Le Gland, François (1955....; chercheur en informatique)
Overview
Works:  39 works in 62 publications in 2 languages and 64 library holdings 

Roles:  Author, Other, Opponent, Thesis advisor 
Classifications:  TJ217.5, 500 
Publication Timeline
.
Most widely held works by
François Le Gland
Consistent parameter estimation for partially observed diffusions with small noise by
Matthew R James(
Book
)
5 editions published between 1989 and 1990 in English and held by 8 WorldCat member libraries worldwide
5 editions published between 1989 and 1990 in English and held by 8 WorldCat member libraries worldwide
Estimation de paramètres dans les processus stochastiques, en observation incomplète : application à un problème de radioastronomie by
François Le Gland(
)
3 editions published between 1981 and 2019 in French and held by 8 WorldCat member libraries worldwide
Présentation d'un problème d'estimation apparaissant en radioastronomie. calcul du rapport de vraisemblance et problème de filtrage non linéaire associé. Problèmes approches: semi discrétisation en temps; discrétisation globale. résultats numériques
3 editions published between 1981 and 2019 in French and held by 8 WorldCat member libraries worldwide
Présentation d'un problème d'estimation apparaissant en radioastronomie. calcul du rapport de vraisemblance et problème de filtrage non linéaire associé. Problèmes approches: semi discrétisation en temps; discrétisation globale. résultats numériques
Timediscretization of the Zakai equation for diffusion processes observed in correlated noise by
Patrick Florchinger(
Book
)
3 editions published in 1990 in English and held by 6 WorldCat member libraries worldwide
3 editions published in 1990 in English and held by 6 WorldCat member libraries worldwide
Asymptotics of the GLRT for the disorder problem in diffusion processes = Comportement asymptotique du test du rapport de
vraisemblance généralisé pour la détection by
Fabien Campillo(
Book
)
4 editions published in 1992 in English and held by 4 WorldCat member libraries worldwide
4 editions published in 1992 in English and held by 4 WorldCat member libraries worldwide
Navigation intégrée d'un engin sousmarin remorqué, filtrage nonlinéaire des systèmes sans bruit d'observation et, ou
mesures parfaites by
Marc Joannides(
Book
)
2 editions published in 1997 in French and held by 3 WorldCat member libraries worldwide
DANS LA PREMIERE PARTIE DE CETTE THESE, NOUS CONSIDERONS LE PROBLEME DE L'ESTIMATION DE LA TRAJECTOIRE D'UN ENGIN SOUSMARIN REMORQUE, EN SE BASANT SUR DEUX SOURCES D'INFORMATION DIFFERENTES: DES MESURES PRECISES D'ACCELERATION DE L'ENGIN (INS) ET DES MESURES DE POSITION DU NAVIRE DE SURFACE (GPS). LES MESURES INS RESTITUENT FIDELEMENT LES HAUTES FREQUENCES DU MOUVEMENT DE L'ENGIN, MAIS DERIVENT AVEC LE TEMPS. NOUS UTILISONS L'INFORMATION GPS, FIABLE A LONG TERME, POUR RECALER LE MOUVEMENT MOYEN DE L'ENGIN. NOUS INTRODUISONS UN MODELE NUMERIQUE DU SYSTEME CABLEENGIN POUR TRANSFERER L'INFORMATION DE POSITIONNEMENT DEPUIS LA SURFACE JUSQU'A L'ENGIN, ET NOUS PROPOSONS UN ESTIMATEUR HYBRIDE DE SA TRAJECTOIRE. NOUS NOUS INTERESSONS ENSUITE AU PROBLEME DE FILTRAGE DES PROCESSUS DE DIFFUSION, DANS LE CAS OU L'ON DISPOSE D'OBSERVATIONS NONBRUITEES, EN TEMPS DISCRET. CE PROBLEME EST SINGULIER CAR LA LOI CONDITIONNELLE EST SUPPORTEE A CHAQUE INSTANT PAR UN ENSEMBLE DE NIVEAU DE LA FONCTION D'OBSERVATION, QUI EST EN GENERAL DE MESURE NULLE (POUR LA MESURE DE LEBESGUE) DANS L'ESPACE D'ETAT. DANS LE CAS OU LA VALEUR OBSERVEE EST REGULIERE, NOUS OBTENONS UNE EXPRESSION EXPLICITE POUR LA DENSITE DE LA LOI CONDITIONNELLE, PAR RAPPORT A LA MESURE CANONIQUE SUR L'ENSEMBLE DE NIVEAU. CE RESULTAT EST D'ABORD OBTENUE PAR UNE APPROCHE DIRECTE. NOUS INTRODUISONS ENSUITE UNE APPROCHE ASYMPTOTIQUE QUI PERMET D'ABORDER LE CAS DES VALEURS SINGULIERES. LA METHODE AINSI DEVELOPPEE PEUT ETRE ADAPTEE POUR RESOUDRE UN PROBLEME VOISIN EN STATISTIQUE DES PROCESSUS: IL S'AGIT DE L'ASYMPTOTIQUE PETIT BRUIT DE L'ESTIMATEUR BAYESIEN DANS LE CAS NON IDENTIFIABLE, LORSQUE L'OBSERVATION EST UN SIGNAL DETERMINISTE PERTURBE. NOUS DONNONS UNE EXPRESSION EXPLICITE POUR LA DENSITE DE LA LOI LIMITE, LORSQUE L'ENSEMBLE DES POINTS MINIMUM DE L'INFORMATION DE KULLBACKLEIBLER EST UNE SOUSVARIETE DE L'ESPACE DES PARAMETRES
2 editions published in 1997 in French and held by 3 WorldCat member libraries worldwide
