Similarly to marine mammals, mankind has developed sonar systems which are able to perceive underwater environments on a range of distances and with more or less detail, depending on its needs. The need addressed in this manuscript is automatic underwater target (ATR) recognition, whether the target is laying on the seafloor or floating in the water column. The chronology of the work which is presented shows how sonar systems and underlying algorithms have progressively been improved in order to adapt to new threats, from the stealthy mines of the 90s to the improvised explosive devices of today. To counter these threats using side-scan or synthetic aperture sonar (SAS) data, two topics in particular are discussed: first, a new mine-hunting ATR approach which takes into consideration the environment in the vicinity of the target; second, the analysis and use of repeat-pass SAS data acquired in a monitoring context in order to perform change detection. Apart from the detection of targets lying on the seabed, obstacle detection issues have more recently been tackled in order to allow underwater drones to operate in complete safety whatever the depth. For each type of mission –seabed tracking, obstacle detection and tracking in the water column, and vehicle surfacing–, the combination of configuration and mode of the obstacle-avoidance sonar systems, and dedicated algorithms are described.The body of work presented is based on high-frequency sonar imaging, ranging from side-scan sonars to forward-looking sonars, and most of it is the result of various collaborations. Initially intended for operators, the use of sonar images needs to be adapted to an increasing quantity of data to be processed, and to the increasing automation of tasks aboard underwater drones. The stealth, flexibility and safety of these small-size vehicles open up new opportunities both of missions and of choice of relevant imaging and navigation sensors, not only in the military but also in the civilian domain. More generally, this adaptation ...
Similarly to marine mammals, mankind has developed sonar systems which are able to perceive underwater environments on a range of distances and with more or less detail, depending on its needs. The need addressed in this manuscript is automatic underwater target (ATR) recognition, whether the target is laying on the seafloor or floating in the water column. The chronology of the work which is presented shows how sonar systems and underlying algorithms have progressively been improved in order to adapt to new threats, from the stealthy mines of the 90s to the improvised explosive devices of today. To counter these threats using side-scan or synthetic aperture sonar (SAS) data, two topics in particular are discussed: first, a new mine-hunting ATR approach which takes into consideration the environment in the vicinity of the target; second, the analysis and use of repeat-pass SAS data acquired in a monitoring context in order to perform change detection. Apart from the detection of targets lying on the seabed, obstacle detection issues have more recently been tackled in order to allow underwater drones to operate in complete safety whatever the depth. For each type of mission –seabed tracking, obstacle detection and tracking in the water column, and vehicle surfacing–, the combination of configuration and mode of the obstacle-avoidance sonar systems, and dedicated algorithms are described.The body of work presented is based on high-frequency sonar imaging, ranging from side-scan sonars to forward-looking sonars, and most of it is the result of various collaborations. Initially intended for operators, the use of sonar images needs to be adapted to an increasing quantity of data to be processed, and to the increasing automation of tasks aboard underwater drones. The stealth, flexibility and safety of these small-size vehicles open up new opportunities both of missions and of choice of relevant imaging and navigation sensors, not only in the military but also in the civilian domain. More generally, this adaptation ...
Similarly to marine mammals, mankind has developed sonar systems which are able to perceive underwater environments on a range of distances and with more or less detail, depending on its needs. The need addressed in this manuscript is automatic underwater target (ATR) recognition, whether the target is laying on the seafloor or floating in the water column. The chronology of the work which is presented shows how sonar systems and underlying algorithms have progressively been improved in order to adapt to new threats, from the stealthy mines of the 90s to the improvised explosive devices of today. To counter these threats using side-scan or synthetic aperture sonar (SAS) data, two topics in particular are discussed: first, a new mine-hunting ATR approach which takes into consideration the environment in the vicinity of the target; second, the analysis and use of repeat-pass SAS data acquired in a monitoring context in order to perform change detection. Apart from the detection of targets lying on the seabed, obstacle detection issues have more recently been tackled in order to allow underwater drones to operate in complete safety whatever the depth. For each type of mission –seabed tracking, obstacle detection and tracking in the water column, and vehicle surfacing–, the combination of configuration and mode of the obstacle-avoidance sonar systems, and dedicated algorithms are described.The body of work presented is based on high-frequency sonar imaging, ranging from side-scan sonars to forward-looking sonars, and most of it is the result of various collaborations. Initially intended for operators, the use of sonar images needs to be adapted to an increasing quantity of data to be processed, and to the increasing automation of tasks aboard underwater drones. The stealth, flexibility and safety of these small-size vehicles open up new opportunities both of missions and of choice of relevant imaging and navigation sensors, not only in the military but also in the civilian domain. More generally, this adaptation ...
