Overblog
Suivre ce blog Administration + Créer mon blog
16 novembre 2015 1 16 /11 /novembre /2015 17:55
Une équipe de douze drones et robots en mission autonome

Architecture décisionnelle multidrone développée dans le projet et intégrée dans chaque véhicule  (AAV = Autonomous Aerial Vehicle, AGV = Autonomous Ground Vehicle). © ONERA

 

06.11.2015 ONERA

 

La sixième et dernière démonstration de l'Étude amont Action* a mis en œuvre 12 drones aériens et terrestres dans un environnement rurbain. Objectif : valider une architecture logicielle distribuée « décisionnelle » pour la coopération de drones évoluant dans une mission commune.

 

La démonstration  a eu lieu les 19 et 20 octobre 2015 sur le village de combat du camp militaire de Caylus (82), en présence de la Direction générale de l'armement (DGA), ainsi que des personnels de l’ONERA et du CNRS/LAAS** acteurs de la démonstration.

 

L'objectif était de mettre en avant la capacité d’une architecture logicielle décisionnelle, embarquée sur une équipe hétérogène de douze véhicules autonomes, à réaliser en s’entraidant une mission « patrouille, localisation, suivi de cible » dans un environnement mixte urbain/campagne.

 

L'équipe de drones était constituée de deux drones aériens Vario Benzin/ReSSAC de l'ONERA, de trois robots terrestres du CNRS/LAAS, de trois robots terrestres Effibot de DGA Techniques terrestres mis en oeuvre par l'ONERA et de quatre véhicules simulés dont un drone aérien de type Ressac, un robot terrestre de type LAAS et deux robots terrestres de type Effibot.

 

Rendez-vous planifié entre les robots Mana (LAAS) et Effibot (DGA) photo ONERA

Rendez-vous planifié entre les robots Mana (LAAS) et Effibot (DGA) photo ONERA

Suivi de la cible par un Vario -Ressac  (drone aérien ONERA) photo ONERA

Suivi de la cible par un Vario -Ressac (drone aérien ONERA) photo ONERA

Cette démonstration a été la plus complexe de la série des six expérimentations terrain du projet Action, qui se sont déroulées de 2012 à 2015 dans des contextes aéroterrestre ou aéromaritime. De nombreux aléas ont été introduits dans cette dernière mission, comme des pannes de véhicules, des retards, des défauts de communication, l’apparition d’une deuxième cible…  Il s’agissait bien de valider l’architecture décisionnelle embarquée qui donne à l'équipe de drones son autonomie pour la réalisation de la mission.

 

Cette démonstration a été un succès, ainsi que le projet dans son ensemble, tous les scénarios joués ayant montré la capacité de l'équipe de drones à réaliser la patrouille de la zone, y compris en présence d'événements perturbateurs. La robustesse de l’architecture multidrone embarquée sur chaque véhicule a été démontrée.

 

 

Le mot de la cheffe de projet, Magali Barbier (ONERA)

 

"Ce projet qui a duré 9 années a permis de faire avancer l'état de la recherche sur la problématique de coopération de drones autonomes hétérogènes. Plus de 30 publications et 5 thèses sont actuellement disponibles sur le site du projet.

L'architecture décisionnelle générique conserve les capacités individuelles d'autonomie de chaque véhicule ; elle planifie et répare les plans en utilisant de l'expertise humaine ; elle prévoit des tâches de rendez-vous entre véhicules pour garantir la synchronisation des actions dans un environnement où les communications ne sont ni continues ni parfaites hormis en ces points ; elle est connectée à un haut niveau sémantique avec un opérateur mission lui permettant d'intervenir pour aider la mission.

Au cours de ces années, le projet a vu avancer les progrès sur les drones : sur les engins eux-mêmes, les batteries, les charges utiles, les processus embarqués, les interfaces logicielles, les moyens de communication, les interfaces homme-machine...

La recherche doit continuer dans les laboratoires sur la thématique de la coopération de drones autonomes. Le projet Action a posé des bases solides notamment sur les fonctions de planification et de supervision, mais de nombreux verrous technologiques doivent encore être étudiés ; on peut citer sans être exhaustif les stratégies de suivi coopératif, la localisation coopérative, la formalisation des communications (quelles informations à qui et quand), le suivi de situation global, une réflexion sur le rôle et la charge de travail des opérateurs mission...

