Séminaire Chaire Drive for All « Conduite autonome : entre modèle et apprentissage »

La chaire Drive for All est heureuse de vous inviter au séminaire “Conduite autonome : entre modèle et apprentissage”. Le séminaire aura lieu le mercredi 29 novembre à MINES ParisTech.

logoDriveForAll

Le séminaire est public et gratuit mais l’inscription est obligatoire https://goo.gl/forms/zTXI5ZFyr5yNN18f2

L’objectif de la journée est de donner des éléments de compréhension et les défis en cours concernant la conduite autonome.

Deux approches sont actuellement concurrentes : d’une part l’apprentissage (ou intelligence artificielle, souvent représentée par l’approche Deep Learning) qui est de nature probabiliste mais offre des performances étonnantes, et d’autre part les algorithmes déterministes, qui offrent plus de garanties mais sont souvent plus limités.

Ecole d’été 2017 « Cooperative Interacting Automobiles »

Une école d’été 2017 sur «Cooperative Interacting Automobiles» est organisée du 8 au 20 août 2017 à Schwäbisch Gmünd, en Allemagne

L’école d’été vise à rassembler le doctorat. Les étudiants et les jeunes professionnels des universités et des institutions de recherche pour des tutoriels, des conférences, des présentations de posters et des séances d’ateliers interactives sur :

  • Perception coopérative
  • Planification de la motion coopérative
  • Interaction implicite et explicite

Il est organisé conjointement par Karlsruhe Institute of Technology (KIT, Allemagne) et MINES ParisTech (France). Nous remercions le programme prioritaire « Cooperative Interacting Automobiles » de la Fondation allemande de la science (DFG), de la Société de systèmes de transport intelligents IEEE et de la Chaire Drive for All (MINES ParisTech) pour leur soutien.

Pour plus d’information.

Séminaire « Infrastructure-Assisted V2V Messaging for cooperative autonomous driving »

La Chaire Drive for All est heureuse de vous inviter au séminaire “Infrastructure-Assisted V2V Messaging for cooperative autonomous driving”. Nous avons l’honneur de recevoir le Dr. Manabu Tsukada, de la Graduate School of Information Science and Technology de l’université de Tokyo, Japon. Le séminaire aura lieu lundi 22 mai à MINES ParisTech.

Le séminaire est public et gratuit mais l’inscription est obligatoire https://goo.gl/forms/DmPDCrZXCzvJwK8L2.

Lieu:

MINES ParisTech – Salle Chevalier, 60, Boulevard Saint-Michel 75006 Paris
Luxembourg station on RER line B

Abstract:

Cooperative Intelligent Transportation Systems (ITS) are systems where the vehicles, the roadside infrastructure, central control centres and other elements exchange information to achieve better road safety, traffic efficiency and comfort of the road users. These cooperative elements share the information by organising Vehicular Ad-hoc Network (VANET) or using the mobile network (e.g. 3G and LTE) and realises various cooperative ITS services. To promote wider deployment of such services, many stakeholders made consensus that the communication platform must be based on a common architecture and drive rapid international standardisation (e.g. ISO, ETSI, IEEE). In the near future, autonomous vehicles also benefit from such communication platforms by enforcing the perception, planing, and decision-making. However the most basic Vehicle-to-Vehicle (V2V) messages standardized in Europe, US and Japan suffer from same issues such as 1) unable to receive messages from the object without V2V transmitter, 2) message loss because of obstacle and wireless range, 3) vulnerable for malfunctioning and malicious node. For the solution, we propose an infrastructure-assisted V2V messaging system to support cooperative autonomous driving. We design the system based on the ITS Station architecture standardized in ISO/ETSI, working with any vehicle sensing technology. Moreover, we implement the prototype roadside system with a stereo vision for the vehicle sensing. The prototype system is evaluated in a field test in the campus of the University of Tokyo. The results show that the proposed system significantly improves the coverage of V2V messaging while the system overhead is limited. The proposal is being integrated to our personal mobility autonomous vehicle system based on an open source software. We are also planning the contribution to the international standardization of the technique.

