Search

The development of software architectures that ensure both a high-level of autonomy in the mission, and the robustness to possible failures, is a challenging task. In this paper, we propose to structure the software architecture around a skill management layer, based on formal skill models. This skill management layer helps to structure and test the underlying functional layer, while it provides a simple abstraction to the decision layer. This architecture has been used to support the implementation of resilient behaviours, using Behaviour Trees, in autonomous UAV missions, when facing sensor failures or communication losses.

Gliders constitute a new generation of autonomous underwater platforms. The SeaExplorer glider, developed and commercialized by Alseamar, is a buoyancy-controlled vehicle that profiles from the surface to 1000 m-depth along reprogrammable routes. In the framework of AGLIMMS (Acoustic GLIders Mission Management System) project, a centralized web-based platform for the SeaExplorer gliders has been developed and is presented here. The development of such a tool aimed to bring together very different stakes such as piloting assistance, help for mission planning or interpretation of glider data in a 4-D context. In this short article, a focus is made on satellite products and concrete examples are given to highlight the platforms capabilities.

Depression is a serious mental illness that is frequently resistant to a first round of pharmacotherapy. Electroconvulsive therapy (ECT) is effective even for such treatment resistant depression but is associated with significant adverse effects. Repetitive transcranial magnetic stimulation (rTMS) over the left dorsolateral prefrontal cortex in comparison causes only mild discomfort but is less effective than ECT. We hypothesize that TMS treatment efficacy could be improved by locking TMS onset to a specific, potentially subject specific phase of the prefrontal alpha rhythm in the electroencephalogram (EEG). Here, we present an instrument that can track and predict phase of the alpha rhythm in the EEG to precisely target TMS while concurrently recording functional magnetic resonance imaging (fMRI) to study local and distributed hemodynamic brain responses to stimulation. Tests of the instrument with three healthy adults indicate that EEG phase-locked TMS can be administered accurately enough to start testing systematically whether specific stimulation protocols can lead to clinically significant improvements in depression. To our knowledge, this is the first system that can deliver TMS phase-locked to the alpha rhythm while concurrently recording fMRI. For patients, such EEG guided TMS treatment could lead to better clinical outcomes and lower incidence of adverse effects.

Mixed criticality systems combine applications with different levels of safety within the same platform. Those applications execute under different modes, each representing an execution condition that can be required at runtime. This work proposes a mixed criticality schedulability analysis which is a modified version of the earliest deadline first scheduling paradigm. The analysis covers every possible execution condition by providing schedulability that is parametrized by the safety level. Besides, sensitivity analysis is developed for evaluating the impact of mixing criticality to schedulability and defining trade-offs between resource usage and schedulability. Test cases of real-time cyber physical systems are investigated to validate the mixed criticality analysis framework presented.

A parametric direction-of-arrival (DoA) estimation technique is investigated for an advanced multi-mode antenna design. It is proven that multi-mode antennas are suitable for this purpose. Possible applications include swarm exploration and indoor positioning.

Future wireless networks put high requirements on many system components. Data rates of the order of 100 Gbps and beyond, currently used in wired optical networks only, are expected for 5G applications. This imposes high effort with respect to code design and decoding algorithms. Due to their low complexity, LDPC codes are an essential part of many modern communication systems. Additionally, high symbol rates will require high-rate analog-to-digital conversion which turns out to be very power consuming. In order to explore trade-offs between hardware complexity and digital baseband processing, this contribution investigates the bit error rate performance and soft Monte Carlo simulation of LDPC codes together with reduced-complexity analog-to-digital conversion.

Developing robotic applications requires to design and implement complex software architectures. These architectures must embed advanced algorithms that include capacities to adapt to unforeseen events like external disturbances, sensor or actuator failures. To improve the system robustness, its behavior should be adapted at runtime by a reconfiguration of its software architecture. Such reconfiguration must be done safely and efficiently, while ensuring functional constraints and a minimal quality of service of the system. Among these constraints, preserving real-time properties of the reconfiguration process is a key feature. In this paper, we present the preliminary design of a new component-based middleware that allows to perform software architecture reconfigurations with a focus on real-time constraints.

Nowadays dangerous, repetitive or precision requiring jobs are done by robots like flying drones, industrial assembly arms or medical assistants. In all these cases, human beings can interact with the machines, therefore it is essential to guarantee that every part of the robot's software and hardware will produce a safe behavior: both the overall behavior and the local behaviors of such embedded systems have to be carefully analyzed. The complexity of embedded systems software architectures increase with more and more tasks involved; state-machines are applied to implement more functional capabilities of the tasks; and the task models used in the analyses gain in complexity. The analysis techniques have to be adapted in order to face such new complexities. This paper focuses on the real-time analysis of state-machine-based software architectures. We propose a method to analyze the temporal behavior of a component-based architecture in which the components are described by state-machines. The method computes an accurate worst-case response time by taking into account the state-machines of the components. Finally, we validate our approach with a real real-time robotic case study.

