Abstract

International audience; This chapter summarizes the one-dimensional modeling of transients in a pipeline, commonly used for detection and location of faults (such as leaks and obstructions) by means of model-based methods. The modeling starts with the discretization via finite-difference method of classical water hammer equations. The result of such a discretization is a system of ordinary differential equations, which is considered together with boundary conditions that represent faults and pipeline accessories. Some illustrative results are finally given based on a test bench.

Abstract

International audience; This paper presents an approach which has for objective to model new tools allowing to help the controllers to assume the incessant increase of the air traffic (actual version of the platform AMANDA V2), as well as help them in the negotiation phase and cooperation with their counterparts of adjacent sector (objectives of the new version of AMANDA). Help them in furnishing some tools able to quickly share information, and to maintain good common situation awareness. An approach is proposed, it is divided in three main phases. A first phase which consists of understand and to model the decision-making process of controllers. The second phase introduces a multiple criteria decision-making methodology. This Methodology has for objective to understand in more details the activities of controllers and the cases of cooperation with adjacent sectors. Finally, the last phase is the operational level of the approach, and consists of an application of repertory grid methodology in order to guide the interviews with the different participants of the study. This will allow realizing the knowledge acquisition, keeping in mind objective to develop new tools. To conclude this paper, the last part presents an example of application of this approach and the first results.

Abstract

The European research project DESERVE (DEvelopment platform for Safe and Efficient dRiVE, 2012-2015) had the aim of designing and developing a platform tool to cope with the continuously increasing complexity and the simultaneous need to reduce cost for future embedded Advanced Driver Assistance Systems (ADAS). For this purpose, the DESERVE platform profits from cross-domain software reuse, standardization of automotive software component interfaces, and easy but safety-compliant integration of heterogeneous modules. This enables the development of a new generation of ADAS applications, which challengingly combine different functions, sensors, actuators, hardware platforms, and Human Machine Interfaces (HMI). This book presents the different results of the DESERVE project concerning the ADAS development platform, test case functions, and validation and evaluation of different approaches. The reader is invited to substantiate the content of this book with the deliverables published during the DESERVE project. Technical topics discussed in this book include: Modern ADAS development platforms; Design space exploration; Driving modelling; Video-based and Radar-based ADAS functions; HMI for ADAS; Vehicle-hardware-in-the-loop validation systems Note: EUR 6,000 BPC fee funded by the EC FP7 Post-Grant Open Access Pilo Document type: Part of book or chapter of book

Abstract

The two first CEAS (Council of European Aerospace Societies) Specialist Conferences on Guidance, Navigation and Control (CEAS EuroGNC) were held in Munich, Germany in 2011 and in Delft, The Netherlands in 2013. ONERA The French Aerospace Lab, ISAE (Institut Superieur de l’Aeronautique et de l’Espace) and ENAC (Ecole Nationale de l’Aviation Civile) accepted the challenge of jointly organizing the 3rd edition. The conference aims at promoting new advances in aerospace GNC theory and technologies for enhancing safety, survivability, efficiency, performance, autonomy and intelligence of aerospace systems. It represents a unique forum for communication and information exchange between specialists in the fields of GNC systems design and operation, including air traffic management. This book contains the forty best papers and gives an interesting snapshot of the latest advances over the following topics: l Control theory, analysis, and design l Novel navigation, estimation, and tracking methods l Aircraft, spacecraft, missile and UAV guidance, navigation, and control l Flight testing and experimental results l Intelligent control in aerospace applications l Aerospace robotics and unmanned/autonomous systems l Sensor systems for guidance, navigation and control l Guidance, navigation, and control concepts in air traffic control systems For the 3rd CEAS Specialist Conference on Guidance, Navigation and Control the International Program Committee conducted a formal review process. Each paper was reviewed in compliance with standard journal practice by at least two independent and anonymous reviewers. The papers published in this book were selected from the conference proceedings based on the results and recommendations from the reviewers. Document type: Part of book or chapter of book

Abstract

International audience; Air traffic management (ATM) is an endless source of challenging optimization problems. Before discussing applications of metaheuristics to these problems, let us describe an ATM system in a few words, so that readers who are not familiar with such systems can understand the problems being addressed in this chapter. Between the moment passengers board an aircraft and the moment they arrive at their destination, a flight goes through several phases: push back at the gate, taxiing between the gate and the runway threshold, takeoff and initial climb following a Standard instrument departure (SID) procedure, cruise, final descent following standard terminal arrival route (STAR), landing on the runway, and taxiing to the gate. During each phase, the flight is handled by several air traffic control organizations: airport ground control, approach and terminal control, en-route control. These control organizations provide services that ensure safe and efficient conduct of flights, from departure to arrival.