The speaker:

Dr. Manabu Tsukada
Dr. Manabu Tsukada

Dr. Manabu Tsukada received his B.S. degree in Environmental and Information Studies from Keio University, Japan, in 2005 and his M.S. degree in Media and Governance from Keio University, Japan, in 2007. He worked in IMARA Team, INRIA, France during his Ph.D course and obtained his Ph.D. degree from Centre de Robotique, Mines ParisTech, France, in 2011. During his pre and postdoc research stages he has participated in multitude of international projects in the networked ITS area, such as GeoNet, ITSSv6, SCORE@F, CVIS, Nautilus6 or ANEMONE. He has been serving as board member of the WIDE Project since 2014. Regarding scientific dissemination, Dr. Tsukada has

published a number of conference and journal papers in the ITS and computer networks areas, and his appearance in IPv6 related events and initiatives is common. He is currently a project assistant professor at Graduate School of Information Science and Technology, the University of Tokyo, Japan. His research interests are mobility support for next generation Internet (IPv6) and communications for intelligent vehicles. http://www.hongo.wide.ad.jp/~tsukada/

 

Séminaire « Some autonomous vehicle issues with a Canadian flavor »

La Chaire Drive for All est heureuse de vous inviter au séminaire “Some autonomous vehicle issues with a Canadian flavor: from current project initiatives to testing, validation and challenges in artificial intelligence.” Nous avons l’honneur de recevoir le Prof. Denis Gingras, directeur du laboratoire sur les véhicules intelligent (LIV) à lUniversité de Sherbrooke, Québec, Canada. Le séminaire aura lieu Mardi 9 mai à MINES ParisTech de 15:00 à 16:00.

Le séminaire est public et gratuit mais l’inscription est obligatoire.

Plus d’informations sur le séminaire.

Dr Gingras has more than 35 years of a unique experience in the fields of information processing, AI and embedded intelligent systems applied to road transportation. In 1980, he was already working as a junior engineer at the BOSCH Research Center in Schwieberdingen-Stuttgart, Germany, on energy recovery dynamic braking control systems for electrical vehicles. In 1990, he founded and directed for nine years the Digital and Optical Systems Division at INO (Institut National d’Optique) in Quebec City, supervising the development of ITS technologies such as fiber optics weight sensors for trucks or intelligent video monitoring systems for

automated road traffic analysis. In 2000, Dr Gingras was appointed Director General of the Intelligent Materials and Systems Institute (IMSI) at Université de Sherbrooke and full professor at the Electrical Engineering and Computer Science Department. During that time, he was also head of the Canadian research program “Intelligent systems and sensors” in the federal AUTO21 Network of Centers of Excellence (NCE) for 15 years. From 2003 to 2008, he was Group Leader of the Intelligent Vehicles Group in the ATLANTIC International Network (Canada-US-Europe). He is a member of different boards and committees such as the board of the CAHR Council (Canadian Automotive Human Resources) or the board of INNOV-ÉÉ in Quebec. He is a member of the APMA Connected Vehicle Working Group and the ITS Canada Autonomous Vehicle Task Force. He was the instigator and Chair of the Canada-California Strategic Innovation Partnership-CCSIP- Committee on “Intelligent Vehicles and Transportation Systems”, allowing Canadian PhDs and postdoc students to complete internships in Californian automotive firms and universities. He was a member of advisory committees of the ANR in France (French Research Agency) on transportation and mobility from 2009 to 2011. He founded the Laboratory on Intelligent Vehicles (LIV) at Sherbrooke in 2012 involved in various R&D industrial projects related to intelligent vehicles (http://www.gel.usherbrooke.ca/LIV). He is a senior associate member of the Canadian Autonomous Vehicle Center of Excellence (CAVCOE) since 2013. Over the years, Pr. Gingras has been invited professor and a guest researcher in various universities across the world (France, US, Mexico, Australia). He was awarded in 2008 a Foreign Researcher Fellowship from the City of Paris, France, working at IFSTTAR at the Intelligent Vehicle Laboratory (LIVIC). As invited guest professor, lecturing a graduate course on intelligent vehicles at Cal Poly, San Luis Obispo, California in 2015. He was also visiting the CAOR at Mines Paristech in 2016. Throughout his career, Dr Gingras has published 2 books and has co-authored more than hundred publications, most of them related to the road transportation arena. He is regularly invited to give talks on the subject at various organizations, such as the Centre Jacques-Cartier in Lyon,or the Canadian Board of Insurance Companies. His research interests cover information processing, telecommunications, multi-sensor fusion and intelligent systems applied to automobile.