A planar superconducting undulator (SCU), which generates horizontal and vertical fields, B x and B y , perpendicular to the beam directions, is inserted in between the magnetic poles of another unit, which generates the vertical field B y . Analytical formulae of the magnetic fields are presented for the inserted, as well as the vertical-field, units. A scaling law may be applied to the SCU. The angle of the coil windings for the inserted unit is analyzed to maximize B x . The range of the optimum rotation angle, for the range of gap/period ratio 0.1 – 0.6, is calculated to be 30° – 40°.

Radiation-induced magnetic field strength losses are seen in undulator permanent magnets in the two sectors with small-aperture (5 mm) vacuum chambers. Initially, simple retuning of the affected undulators could restore them to full operation. As the damage has accumulated, however, it has become necessary to disassemble the magnetic arrays and either replace magnet blocks or remagnetize and reinstall magnet blocks. Some of the damaged magnet blocks have been studied, and the demagnetization was found to be confined to a limited volume at the surface close to the electron beam. Models for the magnetic damage were calculated using RADIA and were adjusted to reproduce the measurements. Results suggest that a small volume at the surface has acquired a weak magnetization in the opposite direction. Simulations of the radiation environment at the undulators have been performed with the MARS15 code.

Superconducting undulators (SCUs) with a period of 14.5 mm are under development for the Advanced Photon Source (APS). The undulators have been designed to achieve a peak field on the beam axis higher than 0.8 T with an 8 mm pole gap and current densities over 1 kA/mm2in the NbTi and Nb 3 Sn coils. Upper-half NbTi SCUs of short sections have been fabricated and were charged up to near the critical current density of 1.43 kA/mm2to achieve a peak field about 1 T. The stability margin of the SCU was measured by imposing steady-state heat fluxes on the pole/coil face of the SCU in a pool-boiling liquid He (LHe) dewar at 4.2 K. Near the critical current density, where the temperature stability margin is minimal, the heat flux density to quench the SCU was about 1.3 mW/mm2, of which 60% was attributed to LHe at the interface of the SCU and the vacuum chamber. The peak fields of the SCU were mapped along the beam axis using a Hall probe in a vertical dewar. The first test of a Nb 3 Sn short-section SCU was charged to an average current density of 1.45 kA/mm2, slightly higher than the critical current density for the NbTi SCU.

A planar superconducting undulator (SCU) with a period of 15 mm is under development at the Advanced Photon Source (APS). The intended users require a photon energy that can be tuned from 19 to 28 keV for inelastic X-ray scattering studies. The SCU design consists of two low-carbon-steel cores that are positioned above and below the beam chamber. There are 20 turns of NbTi/Cu superconducting (SC) wire within a coil cross section of 4.3 /spl times/ 4.0 (w /spl times/ h) mm/sup 2/. At a pole gap of 8 mm, the necessary average current density in the coil will be about 1 kA/mm/sup 2/ to achieve a peak field of 0.8 T on the beam axis. The design and fabrication progress of a 12-period prototype SCU are presented, and some challenging requirements are discussed.

At the Advanced Photon Source (APS) at Argonne National Laboratory (ANL), a free-electron laser (FEL) based on the self-amplified spontaneous emission (SASE) process is nearing completion. Recently, an RF photoinjector gun system was made available to the APS by Brookhaven National Laboratory/Accelerator Test Facility (BNL/ATF). It will be used to provide the high-brightness, low-emittance, and low-energy spread electron beam required by the SASE FEL theory. A Nd:Glass laser system, capable of producing a maximum of 500 /spl mu/J of UV in a 1-10 ps pulse at up to a 10-Hz repetition rate, serves as the photoinjector's drive laser. Here, the design, commissioning, and integration of this gun with the APS are discussed.

In the study presented here, air flows were homogeneously seeded with NO/sub 2/ at low concentrations (few hundreds of ppm). By photodissociation of NO/sub 2/ at 308 nm (XeCl excimer laser at approx. 200 MW/cm/sup 2/) NO was formed, which was imaged by LIF at various delays after its generation. Whereas several other tracers disappear rather quickly due to energy transfer processes (from vibrationally excited molecules) or high reactivity (in the case of OH/sup 4/), due to small reaction cross-sections, NO is stable on the timescale of interest. Therefore, using NO, the movements of volume elements can be imaged at longer times and larger distances compared to other techniques. Furthermore, with the molecular weight of NO being between that of N/sub 2/ and O/sub 2/ and diffusion coefficients close to nitrogen, NO turns out to be a good choice to characterize the small-scale behavior of turbulent air flows.