Et si le choix dans le projet Action a été de distribuer totalement l'architecture décisionnelle sur les drones (drones au sens large, air-terre-mer), certaines fonctions peuvent se partager entre l'embarqué et le sol en fonction des contraintes de communication ; des stratégies de maintien de la communication par un réseau de drones sont aussi à l'étude."

 

Notes

Action – coopération de véhicules hétérogènes autonomes

L'objectif de l'étude amont Action est de développer et d'implémenter sur des véhicules hétérogènes autonomes une architecture logicielle multidrone permettant de les faire coopérer pour la réalisation de leur mission. Le projet est centré sur l'amélioration de la fonction localisation des cibles et des véhicules amis sous contraintes de communication. Les travaux de recherche poursuivis par l'équipe projet s'organisent autour du développement des briques de l'architecture décisionnelle multidrone pour la réalisation des fonctions de fusion de données et de décision (planification et supervision de la mission).

 

** LAAS : Laboratoire d'analyse et d'architecture des systèmes (CNRS)

 

En savoir plus

 

Qui dit autonomie dit capacité de décision, dans "Les pépites de l'ONERA"

Le jour où les drones travailleront ensemble (à propos du projet Action)

action.onera.fr, le site web du projet

Partager cet article

Repost0
10 mars 2015 2 10 /03 /mars /2015 12:20
SAIC to continue USMC AAV upgrade programme

 

9 March 2015 naval-technology.com

 

The US Marine Corps (USMC) Programme Executive Office - Land Systems has exercised options under a contract awarded to Science Applications International Corporation (SAIC) to support its assault amphibious vehicle (AAV) upgrade programme.

 

As part of the prime contract signed in May last year, SAIC is responsible for engineering, designing and test upgrading of ten prototypes and 52 low-rate initial production (LRIP) vehicles.

 

The initial $16m, multiple award included options for prototype vehicle builds and trials, followed by a LRIP.

 

SAIC Navy and Marine Corps Customer Group senior vice-president and general manager Tom Watson said: "This significant win demonstrates SAIC's ingenuity to meet the demands of our customers and to serve the mission interests of the USMC.

 

"SAIC is honoured to have the opportunity to support the USMC on this highly important programme to ensure the safety and amphibious mission success of our nation's marines."

 

The contract value has increased to $69m and, if all options are exercised, will rise to approximately $194m over five years.

 

Work includes armour and engine rebuilds to enhance horsepower and torque, replacement of ageing transmission, modernisation of suspension components and new water jets and blast-resistant seats, as well as upgrades to vehicle control, instrumentation and driver interface systems.

 

SAIC will carry out upgrade works primarily at its facility in Charleston, South Carolina, US.

 

AAVs are capable of attacking any shoreline from the well decks of navy assault ships and can carry marines and cargo through hostile environments.

Partager cet article

Repost0
28 mai 2013 2 28 /05 /mai /2013 12:20
U.S. Marines with 2nd Amphibious Assault Battalion, 2nd Marine Division, storm ashore in an amphibious assault vehicle during Exercise Cobra Gold 2011 in Thailand. The Marine Corps is moving forward with plans to replace the aging AAVs with the Amphibious Combat Vehicle now under development. (Staff Sgt. Leo Salinas / U.S. Marine Corps)

U.S. Marines with 2nd Amphibious Assault Battalion, 2nd Marine Division, storm ashore in an amphibious assault vehicle during Exercise Cobra Gold 2011 in Thailand. The Marine Corps is moving forward with plans to replace the aging AAVs with the Amphibious Combat Vehicle now under development. (Staff Sgt. Leo Salinas / U.S. Marine Corps)

 

May 28, 2013: Strategy Page

 

In April 2013 DARPA (the Defense Advanced Research Projects Agency) awarded a million dollar prize to a three man design team (Team Ground Systems) for proposing the most promising concept for the new Marine Corps Amphibious Combat vehicle (ACV).  The winner beat out a thousand other proposals. DARPA is offering another million dollar prize for the best drive train (propulsion system) design and then a two million dollar prize for a complete vehicle design. This approach may sound either very innovative or very desperate and in reality it is both. In part because the marines recently blew three billion dollars in an unsuccessful attempt to design and develop a high-speed ACV and partly because that failure made it clear that some original thinking was required.  