 

Visite à Shanghai Jiao Tong University

Les mécènes de la Chaire ont été invités à l’Université de Shanghai Jiao Tong pendant deux jours en avril 2017 pour rencontrer formellement les équipes participant aux travaux de la Chaire.

Au-delà de la revue des projets et s’assurer leur pertinence face aux enjeux fixés par la Chaire, cette rencontre était aussi destinée à intensifier les interactions entre les Mécènes et le Département d’Automatique  de l’Université de Shanghai Jiao Tong sur des sujets connexes.

Séminaire « Service-oriented Cooperation of Driverless Vehicles—from model, mechanism to simulation »

La Chaire Drive for All est heureuse de vous inviter à un séminaire sur « Service-oriented Cooperation of Driverless Vehicles—from model, mechanism to simulation » présenté par le Prof. Kailong Zhang de Northwestern Polytechnical University (Xi’an, China).

Ce séminaire aura lieu de 10h30 à 11h30 le mercredi 29 mars à MINES ParisTech.

Pour plus d’informations et pour l’inscription obligatoire

Prof. ZHANG Kailong is an associate professor in school of computer science and technology, Northwestern Polytechnical University (NPU), China, since 2010. He obtained his MS degree and PhD degree at 2003 and 2009 respectively. He has been connected with Networked Embedded Computing and Technology Laboratory in NPU since 2003. Now, he is also a researcher and the assistant director of Shaanxi Provincial Key Laboratory of Embedded System Technology.
He has been conducting studies on intelligent embedded computing and real-time intelligent systems, covering the architecture, adaptive real-time computing, intelligent mechanisms and design methods for autonomous Cyber-Physical Systems. From 2012 to 2014, he was a post-doctor with the CyberCars program of the Center of Robotics in Mines ParisTech, Paris, France, mainly working on the cooperative mechanisms of multiple driverless vehicles by using state-driven automatic control and vehicular ad-hoc networks. In the field, he has undertaken/participated several China national/provincial important research projects, such as NSF projects and NDRC project. He has published more than 60 academic papers on international journals and conferences, such as IEEE Trans on ITS, IEEE Communication Letters, and IEEE ITSC etc. In recent years, he has won several awards for his research, including a HP scholarship (2006), a Shaanxi Provincial first Prize award of Science and Technology (2006), and a Chinese Ten Youths Award in Embedded System Domain (2011).
Now, he also serves as the program committee member and reviewer to several international journals, typically including “IEEE Trans on ITS”, “Elsevier-Physica A: Statistical Mechanics and its Applications”, “Oxford-The Computer Journal”, “Elsevier- Ad Hoc Networks”, “Wiley-Security and Communication Networks”, “Chinese Journal of Computers”, “Chinese Acta Automatica Sinica”, and also to serial international conferences, IEEE PiCOM, IEEE ICESS, IEEE/ACIS ICIS etc.

Soutenance de thèse – Xiangjun QIAN

Xiangjun QIAN a plaisir de vous inviter à sa soutenance de thèse, intitulée : commande prédictive pour conduite autonome et coopérative. Pour un aperçu rapide, cette thèse parle la planification et le contrôle de voiture autonome (et coopératif).

La soutenance aura lieu le jeudi 15 décembre 2016 à 13h30 en salle L118 à l’École des Mines de Paris, 60 boulevard Saint Michel 75272, Paris Cedex 06. Cette dernière sera suivie d’un pot à l’espace Vendôme.