DANS LA PREMIERE PARTIE DE CETTE THESE, NOUS CONSIDERONS LE PROBLEME DE L'ESTIMATION DE LA TRAJECTOIRE D'UN ENGIN SOUSMARIN REMORQUE, EN SE BASANT SUR DEUX SOURCES D'INFORMATION DIFFERENTES: DES MESURES PRECISES D'ACCELERATION DE L'ENGIN (INS) ET DES MESURES DE POSITION DU NAVIRE DE SURFACE (GPS). LES MESURES INS RESTITUENT FIDELEMENT LES HAUTES FREQUENCES DU MOUVEMENT DE L'ENGIN, MAIS DERIVENT AVEC LE TEMPS. NOUS UTILISONS L'INFORMATION GPS, FIABLE A LONG TERME, POUR RECALER LE MOUVEMENT MOYEN DE L'ENGIN. NOUS INTRODUISONS UN MODELE NUMERIQUE DU SYSTEME CABLEENGIN POUR TRANSFERER L'INFORMATION DE POSITIONNEMENT DEPUIS LA SURFACE JUSQU'A L'ENGIN, ET NOUS PROPOSONS UN ESTIMATEUR HYBRIDE DE SA TRAJECTOIRE. NOUS NOUS INTERESSONS ENSUITE AU PROBLEME DE FILTRAGE DES PROCESSUS DE DIFFUSION, DANS LE CAS OU L'ON DISPOSE D'OBSERVATIONS NONBRUITEES, EN TEMPS DISCRET. CE PROBLEME EST SINGULIER CAR LA LOI CONDITIONNELLE EST SUPPORTEE A CHAQUE INSTANT PAR UN ENSEMBLE DE NIVEAU DE LA FONCTION D'OBSERVATION, QUI EST EN GENERAL DE MESURE NULLE (POUR LA MESURE DE LEBESGUE) DANS L'ESPACE D'ETAT. DANS LE CAS OU LA VALEUR OBSERVEE EST REGULIERE, NOUS OBTENONS UNE EXPRESSION EXPLICITE POUR LA DENSITE DE LA LOI CONDITIONNELLE, PAR RAPPORT A LA MESURE CANONIQUE SUR L'ENSEMBLE DE NIVEAU. CE RESULTAT EST D'ABORD OBTENUE PAR UNE APPROCHE DIRECTE. NOUS INTRODUISONS ENSUITE UNE APPROCHE ASYMPTOTIQUE QUI PERMET D'ABORDER LE CAS DES VALEURS SINGULIERES. LA METHODE AINSI DEVELOPPEE PEUT ETRE ADAPTEE POUR RESOUDRE UN PROBLEME VOISIN EN STATISTIQUE DES PROCESSUS: IL S'AGIT DE L'ASYMPTOTIQUE PETIT BRUIT DE L'ESTIMATEUR BAYESIEN DANS LE CAS NON IDENTIFIABLE, LORSQUE L'OBSERVATION EST UN SIGNAL DETERMINISTE PERTURBE. NOUS DONNONS UNE EXPRESSION EXPLICITE POUR LA DENSITE DE LA LOI LIMITE, LORSQUE L'ENSEMBLE DES POINTS MINIMUM DE L'INFORMATION DE KULLBACKLEIBLER EST UNE SOUSVARIETE DE L'ESPACE DES PARAMETRES
Identification d'un systeme nonlineaire partiellement observe par la methode de la distance minimale [Identification of a
partially observed nonlinear system by
Yu. A Kutoyants(
Book
)
3 editions published in 1993 in French and held by 3 WorldCat member libraries worldwide
3 editions published in 1993 in French and held by 3 WorldCat member libraries worldwide
A differential geometric approach to nonlinear filtering : the projection filter by
Damiano Brigo(
)
1 edition published in 1995 in English and held by 2 WorldCat member libraries worldwide
1 edition published in 1995 in English and held by 2 WorldCat member libraries worldwide
Optimisation spatiotemporelle d'efforts de recherche pour cibles manoeuvrantes et intelligentes by
Mathieu Chouchane(
Book
)
2 editions published in 2013 in French and held by 2 WorldCat member libraries worldwide
In this work, we propose a solution to a problem issued by the DGA Techniques navales in order to survey a strategic area: determining the optimal spatiotemporal deployment of sensors that will maximize the detection probability of a mobile and smart target. The target is said to be smart because it is capable of detecting the threat of the sensors under certain conditions and then of adapting its behaviour to avoid it. The cost of a deployment is known to be very expensive and therefore it has to be taken into account. It is important to note that the wide spectrum of applications within this field of research also reflects the need for a highly complex theoretical framework based on stochastic mono or multiobjective optimisation. Until now, none of the existing works have dealt with the cost of the deployments. Moreover, the majority only treat one type of constraint at a time. Current works mostly rely on operational research algorithms which commonly model the constraints in both discrete space and time.In the first part, we present an algorithm which computes the most efficient spatiotemporal deployment of sensors, but without taking its cost into account. This optimisation method is based on an application of the generalised splitting method.In the second part, we first use a linear combination of the two criteria. For our second approach, we use the evolutionary multiobjective optimisation framework to adapt the generalised splitting method to multiobjective optimisation. Finally, we compare our results with the results of the NSGAII algorithm
2 editions published in 2013 in French and held by 2 WorldCat member libraries worldwide
In this work, we propose a solution to a problem issued by the DGA Techniques navales in order to survey a strategic area: determining the optimal spatiotemporal deployment of sensors that will maximize the detection probability of a mobile and smart target. The target is said to be smart because it is capable of detecting the threat of the sensors under certain conditions and then of adapting its behaviour to avoid it. The cost of a deployment is known to be very expensive and therefore it has to be taken into account. It is important to note that the wide spectrum of applications within this field of research also reflects the need for a highly complex theoretical framework based on stochastic mono or multiobjective optimisation. Until now, none of the existing works have