Project

Back to overview

Swarm of Flying Cameras

English title Swarm of Flying Cameras
Applicant Scaramuzza Davide
Number 143607
Funding scheme Project funding (Div. I-III)
Research institution Institut für Informatik Universität Zürich
Institution of higher education University of Zurich - ZH
Main discipline Information Technology
Start/End 01.04.2013 - 30.09.2016
Approved amount 367'584.00
Show all

Keywords (5)

Robotics; Simultaneous Localization and Mapping; Autonomous micro flying robots; Vision-based navigation; Computer vision

Lay Summary (Italian)

Lead
Nei primi dieci anni del 2000 si e’ sentito molto parlare di droni, ovvero velivoli autonomi senza pilota. Nonostante l’utilizzo di droni sia spesso associato ad usi militari, molteplici sono le applicazioni civili. La teconologia attuale e’ basata su sistemi di navigazione GPS. Poiche’ quest’ultimo e’ assente all’interno di edifici e molto debole e impreciso a ridosso di abitazioni e vegetazione, l’impiego di droni e’ attuallmente limitato a spazi aperti.
Lay summary

Soggetto e obiettivo

L’obiettivo di questo progetto e’ contribuire alla guida automatica di droni in ambienti interni o esterni in assenza di GPS. A tal scopo proponiamo lo studio di metodologie di navigazione basate su telecamere a bordo che permettono ai droni, mediante avanzati algoritmi di visione computazionale, di analizzare e riconstruore la scena in una maniera simile a quella utilizzata dagli esseri umani. Allo stato attuale esistono gia’ algoritmi di navigazione basati su visione computazionale, di cui il coordinatore di questo progetto e’ un precursore (vedasi il progetto Europeo www.sfly.org che lo stesso coordinatore ha diretto). Tuttavia, nell’ambito di questo progetto SNSF proponiamo di avanzare lo stato dell’arte nei seguenti punti:

(i) il nostro obbiettivo principale e’ estendere la teconologia attuale (ad oggi in grado di funzionare solo su un singolo velivolo) a molteplici velivoli in grado di coordinarsi tra loro;

(ii) il secondo obiettivo e’ implementare algoritmi di visione computazionale piu’ efficienti e accurati di quelli allo stato attuale, che possano funzionare su droni di dimensioni e peso ridotte (es. 150 grammi invece degli attuali 1-2 Kg). L’utilizzo di droni di piccole dimensioni garantisce maggior sicurezza rispetto a droni di media grandezza ma richiede di dotare i velivoli di computer di bordo con minore potenza di calcolo. Cio’ rende necessario aumentare l’efficienza computazionale degli algoritmi.

(iii) il terzo obiettivo e’ implementare tecniche di esplorazione automatiche, che permettano ai droni, per esempio, la miglior strategia per la ricerca e soccorso di vittime in ambienti a priori sconosciuti.

Contesto socio-scientifico

Il nostro lavoro permettera’ lo sviluppo di tecniche visione computazionali che possono essere applicate anche ad altri sistemi robotici, non necessariamente volanti. Sia istituti di ricerca che industrie potranno trarne vantaggio. Le possibili applicazioni sono molteplici:

  1. Ricerca e soccordo di vittime. I droni possono essre inviati all’interno di edifici colpiti da cataclismi ambientali per i quali l’intervento umanitario e’ a rischio. I robot costruirebbero una mappa interna dell’edificio che aiuterebbe i soccorsi a pianificare il loro intervento, individuando le aree a rischio di cedimento.
  2. Evacuazione di aree densamente affollate (monitoraggio di stadi, parate, manifestazioni ed eventi non programmati).
  3. Trasporto di medicinali in aree remote.
  4. Monitoraggio ambientale: monitoroaggio della qualita’ dell’aria, identificazione di incendi, etc.

Un recente studio del Ministero della Difesa francese ha stimato che nei prossimi 10 anni i droni saranno principalmente usati in settori quali polizia, protezione civile e monitoraggio ambientale.