Composition du jury:

  • M. Sebastien GLASER, Laboratoire sur les interactions véhicules-infrastructure-conducteurs, IFSTTAR Rapporteur
  • M. Denis GILLET Ecole, Polytechnique Fédérale de Lausanne, Rapporteur
  • M. Christoph STILLER, Karlsruher Institut für Technologie, Examinateur
  • M. Philippe MARTINET, Ecole Centrale de Nantes Examinateur
  • M. Arnaud DE LA FORTELLE, MINES ParisTech, Examinateur
  • M. Fabien MOUTARDE, MINES ParisTech, Examinateur

Résumé:
La conduite autonome a attiré une attention croissante au cours des dernières décennies en raison de son potentiel impact socio-économique, notamment concernant la réduction du nombre d’accidents de la route et l’amélioration de l’efficacité du trafic. Grâce à l’effort de plusieurs instituts de recherche et d’entreprises, les véhicules autonomes ont déjà accumulé des dizaines de millions de kilomètres parcourus dans des conditions de circulation réelles. Cette thèse porte sur la conception d’une architecture de contrôle pour les véhicules autonomes et coopératifs dans l’optique de leur déploiement massif. La base commune des différentes architectures proposées dans cette thèse est le Contrôle-Commande Prédictif, reconnu pour son efficacité et sa polyvalence. Nous présentons tout d’abord une architecture classique de contrôle hiérarchique, qui utilise la commande prédictive pour planifier un déplacement (choix de trajectoire), puis déterminer un contrôle permettant de suivre cette trajectoire. Toutefois, comme nous le montrons dans un deuxième temps, cette architecture simple n’est pas capable de gérer certaines contraintes logiques, notamment liées aux règles de circulation et à l’existence de choix de trajectoires discrets. Nous proposons donc approche hybride de la commande prédictive, que nous utilisons pour développer une architecture de contrôle pour un véhicule autonome individuel. Nous étudions le problème de contrôler un ensemble de véhicules autonomes circulant en convoi, i.e. maintenir une formation prédéterminée sans intervention humaine. Pour ce faire, nous utilisons à nouveau une architecture hiérarchique basée sur la commande prédictive, composée d’un superviseur de convoi et de contrôleurs pour chaque véhicule individuel. Enfin, nous proposons encore une architecture pour le problème de coordonner un groupe de véhicules autonomes dans une intersection sans feux de circulation, en utilisant un contrôleur d’intersection et en adaptant les contrôleurs des véhicules individuels pour leur permettre de traverser l’intersection en toute sécurité.

Abstract:
Autonomous driving has been gaining more and more attention in the last decades, thanks to its positive social-economic impacts including the enhancement of traffic efficiency and the reduction of road accidents. A number of research institutes and companies have tested autonomous vehicles in traffic, accumulating tens of millions of kilometers traveled in autonomous driving. With the vision of massive deployment of autonomous vehicles, researchers have also started to envision cooperative strategies among autonomous vehicles. This thesis deals with the control architecture design of individual autonomous vehicles and cooperative autonomous vehicles. Model Predictive Control (MPC), thanks to its efficiency and versatility, is chosen as the building block for various control architectures proposed in this thesis. In more detail, this thesis first presents a classical hierarchical control architecture for individual vehicle control that decomposes the controller into a motion planner and a tracking controller, both using nonlinear MPC. In a second step, we analyze the inability of the proposed planner in handling logical constraints raised from traffic rules and multiple maneuver variants, and propose a hybrid MPC based motion planner that solves this issue. We then consider the convoy control problem of autonomous vehicles in which multiple vehicles maintain a formation during autonomous driving. A hierarchical formation control architecture is proposed composing of a convoy supervisor and local MPC based vehicle controllers. Finally, we consider the problem of coordinating a group of autonomous vehicles at an intersection without traffic lights. A hierarchical architecture composed of an intersection controller and multiple local vehicle controllers is proposed to allow vehicles to cross the intersection smoothly and safely

1er Prix pour le Valeo Innovation Challenge 2016

Une équipe d’étudiants de MINES ParisTech, composée de 4 doctorants du Centre de Robotique et d’un élève de dernière année du cycle ingénieur civil, a remporté un premier Valeo Innovation Challenge 2016 dans la catégorie « innovation technologique ». L’équipe reçoit ce prix après une compétition rassemblant plus de 1300 propositions.

dows_valeo_innovation_challenge
Jacques Aschenbroi remet le premier prix à l’équipe DOWS

Félicitations à Eva Coupeté, Florent Altché, Philip Polack, Sofiane Mahiou et Xiangjun Qian !

Tous les résultats sont disponibles sur le site du Valeo Challenge Innovation.