Search for a funded project
human inspired autonomous navigation in crowds – Hianic
The Hianic project aims to give autonomous vehicule the ability to spaces shared with pedestrians. In these shared spaces, there is no traffic lights or stops to give the priority to pedestrians or cars, and these spaces can be densely populated. The Hianic project proposes to combine works in crowd simulation, proactive navigation, psychological study on the acceptability, experiments with real autonomous vehicles to move forward on this challenging topic.
Visual semantic analysis: toward a smarter urban navigation – AVISE
Toward a safer navigation
Site-specific weeding robot using high electrical voltage actuator combined with UAV hyperspectral imagery for predictive management and post-assessment. – WeedElec2017
The WeedElec project offers an alternative to global chemical weeding, which combines an aerial adventitious detection capability with a terrestrial robotic weeding solution using high-voltage electric power
Sequential Art Image Laboratory – SAIL
analysis, knowledge representation and artificial intelligence. With the ultimate goal of obtaining a complete
ROBOTIC BIOPSY OF THE UTERUS STANDARDIZED TECHNIQUE – ROBUST
The basic idea is to provide the surgeon with intelligent tools that help him/her only when necessary while
Pesticide Free Robotized Pest Control in Agriculture – GREENSHIELD
Dans le cadre de la réduction de l’utilisation des pesticides, le projet Greenshield a pour objectif de développer un robot mobile pour combattre les ravageurs, en se basant sur l’intelligence artificielle pour la détection et un laser pour la destruction
Optimization of Urban Synchromodal Systems – OPUSS
Last mile delivery concerns the last stage of a transport system, from wholesale transport to the delivery to final customers. Because of its synchromodal aspect (several modes of transport subject to synchronization constraints), it is the nightmare of logisticians, especially in urban areas. This project aims at designing advanced optimization algorithms to plan transports in such synchromodal networks.
Notifying Memories for Dynamic Data-Flow Application – Nooman
The current architecture of computers where memory is a servile component subject to processor requests imposes constant and energy-consuming transfers. However, many of these requests are unnecessary. My goal is to bring intelligence to the memory to delete these requests because they are useless and halve the energy consumed, which will have a major impact in the context of a growing number of digital devices and the goal of a carbon-neutral society.
No conventional imagery for secure urban mobility – ICUB
Polarimetric imaging characterizes the reflection of the light. The use of this feature will allow us to go beyond the problems related to only pattern recognition. The light-material interaction and the relationship between polarization, reflection and fog will be a major asset to address the problem of road scene analysis in reduced visibility conditions . In addition, deep neural networks have shown their superiority for obstacle detection over conventional methods.
Network for studying Entrainment and microPHysics of cLouds using Adaptive Exploration – NEPHELAE
that coordinate themselves in real-time as an intelligent network to address long standing questions in [...] goals of Défi 1 / Orientation 1 by creating an intelligent network to monitor the earth system with an