dealt with the cost of the deployments. Moreover, the majority only treat one type of constraint at a time. Current works mostly rely on operational research algorithms which commonly model the constraints in both discrete space and time.In the first part, we present an algorithm which computes the most efficient spatiotemporal deployment of sensors, but without taking its cost into account. This optimisation method is based on an application of the generalised splitting method.In the second part, we first use a linear combination of the two criteria. For our second approach, we use the evolutionary multiobjective optimisation framework to adapt the generalised splitting method to multiobjective optimisation. Finally, we compare our results with the results of the NSGAII algorithm
An adaptive local grid refinement method for nonlinear filtering by
Z Cai(
Book
)
2 editions published in 1995 in English and held by 2 WorldCat member libraries worldwide
2 editions published in 1995 in English and held by 2 WorldCat member libraries worldwide
Small noise asymptotics of the Bayesian estimator in nonidentifiable models by
Marc Joannides(
Book
)
2 editions published in 1999 in English and held by 2 WorldCat member libraries worldwide
2 editions published in 1999 in English and held by 2 WorldCat member libraries worldwide
A differential geometric approach to nonlinear filtering the projection filter by
Damiano Brigo(
Book
)
2 editions published in 1995 in English and held by 2 WorldCat member libraries worldwide
2 editions published in 1995 in English and held by 2 WorldCat member libraries worldwide
Stability and uniform approximation of nonlinear filters using the Hilbert metric, and application to particle filters by
François Le Gland(
Book
)
2 editions published in 2001 in English and held by 2 WorldCat member libraries worldwide
2 editions published in 2001 in English and held by 2 WorldCat member libraries worldwide
Cartes incertaines et planification optimale pour la localisation d'un engin autonome by
Francis Celeste(
Book
)
2 editions published in 2010 in French and held by 2 WorldCat member libraries worldwide
Important improvements have been done in robotics. More and more mobile robots and small unmanned aerial vehicles are planned to be used in different applications due to their ability to move autonomously in their environment. The localization task, through the matching between measurements provided by embedded sensors and features given in a map, is essential. This process, which is based on data fusion techniques, has been well studied. In this thesis, our main goal is to define a methodology for offline path planning, in order to guarantee the best performance of localization of the robot during motion execution. This performance takes into account the uncertainty of the system dynamics and also of the given environment map representation. First of all, we introduce a way to generate random outcomes of the uncertain map from a model of errors using the point process theory and we produce a multilevel uncertain map for localization. The criterion for planning is built from the posterior CramèrRao bound for the estimation of the system dynamics and the map uncertainty. Both discrete and continuous system dynamical models are considered. The planning problem is solved via heuristic approaches based on the crossentropy method. The analysis of the performance of the derived path solution is then made using results from the extreme values theory. Finally, some ideas are introduced to demonstrate that the map quality can be improved under resources constraints regarding the localization performance criterion
2 editions published in 2010 in French and held by 2 WorldCat member libraries worldwide
Important improvements have been done in robotics. More and more mobile robots and small unmanned aerial vehicles are planned to be used in different applications due to their ability to move autonomously in their environment. The localization task, through the matching between measurements provided by embedded sensors and features given in a map, is essential. This process, which is based on data fusion techniques, has been well studied. In this thesis, our main goal is to define a methodology for offline path planning, in order to guarantee the best performance of localization of the robot during motion execution. This performance takes into account the uncertainty of the system dynamics and also of the given environment map representation. First of all, we introduce a way to generate random outcomes of the uncertain map from a model of errors using the point process theory and we produce a multilevel uncertain map for localization. The criterion for planning is built from the posterior CramèrRao bound for the estimation of the system dynamics and the map uncertainty. Both discrete and continuous system dynamical models are considered. The planning problem is solved via heuristic approaches based on the crossentropy method. The analysis of the performance of the derived path solution is then made using results from the extreme values theory. Finally, some ideas are introduced to demonstrate that the map quality can be improved under resources constraints regarding the localization performance criterion
