July 29, 2013 | Optimising design of multitasking applications on multiprocessor architectures
With the emergence of multiprocessor systems and distributed architectures on the consumer market (multimedia and automotive), optimising embedded architectures has become a major challenge to ensure operating performance.
CEA LIST tackles this from several angles. If the systems can be built on architectures integrating specific hardware elements (computing accelerators), the aim is to research which combinations of computing architecture / processing algorithms optimise computing time, performance and cost (particularly energy costs). For this purpose, CEA LIST has developed a parallel architecture evaluation and programming environment as part of the Carnot COTS project, which enables users to optimise their choice of programming style and target architecture for a given application.
In a less “free” context (constraints on the hardware architecture, compliance with a specific standard) or if the functions to be performed are not amenable to parallel data processing (video, signal), the problem has to be addressed in a more general way, as a set of functions to be performed over time on distributed computing resources. CEA LIST has thus developed an original approach, combining exact and heuristic formal methods by approximation – to meet this general requirement – with the analysis software Qompass.
Optimising resources involves three levels of architecture: applications (what users want and see), processors (hardware computing resources) and tasks (normal concept aimed at creating virtual subdivisions of processes on the same resource).
The design of the final system is then based on two allocation levels and one validation action:
- placement associates functions with tasks (level of parallelism required),
- partitioning allocates available computing resources to tasks,
- scheduling calculates an implementation plan (typically allocation of priorities to tasks) to guarantee the quality of service expected (typically adherence to deadlines).
To deal with these problems in a general way, CEA LIST, in collaboration with two researchers and international experts (Prof. Marco di Natale and Haibo Zeng), has developed a mixed approach combining formal and heuristic calculations. By scaling up to hundreds of functions deployed on about ten processors, this general solution can be regarded as innovative compared with exact solutions, which are quickly swamped by the effect of combinatorics as soon as distribution is over several processors.
This research, initiated as part of the European INTERESTED project (2008-2011), particularly addresses the competitive challenges facing the automotive industry by optimising embedded resources, thereby guaranteeing application performance and real-time behaviour.
Parallel studies are also underway to apply and integrate these results with the AUTOSAR standard. They could then be incorporated into an optimisation module associated with the AUTOSAR design tools developed by various software publishers in the field, such as Esterel Technologies.
Pr. Marco di Natale – Scuola Superiore Santa Anna & Berkeley University
Haibo Zeng - Mac Gill University
The FP7 INTERESTED project – INTERoperable Embedded Systems Toolchain for Enhanced rapid Design – brought together a consortium of leading-edge European embedded system tool vendors (AbsInt Angewandte Informatik (Germany), Atego (UK), CEA (France), Esterel Technologies (France), UNIS (Czech Republic), Evidence (Italy), Symtavision (Germany), Sysgo (Germany) and TTTech Computertechnik (Austria)), as well as major European tool users (Airbus (France), Magneti Marelli Powertrain (Italy), Siemens Mobility Division, Rail Automation (Germany) and Thales (France)).
July 23, 2013 | Fullbody Exoskeleton at SIGGRAPH 2013
EMY (ENHANCING MOBILITY) | FULL BODY EXOSKELETON
Developed entirely in the Ile-de-France region by CEA LIST, EMY (Enhancing MobilitY) is a full-body exoskeleton designed to help quadriplegic people regain mobility.
Packed full with innovative technology, EMY is the fruit of 10 years of research by CEA LIST's interactive robotics unit and draws on extensive expertise in force control actuation, as well as the associated innovative robotic architectures.
EMY's architecture features four limbs: two legs with three degrees of freedom each, and two ABLE anthropomorphic arms with seven degrees of freedom each. To enable control by a quadriplegic patient, the exoskeleton will ultimately be controlled via a brain-computer interface - or BCI - called WIMAGINE®, developed by CEA LETI at Clinatec. The interface between the BCI and the exoskeleton will be ensured by a physics-simulation engine called XDE, also developed by CEA LIST. XDE is a haptic tool that simulates movement, contact and friction between virtual objects. This simulation layer allows EMY to be controlled at different levels of complexity, from the control of simple joint movements to abstract tasks coordinating the use of several limbs. As of 2014, it will also be capable of keeping the machine properly balanced.
The exoskeleton can also be used to control virtual reality physics simulation for industrial applications associated with “digital factory” design. These applications are being rapidly developed to reduce production set-up costs.
Thus, ABLE arms can not only control the movements of simulated objects, but also feed back the contact forces between them. Each of EMY's seven joints is equipped with a screw-cable system that exactly reproduces contact forces along the entire arm, doing away with the need for a force sensor (this function is fulfilled by the motor). Such outstanding performance is also due to the unique, lightweight, streamlined architecture which fits the human arm perfectly, without closing it in.
The ABLE exoskeleton can thus perform realistic simulations of manual tasks involving contacts distributed along the entire length of the arm, for example when fitting parts inside “crowded” structures. It can also be used as an operator workstation assistant to make work less strenuous and reduce the risk of repetitive strain injuries. It does this by evenly distributing the effort when operators are holding tools and balancing part of their arm.