A magnet measurement facility for semi-automatic measurement control and real-time data analysis has been developed to measure more than 1000 magnets for the Advance Photon Source (APS). One dipole and three rotating coil measurement systems and corresponding probe coils are described. >

The pre-production quadrupole magnet for the Advanced Photon Source (APS) storage ring has been tested. The multipole coefficients meet the tolerance requirements. The field-gradient integrals are measured relative to a reference quadrupole. By using two laser beam units, the magnetic and geometrical axes of the magnet are aligned within 0.2 mrad. The dependencies of the sextupole coefficient and the magnetic center on the excitation current are corrected by shunting the magnetic potentials of the top and bottom yokes. >

The magnetic measurement data of the first 34 of the required 68 production magnets for the injector synchrotron are summarized. The magnetic measurement method of the field strength and field shape relative to a reference magnet is described. The standard deviation of the integrated field strength for the 34 magnets is 3.3/spl times/10/sup -4/ and the variation of the integrated field with transverse displacement of /spl plusmn/25 mm is less than 2.5/spl times/10/sup -4/. >

Sources of transverse beam motion in the APS storage ring have been investigated for ground-motion- and water-system-induced vibrations of the magnet and vacuum systems, and for power supply ripple. The displacement of magnets in a bandwidth of 4-30 Hz have been reduced significantly by inserting viscoelastic damping pads between the girder supports and pedestals, and by welding the magnet cooling headers to the ceiling of the storage ring tunnel. Current ripple on magnet power supplies was identified as a source of horizontal beam motion. Beam motion was measured without the closed-orbit feedback system activated. At /spl beta//sub x/=15.4 m and /spl beta//sub y/=10.4 m the rms beam motion in the 0.02-30 Hz band was 22.7 /spl mu/m and 6.3 /spl mu/m in the horizontal and vertical planes, respectively. A few narrow-band structures of the horizontal beam motion spectrum in the 1-4 Hz band have to be investigated further to identify the sources.

Permanent magnet blocks of NdFeB have a relatively high maximum energy product. Because of its relatively low Curie temperature, however, NdFeB has a large temperature coefficient for its residual induction. The temperature coefficients of the relative magnetic fields (/spl Delta/B/B)//spl Delta/T in the air gap of NdFeB dipole magnets were reduced from -1.1/spl times/10/sup -3///spl deg/C to less than 2/spl times/10/sup -5///spl deg/C under operating temperatures of /spl plusmn/6/spl deg/C. This was achieved passively by using 1.25-mm-thick strips of 30%-Ni-Fe alloy as flux shunts for the NdFeB blocks. The magnets with soft-steel poles and flux-return yokes were assembled and measured in a temperature-controlled environment.

A method for measuring the relative surface resistivity and quality of conductive coatings on ceramic vacuum chambers was developed. This method is unique in that it allows one to test the coating even after the ceramic chamber is installed in the accelerator and under vacuum; furthermore, the measurement provides a localized surface reading of the coating conductance. The method uses a magnetic probe of wire wound on a ferrite and an LCR meter. The probe is calibrated using the measured DC end-to-end resistance of the tube under test and by comparison to a high quality test surface. The measurement method has also been verified by comparison to high frequency impedance measurements. A detailed description, results, and sensitivity of the technique are given here.

Local feedback for the APS storage ring uses local bumps to control the position and angle of the positron beam through each X-ray source point. Induced eddy currents in the aluminum vacuum chamber dominate the AC characteristics of the corrector magnetic fields. Small differences in the geometries at each magnet location change the eddy current effects and result in bump closure errors which must be reduced in order to minimize the coupling between each of the many local loops and the global control loop. By a combination of flux-damping coils, flux-shielding copper sheets, and a set of steel laminations for end-flux clamping, the differences of the eddy current effects between two corrector magnets were reduced from 0.18 Gm/A to 0.035 Gm/A in the frequency span of 0.1-100 Hz.

The statistics of the measured magnetic data of 80 dipole, 400 quadrupole, and 280 sextupole magnets of conventional resistive designs for the APS storage ring is summarized. In order to accommodate the vacuum chamber, the curved dipole has a C-type cross section and the quadrupole and sextupole cross sections have 180* and 120* symmetries, respectively. The data statistics include the integrated main fields, multipole coefficients, magnetic and mechanical axes, and roll angles of the main fields. The average and rms values of the measured magnet data meet the storage ring requirements.

A new direct-drive, out-of-vacuum synchrotron injection septum magnet has been built at the Advanced Photon Source to replace the existing eddy-current-shielded, in-vacuum magnet. We describe the design, assembly, construction, and measurements.

The conductive coating on the kicker ceramic vacuum chambers was found to affect the pulse width, amplitude, and phase of the kicker magnetic field. Differences in the surface resistivity between the chambers caused each kicker magnetic field to have a slightly different pulse shape, resulting in unwanted betatron oscillations of the stored beam during injection into the storage ring at the Advanced Photon Source. Orbit oscillations were measured with the present kicker mismatch and optimized for a closed bump. Tracking simulations of the closed bump provided an idealized tolerance for the variation of chamber surface resistivity consistent with minimizing the betatron oscillations. The pulse shape of the ticker current Was dependent on the current level and the coupling between the kicker magnet and chamber. Local magnetic fields in two prototype kicker chambers were measured and compared with eddy current calculations.