Geometric ergodicity in hidden Markov models by
F Le Gland(
Book
)
2 editions published in 1996 in English and held by 2 WorldCat member libraries worldwide
2 editions published in 1996 in English and held by 2 WorldCat member libraries worldwide
Small noise asymptotics of the bayesian estimator in nonidentifiable models by
Marc Joannides(
)
1 edition published in 1999 in English and held by 1 WorldCat member library worldwide
1 edition published in 1999 in English and held by 1 WorldCat member library worldwide
Méthodes d'estimation statistique pour le suivi de cibles à l'aide d'un réseau de capteurs by
Adrien Ickowicz(
Book
)
2 editions published in 2010 in French and held by 1 WorldCat member library worldwide
This thesis is concerned with the estimation of the dynamical parameters of one or multiple targets moving through an area which is "watched'' by a sensor network. The first part deal with performance analysis for data association, in a target tracking environment. Effects of misassociation are considered in a simple (linear) multiscan framework so as to provide closedform expressions of the probability of correct association. Via rigorous calculations the effect of dimensioning parameters is analyzed. Remarkably, it is possible to derive very simple expressions of the probability of correct association which are independent of the scenario kinematic parameters. In a second part, we are especially interested by fusing binary information at the network level. This binary information is related to the local target behavior; i.e. its distance from a given sensor is increasing () or decreasing (+). However, in this rich framework we choose to focus on even simpler observations so as to put in evidence the limits and the difficulties of the decentralized binary framework. More specifically, the binary sequences {,+} can be (locally) summarized by the times of closest point approach (cpa). So, we consider that the available observations, at the network level, are the estimated values of the cpa times. The analysis is also greatly simplified if we assume that the target motion is rectilinear and uniform or a legbyleg one. In the case of a singleleg trajectory, we resort to a simple maximumlikelihood estimator, while for the case of multipleleg trajectories other methods are presented. Then, we finally present a new algorithm for target tracking within a binary sensor network. The novel tracking method is proposed and its performance through a very classical trajectory model is evaluated. We finally try to extend the algorithm to multiple target tracking
2 editions published in 2010 in French and held by 1 WorldCat member library worldwide
This thesis is concerned with the estimation of the dynamical parameters of one or multiple targets moving through an area which is "watched'' by a sensor network. The first part deal with performance analysis for data association, in a target tracking environment. Effects of misassociation are considered in a simple (linear) multiscan framework so as to provide closedform expressions of the probability of correct association. Via rigorous calculations the effect of dimensioning parameters is analyzed. Remarkably, it is possible to derive very simple expressions of the probability of correct association which are independent of the scenario kinematic parameters. In a second part, we are especially interested by fusing binary information at the network level. This binary information is related to the local target behavior; i.e. its distance from a given sensor is increasing () or decreasing (+). However, in this rich framework we choose to focus on even simpler observations so as to put in evidence the limits and the difficulties of the decentralized binary framework. More specifically, the binary sequences {,+} can be (locally) summarized by the times of closest point approach (cpa). So, we consider that the available observations, at the network level, are the estimated values of the cpa times. The analysis is also greatly simplified if we assume that the target motion is rectilinear and uniform or a legbyleg one. In the case of a singleleg trajectory, we resort to a simple maximumlikelihood estimator, while for the case of multipleleg trajectories other methods are presented. Then, we finally present a new algorithm for target tracking within a binary sensor network. The novel tracking method is proposed and its performance through a very classical trajectory model is evaluated. We finally try to extend the algorithm to multiple target tracking