Abstract

This manuscript aims to describe my career in the transportation domain, putting in evidence my contributions in different levels, as for example thesis advising, teaching, research animation and coordination, projects construction and participation in expert committees, among others, besides my scientific research itself. The goal, besides the HDR diploma itself, is to show very clearly, including to myself, this 'pack' of contributions in order to look for better contributions to the transportation and control communities or to other communities in the future, and also which research directions I will define to work on in the following. I obtained my PhD degree in the Laboratoire des Signaux et Systèmes - L2S 1 in collaboration with MIT, in 2001, having worked in a purely theoretical automatic control topic scarcely known in the literature - the adaptive control of systems with nonlinear parameterization problem. Arriving in 2002 as a permanent researcher to the former LCPC (Laboratoire Central des Ponts et haussées), now called IFSTTAR (Institut Français des Sciences et Technologies des Transports, de l'Aménagement et des Réseaux), I have been faced to real problems to solve in practice, and faced to the new community of transportation, with a completely different philosophy of work. I have nowadays this double vision - of the very applied transportation domain with concrete problems to be solved that touch the citizen every day, and the vision of a very rich high-level theoretical research in automatic control with powerful tools to solve the real problems, or on the other hand, with control problems that appear because of the need for new tools to solve the real problems. I consider this as an important characteristic for my future contributions. Besides the knowledge in Transportation itself, my eleven years of career in IFSTTAR gave me as well the following new features : 1. From the individual research, I have learned also how to coordinate work (in projects for example, as in the PReVAL sub-project of the European PReVENT project, in which I co-leaded one workpackage, or for research teams, as the control team of LIVIC, coordinated by myself from 2006 to 2009). I have also learned how to animate research (by coordinating research working groups or organizing scientific events and workshops - see for example the working group RSEI and the related scientific event below that I have organized in June 2012) and how to advise students. 2. Besides the double vision I have described above, the experience gave me also the acquisition of a quite multidisciplinary view of the problems in the domain. Firstly, arriving in LIVIC, in the frame of the French consortium ARCOS, I have worked for two years in close cooperation with experts in cognitive sciences (the PsyCoTech group from IRCCyN, Nantes) on designing driving assistance systems to a human driver. After this work, I have continued the collaboration with experts in human sciences within the PReVAL subproject of PReVENT on driving assistance systems evaluation and within the French ANR PARTAGE project, that I have constructed together with the PsyCoTec team of IRCCyN and leaded the IFSTTAR partner for one year. In a dition, through my participation in PReVENT at dirent levels (in two meetings of the Core Group, in PReVAL by co-leading the workpackage 3 on Technical Evaluation of ADAS - ADAS is the shortcut for Advanced Driving Assistance Systems - and in the SAFELANE subproject), I have learned many different aspects of ITS systems. I consider this as an add-on value for my 'pack of knowledge'. 3. What I call "from abstractions to real problems : coming back and forth to solve these real problems" has been matured in my mind, and I am very grateful to my students, with whom I have learned and that helped me in this maturing process. By this sentence, I mean, with a problem to solve in hands, and after building an abstraction, or a simplified view of the problem, and the design of a solution, how to apply it, and to come back again to the theory to change it and to come back to the practice, and so on. This is exactly one of the pillars of the NoE HYCON2, for making interact the theory with the application domains. 4. Considering a problem inserted into the societal context, or inserted within its related context, has been another maturing for myself that I consider very important, notably in the transportation domain, that represents a very complex context containing many different parameters, scenarios and objectives and in addition all the uncertainties linked to the human behavior. I think that it is very important to have a very large view of the context in which the specific problem we are treating is placed. Without this, one cannot say in most of the cases, from my point of view, that the problem is solved. This point will be discussed in Chapter 9.5. 5. Another point that I consider important and where I have been contributing recently is the road mapping work. The acquisition of the multidisciplinary knowledge and a larger view of the domain that I have mentioned in the preceding items, together with my theoretical knowledge in automatic control, allowed myself to start contributing to theroad mapping work in Transportation (through my participation in the imobility forum, in HYCON2 and the in the support action T-Area-SoS on Systems of Systems - all these actions to give advice to the European Commission on the priority areas to be considered in the new Calls, notably in the frame of the H2020 program). I had also the pleasure of opening again books and thesis that I had studied in my PhD work, this time now for advising students in the frame of other very different problems. The very beautiful thesis of Mikael Johansson, Lund University, on piecewise linear systems stability theory is an example. My previous study on switched systems, and the implication of switched Lyapunov functions on stability helped me also in advising my students (Post-Docs, PhD, and M.Sc. students), this time for real applications, with very interesting results blooming up from their work. I realize also that the experience that I have described in the five items above must be put in favor of students since this kind of knowledge cannot be found in the books. Concluding, in these last eleven years, from 2002 to 2013, I could bring to the scientic community and to my students a set of contributions of different kinds. I will try to make clear these contributions for the reader in the next two chapters (written in English and in French). This document is organized in the following way : Part II contains my complete curriculum vitae (in french) where all these contributions will be described in detail. Part III contains then the scientific contributions of the manuscript. What I aim in this chapter is to describe, but further, to analyze them with a distanced look and providing a critical view, announcing perspectives, and placing and discussing the obtained results in the societal context. This is in straight relation with item 4 above. Also, I prefer to adopt, as far as possible, a form comprehensible to the non-automatic control expert, with, as far as possible as well, qualitative explanations and then appropriated references containing the theorems and the definitions corresponding to the qualitative explanations will be provided. In the case it is necessary, they are provided within the text. The Part III is structured in the following chapters. Chapter 8 contains an overview of the global transportation scenario with the associated challenges and a description of the driving assistance systems context. Chapter 9 contains my scientific contributions. These include my research results, my contributions in students advising, in the coordination of research groups, and the collaborative works. It is structured in 3 sections : Section 9.1 introduces what will be the greed for a part of the main contributions, that are described in Sections 9.2 and 9.3. Section 9.1 is also dedicated to showing to the reader how theory and abstractions can be very important for solving real problems. Chapter 9.4 describes other contributions that are the result of collaborative works. A discussion from a multidisciplinary view is provided in Chapter 9.5 based on a survey paper of myself. Chapter 10 will be finally dedicated to the perspectives and the general conclusions. Then last Part contains as annexes a selection of the publications that I consider the most illustrative of my contributions described in Chapter 9. Finally, since the described work is in the intersection of two communities - the transportation and the control theory communities - I decided to write a part of the document dedicated to the non control experts readers. This is Part VI of the document whose aim is to provide some fundamental notions on control theory in a very simple qualitative description whose understanding will help the different readers to understand the contributions.