 

For over a decade now DARPA has used this competitive (or “crowdsourcing”) approach, especially in several competitions to develop UGVs (unmanned ground vehicles.) DARPA has been using this crowdsourcing approach successfully so the marines saw it as a possible solution to their ACV problem. The basic problem is that the marines insist that the new ACV be able move towards shore at twice the speed of the older AAV7. The inability of the previous EFV design to accomplish that cost the marines three billion dollars and over a decade of development effort.

American Marines Seeking New Ideas

Two years ago the marines cancelled their EFV (Expeditionary Fighting Vehicle) and have been hustling to come up with a replacement ACV design. Meanwhile they must extend the life of their current 1,057 AAV7 amphibious armored vehicles. These entered service three decades ago and are falling apart. Moreover, some two thirds of the AAV7s saw service in Iraq, where they got as much use in two months as they normally did in two years of peacetime operations. Most AAV7s are already scheduled for refurbishing, so they can still be used until the end of the decade, or whenever a permanent replacement can be found.

 

The marines how have two replacement vehicle projects going. The MPC (Marine Personnel Carrier) is a $4.5 million wheeled, amphibious armored vehicle. This would be similar to the Stryker, but a bit larger and modified for amphibious operations. This project is proceeding because it is low-risk (in the technology department) and the marines need some kind of armored vehicle to replace AAV7s that are dying of old age. The $12 million ACV is the EFV without most of the expensive stuff that didn't work. In effect, the ACV will be a 21st century version of the AAV7, optimized to pass all its development tests and get into service as quickly as possible. The marines do not want to be reminded of the EFV.

AAV7s Come Ashore Somalia

AAV7s Come Ashore Somalia

The cancelled EFV ended up costing over ten times as much as the $2.5 million AAV7 (taking inflation into account). The marines apparently felt they could get by with half as many amphibious armored vehicles because future wars are likely to be more dependent on delivering troops by air, or moving them around in armored hummers. While there was some thought of dispensing entirely with vehicles like this, which were first used in 1943, more traditionalist minds prevailed. That may change, especially since the cheaper MPC is more likely to survive the budget battles than the ACV.

 

The EFV had been threatened with cancellation for several years, mainly because the vehicle was too expensive and didn't work. Well, parts of it worked. Three years ago, tests revealed that the EFV had similar survivability characteristics to MRAPs, when hit with roadside bombs or anti-vehicle mines. The EFV needed all the good news it could get, but marines were already using MRAPs in Afghanistan, and are quite happy with them.

 

The EFV was previously called the AAAV (Advanced Amphibious Assault Vehicle). Weighing nearly 36 tons, the EFV was 3.4 meters (10.5 feet) tall, 3.9 meters (12 feet) wide and just under 10 meters (30 feet) long. It was armed with a 30mm automatic cannon (MK34 Bushmaster) and a 7.62 mm co-axial machine gun. The EFV also had better armor protection and electronics than the AAV7. The EFV was about 25 percent heavier than the AAV7, and somewhat larger.

American Marines Seeking New Ideas

The EFV had been in development for over a decade and delays were mostly because of a complex water-jet propulsion system which, when it worked, allowed it to travel at 60 kilometers an hour while in the water. This capability was specified to reduce the danger (from enemy fire) when the EFVs were moving from their transports to shore, a distance of 30-50 kilometers. The additional gear required for the water jet system made the vehicle less robust and reliable, and fixing those problems took too much time. Otherwise, the EFV was basically a truly amphibious Infantry Fighting Vehicle (IFV), similar to the army's smaller M-2 Bradley. The EFV had a crew of three, and carried 18 passengers.

 

In retrospect, the marines could have just built the ACV, using mature technologies and staying away from the high speed (and high tech) water jet system that provided a capability that was not really critical. But that's hindsight. Lesson, hopefully, learned. But with much tighter budgets looming, the marines may run out of money, not patience, this time around. The proposed ACV is also very expensive, and the MPC is not as capable (for amphibious operations) as the current AAV7. All they may end up with is some refurbished AAV7s, and maybe not many of those either. The budget situation is grim, leaving the usually unstoppable Marine Corps running into an immovable object.

Partager cet article

Repost0

Présentation

  • : RP Defense
  • : Web review defence industry - Revue du web industrie de défense - company information - news in France, Europe and elsewhere ...
  • Contact

Recherche

Articles Récents

Categories