Similarly to marine mammals, mankind has developed sonar systems which are able to perceive underwater environments on a range of distances and with more or less detail, depending on its needs. The need addressed in this manuscript is automatic underwater target (ATR) recognition, whether the target is laying on the seafloor or floating in the water column. The chronology of the work which is presented shows how sonar systems and underlying algorithms have progressively been improved in order to adapt to new threats, from the stealthy mines of the 90s to the improvised explosive devices of today. To counter these threats using side-scan or synthetic aperture sonar (SAS) data, two topics in particular are discussed: first, a new mine-hunting ATR approach which takes into consideration the environment in the vicinity of the target; second, the analysis and use of repeat-pass SAS data acquired in a monitoring context in order to perform change detection. Apart from the detection of targets lying on the seabed, obstacle detection issues have more recently been tackled in order to allow underwater drones to operate in complete safety whatever the depth. For each type of mission –seabed tracking, obstacle detection and tracking in the water column, and vehicle surfacing–, the combination of configuration and mode of the obstacle-avoidance sonar systems, and dedicated algorithms are described.The body of work presented is based on high-frequency sonar imaging, ranging from side-scan sonars to forward-looking sonars, and most of it is the result of various collaborations. Initially intended for operators, the use of sonar images needs to be adapted to an increasing quantity of data to be processed, and to the increasing automation of tasks aboard underwater drones. The stealth, flexibility and safety of these small-size vehicles open up new opportunities both of missions and of choice of relevant imaging and navigation sensors, not only in the military but also in the civilian domain. More generally, this adaptation process to underwater mobile robotics will require the analysis of perception-action mechanisms in an uncertain and unstructured environment, with limited communication. ; A l'instar des mammifères marins, l'homme a développé des systèmes sonar capables de percevoir l'environnement marin sur des distances plus ou moins longues et avec plus ou moins de détails en fonction de ses besoins. Le principal besoin traité dans ce travail préparé pour l'habilitation à diriger les recherches est la reconnaissance automatique de cibles sous-marines (plus connue sous l'acronyme ATR pour Automatic Target Recognition), qu'elles soient posées sur le fond ou flottantes entre deux eaux.La chronologie de la sélection de travaux présentés suit le perfectionnement au fil des années des systèmes d'imagerie et des algorithmes de traitement associés pour répondre à l'évolution de la menace, depuis les mines furtives des années 90 aux engins explosifs improvisés d'aujourd'hui. Pour contrer ces menaces à partir de données sonar latéral ou sonar à antenne synthétique (dites données SAS pour Synthetic Aperture Sonar), deux points sont en particulier exposés : une nouvelle approche de l'ATR pour les missions de chasse aux mines qui passe par la qualification de l'environnement proche de la mine d'une part, la compréhension et l'exploitation de données SAS acquises lors de passes répétitives pour la mise en œuvre d'algorithmes de détection (incohérente puis cohérente) de changements dans un contexte de surveillance d'autre part. Outre la reconnaissance de cibles sur le fond marin, la problématique de la détection d'obstacles s'est imposée plus récemment afin de permettre à des drones sous-marins d'autonomie croissante de mener leur mission en toute sécurité et ce, quelle que soit leur immersion. Dans ce manuscrit, la configuration et le mode des systèmes sonar d'évitement, de même que les algorithmes dédiés à chaque mission - suivi de fond, pistage d'obstacles dans la colonne d'eau, et enfin reprise de vue-, sont discutés.La synthèse des travaux effectuée pour cette habilitation à diriger les recherches repose ainsi sur l'exploitation des systèmes d'imagerie sonar haute fréquence (HF), des sonars latéraux aux sonars frontaux, et est le plus souvent le fruit de diverses collaborations. Initialement destiné aux opérateurs, l'usage des images sonar HF est aujourd'hui à adapter face au flot croissant de données à traiter et à l'automatisation progressive des tâches à bord de drones sous-marins. La discrétion, la souplesse et la sécurité de ces véhicules ouvrent de nombreuses possibilités en termes de missions et d'utilisation de capteurs ad hoc d'imagerie et de navigation, y compris pour le domaine civil. Plus globalement, cette adaptation à la robotique mobile sous-marine exigera l'étude des mécanismes de perception-action dans un environnement incertain et peu structuré avec une communication limitée.
