Armée de l'air et de l'espace

The Android operating system is the most spread mobile platform in the world. Therefor attackers are producing an incredible number of malware applications for Android. Our aim is to detect Android's malware in order to protect the user. To do so really good results are obtained by dynamic analysis of software, but it requires complex environments. In order to achieve the same level of precision we analyze the machine code and investigate the frequencies of ngrams of opcodes in order to detect singular code blocks. This allow us to construct a database of infected code blocks. Then, because attacker may modify and organized differently the infected injected code in their new malware, we perform not only a semantic comparison of the tested software with the database of infected code blocks but also a structured comparison. To do such comparison we compute subgraph isomorphism. It allows us to characterize precisely if the tested software is a malware and if so in witch family it belongs. Our method is tested both on a laboratory database and a set of real data. It achieves an almost perfect detection rate.

There is no consensus in the literature on how to apply Lean Product Development (LPD) successfully. This paper presents a literature review aiming to fill this gap. The key success factors in and barriers to successful implementation of LPD were gathered and categorized by means of a thematic reading. The question of the context’s influence on these factors is also posed, and a proposition is made to prioritize the notion of configurations rather than frameworks. From an academic perspective, this research gives an overview of the LPD implementation’s key factors and opens up a discussion on the pertinence of the notion of framework in this field. On the practitioner side, this paper helps organizations to reflect on the main factors to consider when implementing LPD.

L’armée de l’air française est dotée de systèmes de drones de surveillance contrôlés par des équipages au sol. Un équipage est constitué d’un Pilote A Distance en charge du contrôle du vecteur et de la caméra. Il est secondé par un Opérateur Images en charge de l’analyse des données recueillies. L’ensemble de la mission est géré par le Coordinateur Tactique en charge de communiquer avec les quartiers généraux. Une connaissance précise des compétences mises en œuvre à chaque poste est indispensable pour réfléchir à la mise en place d’une procédure efficace de sélection des candidats. La présente étude a analysé les profils optimaux de chaque poste en termes d’aptitudes cruciales sollicitées au plus haut niveau pour développer les compétences nécessaires à ces postes. Les Pilotes A distances présentent un profil complexe sollicitant des aptitudes variées mais majoritairement issues de la sphère perceptivo-cognitive. Le profil optimal des Coordinateurs Tactiques se caractérise par des niveaux élevés d’aptitudes dans les sphères inter- et intra-personnelles autant que dans la sphère perceptivo-cognitive. Quant aux Opérateurs Images, leur profil optimal est très spécialisé et centré sur la sphère perceptivo-cognitive. Les résultats sont interprétés en rapport avec la littérature scientifique quant à l’évolution des situations de travail ainsi que des caractéristiques de la procédure de sélection.

The Android operating system is the most spread mobile platform in the world. Therefor attackers are producing an incredible number of malware applications for Android. Our aim is to detect Android's malware in order to protect the user. To do so really good results are obtained by dynamic analysis of software, but it requires complex environments. In order to achieve the same level of precision we analyze the machine code and investigate the frequencies of ngrams of opcodes in order to detect singular code blocks. This allow us to construct a database of infected code blocks. Then, because attacker may modify and organized differently the infected injected code in their new malware, we perform not only a semantic comparison of the tested software with the database of infected code blocks but also a structured comparison. To do such comparison we compute subgraph isomorphism. It allows us to characterize precisely if the tested software is a malware and if so in witch family it belongs. Our method is tested both on a laboratory database and a set of real data. It achieves an almost perfect detection rate.

International audience

International audience

International audience

The Android operating system is the most spread mobile platform in the world. Therefor attackers are producing an incredible number of malware applications for Android. Our aim is to detect Android's malware in order to protect the user. To do so really good results are obtained by dynamic analysis of software, but it requires complex environments. In order to achieve the same level of precision we analyze the machine code and investigate the frequencies of ngrams of opcodes in order to detect singular code blocks. This allow us to construct a database of infected code blocks. Then, because attacker may modify and organized differently the infected injected code in their new malware, we perform not only a semantic comparison of the tested software with the database of infected code blocks but also a structured comparison. To do such comparison we compute subgraph isomorphism. It allows us to characterize precisely if the tested software is a malware and if so in witch family it belongs. Our method is tested both on a laboratory database and a set of real data. It achieves an almost perfect detection rate.

International audience

International audience

This present work is deliberately placed in the context capable of defining the requirements expressed by machine decision-making calculations. The informational nature of a decision requires abandoning any invariant preserving the structure but on the contrary switching into total chaos, a necessary and sufficient condition for exploiting the symmetries allowing the calculation to converge. Decision arithmetic is the best way to precisely define the nature of these symmetries.

International audience

This case study investigated the experience of a military helicopter pilot trainee during formation flight. Formation flight is a technique used in military operations, consisting of maneuvering safely around a Lead helicopter by controlling the rate and direction of motion to avoid collisions. Using the Course-of-Action framework, we described the pilot's cognitive activity during formation-flight maneuvers (join-up patterns) in a practice session from his own perspective to provide insights into his lived experience. Focus was placed on the situational elements that were meaningful to the pilot at a given moment (i.e., Representamen), and how these meaningful situational elements were guided by his situated concerns (i.e., Involvement). Data were collected in two steps: (1) collection of activity traces during formation flight training and (2) self-confrontation interviews using these activity traces in which the pilot was invited to relive his experience and describe his activity. The results indicated five typical representamen and four typical involvements, and revealed eight different associations between these typical representamen and typical involvements over the course of the maneuvers. The discussion addresses how the description of these associations provides a better understanding of the pilot's activity during formation maneuvers and proposes possible extensions of this study.

We describe the development of a prototype to support the consideration of multiple objectives in vessel routing. Our aim is to offer shipping companies alternative routes that take into account various criteria, including costs, route safety and CO<inf>2</inf> emissions, so that they can make informed decisions about their itinerary. This work, conducted in collaboration with the start-up MaritimeAPI, is highly topical: Maritime transport represents about 3 percent of CO<inf>2</inf> emissions in the world. Today, a large number of organisations and countries wish to reduce CO<inf>2</inf> emissions in order to slow down climate change. Our proposed decision tool wraps around the Application Programming Interface (API) of the company Seametrix. This API generates shortest routes for a user-defined set of parameters, governing access to areas affected by piracy, emission restrictions or the price of passage. We use a combination of machine learning and mathematical calculations, operating on the API outputs, to obtain seven distinct criteria, deemed relevant by domain experts. Using multiobjective optimization over the API parameters, we then illustrate how the existing API can be employed to explore the trade-offs between these criteria. Finally, we discuss the use of preference elicitation to support practitioners in the selection of a final solution, and contrast the results from this a posteriori approach to those obtained using a priori preference elicitation.

International audience

Abstract: To find out whether it is possible to train pilots and controllers to adequately trust their partners, and thus modify their behavior and increase their performance, this study experimentally tested whether individuals’ disposition to trust is influenced by their social learning. Objective measurements related to the task to perform, collected over two successive phases, were used for two different groups of aviation staff. The results highlight a link between swift trust, supervision strategies, and time management and point to the social nature of the disposition to trust. This opens the way to a new experimental approach to the study of behaviors induced by interpersonal trust and to new pedagogical levers for optimizing the activity of ephemeral teams through training.