Abstract

Co-encadrement de la thèse : Bogdan Stanciulescu; Pedestrian Detection and Traffic Sign Recognition (TSR) are important components of an Advanced Driver Assistance System (ADAS). This thesis presents two methods for eliminating false alarms in pedestrian detection applications and a novel three stage approach for TSR. Our TSR approch consists of a color segmentation, a shape detection and a content classification phase. The red color enhancement is improved by using an adaptive threshold. The performance of the K-d tree is augmented by introducing a spatial weighting. The Random Forests yield a classification accuracy of 97% on the German Traffic Sign Recognition Benchmark. Moreover, the processing and memory requirements are reduced by employing a feature space reduction. The classifiers attain an equally high classification rate using only a fraction of the feature dimension, selected using the Random Forest or Fisher's Criterion. This technique is also validated on two different multiclass benchmarks: ETH80 and Caltech 101. Further, in a static camera video surveillance application, the immobile false positives, such as trees and poles, are eliminated using the correlation measure over several frames. The recurring false alarms in the pedestrian detection in the scope of an embedded ADAS application are removed using a complementary tree filter.; La détection de piétons et la reconnaissance des panneaux routiers sont des fonctions importantes des systèmes d'aide à la conduite (anglais : Advanced Driver Assistance System - ADAS). Une nouvelle approche pour la reconnaissance des panneaux et deux méthodes d'élimination de fausses alarmes dans des applications de détection de piétons sont présentées dans cette thèse. Notre approche de reconnaissance de panneaux consiste en trois phases: une segmentation de couleurs, une détection de formes et une classification du contenu. Le color enhancement des régions rouges est amélioré en introduisant un seuil adaptatif. Dans la phase de classification, la performance du K-d tree est augmentée en utilisant un poids spatial. Les Random Forests obtiennent un taux de classification de 97% sur le benchmark allemand de la reconnaissance des panneaux routiers (German Traffic Sign Recognition Benchmark). Les besoins en mémoire et calcul sont réduits en employant une réduction de la dimension des caractéristiques. Les classifieurs atteignent un taux de classification aussi haut qu'avec une fraction de la dimension des caractéristiques, selectionée en utilisant des Random Forests ou Fisher's Crtierion. Cette technique est validée sur deux benchmarks d'images multiclasses : ETH80 et Caltech 101. Dans une application de vidéo surveillance avec des caméras statiques, les fausses alarmes des objets fixes, comme les arbres et les lampadaires, sont éliminées avec la corrélation sur plusieurs trames. Les fausses alarmes récurrentes sont supprimées par un filtre complémentaire en forme d'arbre.