International audience ; To assess seabed geoacoustic properties, Ocean Acoustic Tomography (OAT) uses powerful active emissions of repetitive signals causing problems when acoustic discretion is required as in military operations. A solution to avoid this disadvantage consists in developing a new concept of OAT, called "discreet acoustic tomography", which is based on stealthy acoustic signals emissions. In this paper, we propose an innovative strategy to synthesize, thanks to a global optimization method, signals which on the one hand maximize the estimation accuracy of underwater acoustic channel parameters and on the other hand, minimize the detection probability of active emissions by a nearby unknown interceptor. Finally, this procedure is applied to a realistic shallow water scenario of which objective consists in hiding a synthetic signal in ship noise with the constraint to have an accurate estimation of the channel parameters. Results obtained illustrate the validity and the potential of the proposed method.
International audience ; To assess seabed geoacoustic properties, Ocean Acoustic Tomography (OAT) uses powerful active emissions of repetitive signals causing problems when acoustic discretion is required as in military operations. A solution to avoid this disadvantage consists in developing a new concept of OAT, called "discreet acoustic tomography", which is based on stealthy acoustic signals emissions. In this paper, we propose an innovative strategy to synthesize, thanks to a global optimization method, signals which on the one hand maximize the estimation accuracy of underwater acoustic channel parameters and on the other hand, minimize the detection probability of active emissions by a nearby unknown interceptor. Finally, this procedure is applied to a realistic shallow water scenario of which objective consists in hiding a synthetic signal in ship noise with the constraint to have an accurate estimation of the channel parameters. Results obtained illustrate the validity and the potential of the proposed method.
International audience ; Military Autonomous Underwater Vehicles (AUV) shall be able to execute survey missions in both known and unknown environments in order to detect a potential threat. These robots will significantly improve our exploration, analysis and intervention capability and will have a large decisional autonomy. While the primary mission of an AUV is data acquisition and collection (up to now commonly done using side scan sonar or a multibeam echosounder), another important task is to guaranty its own security. To do that, it must be able to know in advance its environment, to detect unexpected events, to analyse them, and to react. The paper occurs after the DEVITOBS'06 (Détection et EVITement d'OBStacles) experiment using two types of sensors mounted on the GESMA Redermor experimental AUV: the forward looking sonar Reson Seabat 8101 with a depression angle of 15° and an echosounders network. It has been divided in three main parts: analysis of the "Obstacle Detection and Avoidance" problem for AUV, information extraction techniques assessment, and discussion about behaviour strategies and mission planning.
International audience ; Military Autonomous Underwater Vehicles (AUV) shall be able to execute survey missions in both known and unknown environments in order to detect a potential threat. These robots will significantly improve our exploration, analysis and intervention capability and will have a large decisional autonomy. While the primary mission of an AUV is data acquisition and collection (up to now commonly done using side scan sonar or a multibeam echosounder), another important task is to guaranty its own security. To do that, it must be able to know in advance its environment, to detect unexpected events, to analyse them, and to react. The paper occurs after the DEVITOBS'06 (Détection et EVITement d'OBStacles) experiment using two types of sensors mounted on the GESMA Redermor experimental AUV: the forward looking sonar Reson Seabat 8101 with a depression angle of 15° and an echosounders network. It has been divided in three main parts: analysis of the "Obstacle Detection and Avoidance" problem for AUV, information extraction techniques assessment, and discussion about behaviour strategies and mission planning.
International audience ; Ocean Acoustic Tomography (OAT) uses powerful active emissions of repetitive signals causing problems when acoustic discretion is required as in military operations. In this paper, we propose to develop a new concept of OAT, called Discreet Acoustic Tomography (DAT), which is based on a stealthy acoustic signals emission. An appropriate global procedure to synthesize a signal waveform in accordance with the compromise between interception probability and accuracy in channel parameters estimation, is proposed. Finally, this procedure was applied to a realistic scenario of which objective consists in hiding a synthetic signal in ship noise with the constraint to have an accurate estimation of the channel parameters. Results obtained illustrate the interest and the potential of the proposed method.
International audience ; Ocean Acoustic Tomography (OAT) uses powerful active emissions of repetitive signals causing problems when acoustic discretion is required as in military operations. In this paper, we propose to develop a new concept of OAT, called Discreet Acoustic Tomography (DAT), which is based on a stealthy acoustic signals emission. An appropriate global procedure to synthesize a signal waveform in accordance with the compromise between interception probability and accuracy in channel parameters estimation, is proposed. Finally, this procedure was applied to a realistic scenario of which objective consists in hiding a synthetic signal in ship noise with the constraint to have an accurate estimation of the channel parameters. Results obtained illustrate the interest and the potential of the proposed method.
