechanical and control components are presented. These models involve mechanical and control design variables such as the motor parameters, power screw (including back lash), heliostat mass, load forces, and wind forces. Matlab, Solidwork, and Simulink are chosen to apply PID tracking control to heliostats, due to the ability to arbitrarily model complex mechanical systems, directly import properly constructed, third-party 3D CAD models, simulate integrated control, handle a variety of robotics nomenclature, and other features. The present methodology is employed for integrated design of a single facet small size heliostat with mirror area of 3 m2.The methods described in this article also show a way to rapidly simulate novel and complex heliostat geometries. Analysis of the heliostat drive system performance and dynamic characteristics according to mechanical and control design variables is conducted for the purpose of control system design and performance optimization. The drive system performance is evaluated in terms of positioning tracking errors, system response, and control system behavior. It is shown that the mechanical characteristics of the ball power screw actuator such as ball-screw diameter, lead, overall flexibility, stiffness, backlash, and inertia significantly influence the performance of drive system.