Particlebased methods for parameter estimation and tracking : numerical experiments by Johann Fichou(
)
1 edition published in 2004 in English and held by 1 WorldCat member library worldwide
1 edition published in 2004 in English and held by 1 WorldCat member library worldwide
Geometric ergodicity in hidden markov models by
François Le Gland(
)
1 edition published in 1996 in English and held by 1 WorldCat member library worldwide
1 edition published in 1996 in English and held by 1 WorldCat member library worldwide
An adaptive local grid refinement method for nonlinear filtering by Zhiqiang Cai(
)
1 edition published in 1995 in English and held by 1 WorldCat member library worldwide
1 edition published in 1995 in English and held by 1 WorldCat member library worldwide
Estimation de probabilités d'évènements rares et de quantiles extrêmes : applications dans le domaine aérospatial by
Rudy Pastel(
Book
)
2 editions published in 2012 in English and held by 1 WorldCat member library worldwide
Rare event dedicated techniques are of great interest for the aerospace industry because of the large amount money that can be lost because of risks associated with minute probabilities. This thesis is focused on the search of probability techniques able to estimate rare event probabilities and extreme quantiles associated with a black box system with static random inputs through two case studies from the industry. The first one is the estimation of the probability of collision between satellites Iridium and Cosmos. The CrossEntropy (CE), the Nonparametric Adaptive Importance Sampling (NAIS) and an Adaptive Splitting Technique (AST) are compared. Through the comparison, an improved version of NAIS is designed. Whereas NAIS needs to be initiated with a auxiliary random variable which straight away generates rare events, the Adaptive NAIS (ANAIS) allows one to use the original input random as initial auxiliary density and therefore does not require a priori knowledge. The second case is the estimation of the safety zone with respect to the fall of a spacecraft booster. Though they can be estimated via ANAIS or AST, extreme quantiles are shown to be not adapted to spatial distribution. For that purpose, the Minimum Volume Set (MVS) is chosen from the literature. The Crude Monte Carlo (CMC) plugin MVS estimator being not adapted to extreme level MVS estimation, both ANAIS and AST are adapted into plugin extreme MVS estimators.These two algorithms outperform the CMC plugin MVS estimator
2 editions published in 2012 in English and held by 1 WorldCat member library worldwide
Rare event dedicated techniques are of great interest for the aerospace industry because of the large amount money that can be lost because of risks associated with minute probabilities. This thesis is focused on the search of probability techniques able to estimate rare event probabilities and extreme quantiles associated with a black box system with static random inputs through two case studies from the industry. The first one is the estimation of the probability of collision between satellites Iridium and Cosmos. The CrossEntropy (CE), the Nonparametric Adaptive Importance Sampling (NAIS) and an Adaptive Splitting Technique (AST) are compared. Through the comparison, an improved version of NAIS is designed. Whereas NAIS needs to be initiated with a auxiliary random variable which straight away generates rare events, the Adaptive NAIS (ANAIS) allows one to use the original input random as initial auxiliary density and therefore does not require a priori knowledge. The second case is the estimation of the safety zone with respect to the fall of a spacecraft booster. Though they can be estimated via ANAIS or AST, extreme quantiles are shown to be not adapted to spatial distribution. For that purpose, the Minimum Volume Set (MVS) is chosen from the literature. The Crude Monte Carlo (CMC) plugin MVS estimator being not adapted to extreme level MVS estimation, both ANAIS and AST are adapted into plugin extreme MVS estimators.These two algorithms outperform the CMC plugin MVS estimator
more
fewer
Audience Level
0 

1  
General  Special 
Related Identities
 James, Matthew R. Author
 Université Paris DauphinePSL Degree grantor
 Kutoyants, Yurii Author
 Florchinger, Patrick Author
 Joannides, Marc Author
 IRISA (117)
 Brigo, Damiano (1966....). Author
 Campillo, Fabien Author
 Institut national de recherche en informatique et en automatique (France)
 Université de Rennes 1 Degree grantor
Associated Subjects