Titel
Accueil
Navigation principale
Contenu
Recherche
Aide
Fonte
Standard
Gras
Identifiant
Interrompre la session?
Une session sous le nom de
InternetUser
est en cours.
Souhaitez-vous vraiment vous déconnecter?
Interrompre la session?
Une session sous le nom de
InternetUser
est en cours.
Souhaitez-vous vraiment vous déconnecter?
Accueil
Plus de données
Partenaires
Aide
Mentions légales
D
F
E
La recherche est en cours.
Interrompre la recherche
Recherche de projets
Projet actuel
Projets récents
Graphiques
Identifiant
Titel
Titel
Unité de recherche
INNOSUISSE
Numéro de projet
11116.1;6 PFNM-NM
Titre du projet
Vision-based autopilot for mini-drones (Feasibility Study)
Titre du projet anglais
Vision-based autopilot for mini-drones (Feasibility Study)
Données de base
Textes
Participants
Titel
Textes relatifs à ce projet
Allemand
Français
Italien
Anglais
Description succincte
-
-
Résumé des résultats (Abstract)
-
-
Textes saisis
Catégorie
Texte
Description succincte
(Anglais)
Vision-based autopilot for mini-drones (Feasibility Study)
Description succincte
(Français)
Vision-based autopilot for mini-drones (Feasibility Study)
Résumé des résultats (Abstract)
(Anglais)
This project aims at implementing and testing a novel autopilot for mini-drones to enable them to fly autonomously in cluttered environments by actively avoiding collisions with obstacles such as buildings. The basis of this project is a patent-pending technology recently developed at EPFL, which takes inspiration from flying insects to devise an efficient vision-based control strategy that maps visual signals (optic flow) into commands in real-time. However, the current research prototype works only with natural obstacles. This feasibility study aims at assessing the use of a CSEM vision chip to improve signal quality and enable flight in man-made environments, which would significantly increase its commercial potential.
Résumé des résultats (Abstract)
(Français)
This project aims at implementing and testing a novel autopilot for mini-drones to enable them to fly autonomously in cluttered environments by actively avoiding collisions with obstacles such as buildings. The basis of this project is a patent-pending technology recently developed at EPFL, which takes inspiration from flying insects to devise an efficient vision-based control strategy that maps visual signals (optic flow) into commands in real-time. However, the current research prototype works only with natural obstacles. This feasibility study aims at assessing the use of a CSEM vision chip to improve signal quality and enable flight in man-made environments, which would significantly increase its commercial potential.
SEFRI
- Einsteinstrasse 2 - 3003 Berne -
Mentions légales