Search

Complex systems have become a particular area of interest in the past years and the study of such systems, seen as a whole, has yielded many and varied approaches. One of the difficulties frequently encountered is that such systems display emergence, i.e. their global effect or behavior is greater than the sum of the behaviors of their agents and depends strongly on the interactions between these agents. Simulation is a very interesting approach for this study, since it allows focusing on the dynamic study of the system, thus pointing out these behaviors and interactions. Moreover, it is particularly interesting for aerospace systems, which are nowadays clearly studied as complex systems, since these systems are generally poor candidates for real experimentation, due to many diverse reasons (e.g. cost for space systems, criticality in terms of human life for air transport systems, etc.). Hence, they must be studied quite thoroughly before their full design or realization is possible and simulation plays an important part in this. This paper presents methods that can be used to study complex systems by simulation, as well as some techniques that take into account their specific nature, in particular distributed simulation, a paradigm quite intensively used at Onera. It also deals with the difficulties that may be encountered in setting such large simulations up and also addresses the problem of data in such a large simulation, from the handling to the exploitation., Journal Aerospace Lab, Issue 4, May 2012; ISSN: 2107-6596

The fourth issue of the journal Aerospace Lab is devoted to various techniques of information processing that manage complex interactions. The term "complex" is a bit overused but here it refers to the size of the problems that cannot be tackled using more straightforward approaches. The papers presented address a range of basic techniques. The first group addresses embedded information processing algorithms for control and perception, and also some issues about interaction between increasingly complicated systems and operators who have to manage them in real time. A second group of papers is devoted to engineering practices underlying the design work., Journal Aerospace Lab, Issue 4, May 2012; ISSN: 2107-6596

In this paper we propose and describe how we have built a tool that enables a user to interactively monitor and manage a many-core system like the 48-core experimental Singlechip Cloud Computer (SCC), which was created by Intel Labs targeting the many-core research community. We provide the user with a visual representation of the current state of the system on multiple levels of detail, such as chip, core and task. We allow the user to create, start, pause and migrate tasks across different cores. We also allow the user to easily adjust the voltage and frequency of the chip. However this tool can run on any PC with a screen and input devices, we have optimized the interface to run on a multi-touch device for the best ease of use.

Electronic sets operated on aircraft are usually summarized as “avionic architectures” (for "aviation electronic architecture"). Since the 70s, avionic architectures, composed of digital processing modules and communication buses, are supporting more and more avionic applications such as flight control, flight management, etc. Hence, avionic architectures have become a central component of an aircraft. They have to ensure a large variety of important requirements: safety, robustness to equipment failures, determinism, and real-time. In response to these requirements, aircraft manufacturers have proposed several solutions. This article has a twin objectives: firstly to survey the state of the art of existing avionic architectures, including the IMA (for Integrated Modular Avionic) architecture of the most recent aircraft; and secondly to discuss two challenges for the next generation of avionic architectures: reconfiguration capabilities, and integrating COTS processing equipment such as multi-core processors. We believe that these two challenges will be central to the next generation of IMA architectures (called IMA-2G for IMA 2d generation)., Journal Aerospace Lab, Issue 4, May 2012; ISSN: 2107-6596

CERTI is an HLA RTI developed since 1996 by ONERA, the French Aerospace Lab. The initial purpose of CERTI was to develop a home made RTI in order to: learn HLA usage and HLA RTI internals (e.g. time management), have total control over source code in order to use this particular RTI with specific modifications in several research projects (security mechanism, multi-resolution, high performance distributed simulation...). CERTI became open source in 2002: https://savannah.nongnu.org/projects/certi. Since then, Open Source CERTI project has had variable activity periods, mostly driven by research project needs and funds. CERTI development has started again since the end of 2006, with an increased interest from the open source user community. After a brief status survey of CERTI, this presentation will focus on the Open Source objectives of CERTI and explain why this is not a product but a project driven OSS initiative, pushed by a Public establishment like ONERA. We will further explain how open sourceness CERTI stimulates its development and the community itself and why every stakeholder benefits from this.

This paper presents some experiments and some results to enforce real time distributed simulations in accordance with the High Level Architecture (HLA). Simulations were run by using CERTI, an open source middleware, as the Run Time Infrastructure (RTI). Models were distributed over computers under various available versions of the 2.6 Linux kernel. Studies and experiments relied on a real case study. The chosen case study was the simulation of an "in formation" flight of observation satellites. This case study brings up some real applicative needs in real time distributed simulations and real configurations of simulators and models. Two simulations of "in formation" flight of satellites were studied. The study consisted in modeling the behaviour of the simulators and in running these models by using various kernel or middleware operating mechanisms and services. Time measurements were performed at each test giving some results on the ability of the simulation to meet its real time requirements.

We show that, when solving a linear system with an iterative method, it is necessary to measure the error in the space in which the residual lies. We present examples of linear systems which emanate from the finite element discretization of elliptic partial differential equations, and we show that, when we measure the residual in H−1(Ω), we obtain a true evaluation of the error in the solution, whereas the measure of the same residual with an algebraic norm can give misleading information about the convergence. We also state a theorem of functional compatibility that proves the existence of perturbations such that the approximate solution of a PDE is the exact solution of the same PDE perturbed.