July 22, 2013 | Awards for CEA LIST at the national competition for new innovative technology companies
On Tuesday 2 July 2013, the French Research Ministry presented the awards at the 15th National competition for new innovative technology companies. Eight start-up projects submitted by CEA researchers won accolades, representing a variety of industrial sectors ranging from medical diagnostics and energy to system security. The 2013 results demonstrate the dynamism of the CEA and its research team in the areas of innovation, technology transfer and entrepreneurship.
In the “start-up/development” category, two projects based on CEA LIST technology won awards:
PRONE Systems designs, develops and markets “electronic noses” capable of learning and recognising odours. These work in a similar way to the human nose: an array of sensors based on nanodiamonds – a field in which CEA LIST has acquired a great deal of expertise – calculates the olfactory signature of an odour during the learning phase and identifies it during the recognition phase. Each time an odour is characteristic of the state of an object or a process of change, PRONE Systems proposes a suitable way of detecting and identifying the formation of this odour or any variations in it. There are a whole host of practical applications. Agri-food, air quality monitoring and security (explosives and toxic gases) are the first sectors to express a strong interest. In fact, the particularly low detection thresholds (< 10 ppb) combined with the selectivity and robustness of the products provide an answer to many problems hitherto unsolved.
The TrustInSoft project ensures software is safe and secure, whether it is in a mobile phone, a games console, an industrial device or an information system. The project is based on Frama-C technology, already deployed in aeronautical and nuclear fields. The aim is to develop solutions to eliminate any security breaches that may be exploited by hackers. The team’s ambition is to ensure that all software has the same high level of security as that used in the aeronautical and nuclear sectors. Devices like telephones, smart electric meters and electricity distribution systems will thus benefit from all this technology has to offer.
Find out more:
- Press release by the Ministry of Higher Education and Research (in French)
- List of award winners (in French)
July 5, 2013 | Improving iodine-125 calibration for more effective brachytherapy
CEA LIST’s Laboratoire National Henri Becquerel has developed a new method to more accurately determine the iodine-125 dosages delivered to cancer patients during brachytherapy.
Brachytherapy, which entails placing radiation sources like iodine-125 as close as possible to the tumor being treated, is currently one of the most commonly-used techniques to treat ophthalmic and prostate cancers. One of the key factors in treatment efficacy is knowing the exact dosage delivered by each radiation source placed in the patient’s body. CEA LIST’s Laboratoire National Henri Becquerel has developed a new method using an air-wall ionization chamber to better calibrate the radiation sources.
Measuring photon spectra
To determine the measurements of absorbed-dose-to-water at specific geometries, a number of calculated correction factors must be applied. Therefore, the energy of the X and γ photons emitted by the source must be accurately known. This is particularly true for low-energy radiation sources like iodine-125 (maximum energy 35 keV). In this case, small variations in energy result in large variations in photon-matter interaction probabilities. Finally, the confinement of the radioactive iodine in seeds generates other photons resulting from fluorescence and diffusion.
Knowing the detector’s response
To effectively measure the energy of the photons, the lab used a high-purity germanium (HPGe) detector. And, to take into account the phenomena that occur inside the detector (distorting the spectra), the researchers developed a correction method. They studied the detector’s “response” with a source of photons of the same energy (and whose value is accurately known). The experiment was repeated, varying the energy over a range covering the detector’s scope of use. This required using specific radiation sources: the LNHB Solex source (6 keV to 17 keV) and the ESRF’s ID17 source (30 keV to 60 keV). Based on the data obtained, the researchers developed mathematical models to calculate the “spectra” of energy obtained with the HPGe detector.
Reconstructing the actual spectrum of iodine seeds
Thanks to the new technique, it is now possible to match a spectrum of photons (such as that emitted by an iodine-125 seed) to the spectrum as it would be “seen” by the HPGe detector. By comparing the spectra “seen” by the detector (obtained by the measurement and reconstruction process), it is possible to extract the spectra actually emitted by the iodine-125 source.
June 10, 2013 | DESKOLO: Reducing the power consumption of computer networks
Completed in late 2012, the Deskolo project has led to the development of a complete infrastructure for managing the power consumption of a computer network, without the need for external sensors.
The proposed solution implements a statistical model-based learning algorithm that calculates power consumption as a linear combination of various variables characteristic of a computer's operating status at a given moment (processor activity, number of applications executed, memory usage, etc.).
Combined with a custom version of the Mandriva standalone operating system, a generic script determines the reference power consumption profile of a given category of computer systems and exports it to other computers in the network.
A dedicated interface allows individual users to view, store and modify their power consumption parameters so as to ensure minimum consumption. DESKOLO then interacts with the network server and sends it each user's parameters so as to schedule the operating times of each computer, thereby reducing the overall power consumption of the entire network.
Funding: European Regional Development Fund (ERDF) | Ile de France region
Duration: 30 months