Abstract

International audience; TCAS resolution advisories generation process during level-off situation is overviewed and illustrated on a real situation captured through mode S radar with RA downlink service. It underlines that one common type of resolution advisories is that which is issued when aircraft are expected to level-off 1000 feet apart, and are converging at the same time. These resolution advisories, often subsequently classed as ?operationally unnecessary? from an air traffic control standpoint, can be perceived as disturbing by controllers and by pilots. This issue has been tackled in this paper by proposing a method to prevent RA triggering during level off situation. This method is based on the setting of the altitude at which the autopilot switches towards the altitude capture mode and on an intrinsic parameter of the autopilot altitude capture mode, namely its natural frequency. It is shown that applying the proposed method to the illustrative example prevents the issuance of resolution advisories when aircraft are expected to level-off 1000 feet apart. The method which is presented is one of the simplest options, and allows altitude capture without triggering resolution advisories for vertical speeds up to 3000 feet/min. Nevertheless, the time needed by the aircraft to capture the cleared flight level is increased. Consequently, future work will focus on improving the time needed to capture the cleared flight level under the constraint to keep positive the distance between the autopilot altitude capture dynamic and the boundary of the RA zone.

Abstract

ording to a 2013 AUVSI report, delays in integrating Unmanned Aerial Systems (UAS) into the National Airspace System (NAS) could cost more than $10 billions a year for the United States alone. Worldwide regulatory bodies are under pressure by the UAS industry to accelerate the regulation process, but safety remains their main objective. One condition for the safe introduction of UAS in the NAS is for them to be equipped with a collision avoidance system. Though the existing Airborne Collision Avoidance System II (ACAS II) could have been an option, the transformations of air traffic management engaged through NextGen (US) and SESAR (Europe) led to the definition of a new ACAS based on new logics, namely ACAS X. Its definition contains in particular two variations : ACAS Xa, for large aircraft, and ACAS Xu, for unmanned aircraft. As noted in a 2014 RTCA annual report, divide in technological knowledge between those experienced in ACAS II and those involved in the development of ACAS X is a concern. To help preventing this divide we believe it is essential to keep the community updated with the latest evolutions of the ACAS X standards. As work on Minimum Operational Performance Standards (MOPS) for ACAS Xu just started, it is of interest to know which parts of the MOPS are already decided, which remain flexible for the industries to make the difference and which are open research problems. Being a member of the ACAS X family, ACAS Xu lays on the same foundations as the well defined ACAS Xa standard. This work proposes an introduction to the ACAS Xa/Xu common basis, as it is unlikely to change, including the general architecture and Collision Avoidance (CA) logics. It is followed by a presentation of concepts specific to ACAS Xu such as the tailored threat logic, horizontal CA logic, CA coordination and automatic responses. For the flexible part, we believe it mainly concerns the surveillance sources. Instead of a precise standard, the regulation is likely to ask for requirements on the sensors capabilities. A state of the art of recent works allows proposing minimum sensor performances and focusing on an essential set of sensors. This work is concluded by presenting future challenges that need to be addressed to build a safe ACAS Xu baseline and to extend it to smaller and lower altitude UAS.

Abstract

6 pages; International audience; The paper proposes a control system with the brake and the steering for developing a driver assistance system. The purpose is to design a cruise control method to track the road geometry with a predefined velocity and guarantee the road stability of the vehicle simultaneously. An actuator selection method is developed in the control design, in which the actuator limits, energy requirements and vehicle operations are taken into consideration. The method is extended with a fault-tolerant feature based on a robust LPV method, into which the actuator selection procedure and the detected fault information are incorporated. The operation of the reconfigurable control system is illustrated through various vehicle manoeuvres.