Direct link to Lay Summary Last update: 15.05.2014

Responsible applicant and co-applicants

Employees

Publications

Publication
On-Manifold Preintegration for Real-Time Visual-Inertial Odometry
Forster Christian, Carlone Luca, Dellaert Frank, Scaramuzza Davide (2016), On-Manifold Preintegration for Real-Time Visual-Inertial Odometry, in IEEE Transactions on Robotics, PP(99), 1-21.
Air-ground Matching: Appearance-based GPS-denied Urban Localization of Micro Aerial Vehicles
Majdik Andras, Verda Damiano, Albers-Schoenberg Yves, Scaramuzza Davide (2015), Air-ground Matching: Appearance-based GPS-denied Urban Localization of Micro Aerial Vehicles, in Journal of Field Robotics, 32(7), 1015-1039.
Automatic Re-Initialization and Failure Recovery for Aggressive Flight with a Monocular Vision-Based Quadrotor
Faessler Matthias, Fontana Flavio, Forster Christian, Scaramuzza Davide (2015), Automatic Re-Initialization and Failure Recovery for Aggressive Flight with a Monocular Vision-Based Quadrotor, in IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015., IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015..
Autonomous, Vision-based Flight and Live Dense 3D Mapping with a Quadrotor Micro Aerial Vehicle
Faessler Matthias, Fontana Flavio, Forster Christian, Mueggler Elias, Pizzoli Matia, Scaramuzza Davide (2015), Autonomous, Vision-based Flight and Live Dense 3D Mapping with a Quadrotor Micro Aerial Vehicle, in Journal of Field Robotics, 33(4), 431-450.
Continuous On-Board Monocular-Vision-based Elevation Mapping Applied to Autonomous Landing of Micro Aerial Vehicles
Forster Christian, Faessler Matthias, Fontana Flavio, Werlberger Manuel, Scaramuzza Davide (2015), Continuous On-Board Monocular-Vision-based Elevation Mapping Applied to Autonomous Landing of Micro Aerial Vehicles, in IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015., IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015..
Continuous-Time Trajectory Estimation for Event-based Vision Sensors
Muegggler Elias, Gallego Guillermo, Scaramuzza Davide (2015), Continuous-Time Trajectory Estimation for Event-based Vision Sensors, in Robotics: Science and Systems (RSS), Rome, 2015., Robotics: Science and Systems (RSS), Rome, 2015..
IMU Preintegration on Manifold for Efficient Visual-Inertial Maximum-a-Posteriori Estimation
Forster Christian, Carlone Luca, Dellaert Frank, Scaramuzza Davide (2015), IMU Preintegration on Manifold for Efficient Visual-Inertial Maximum-a-Posteriori Estimation, in Robotics: Science and Systems (RSS), Rome, 2015., Robotics: Science and Systems (RSS), Rome, 2015..
Lifetime Estimation of Events from Dynamic Vision Sensors
Mueggler Elias, Forster Christian, Baumli Nathan, Gallego Guillermo, Scaramuzza Davide (2015), Lifetime Estimation of Events from Dynamic Vision Sensors, in IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015., IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015..
Low Computational-Complexity Algorithms for Vision-aided Inertial Navigation of Micro Aerial Vehicles
Troiani Chiara, Martinelli Agostino, Laugier Christian, Scaramuzza Davide (2015), Low Computational-Complexity Algorithms for Vision-aided Inertial Navigation of Micro Aerial Vehicles, in Robotics and Autonomous Systems, 69, 80-97.
Towards Evasive Maneuvers with Quadrotors using Dynamic Vision Sensors
Mueggler Elias, Baumli Nathan, Fontana Flavio, Scaramuzza Davide (2015), Towards Evasive Maneuvers with Quadrotors using Dynamic Vision Sensors, in European Conference on Mobile Robots (ECMR), Lincoln, 2015., European Conference on Mobile Robots (ECMR), Lincoln, 2015..
2-Point-based Outlier Rejection for Camera-IMU Systems with applications to Micro Aerial Vehicles
Troiani Chiara, Martinelli Agostino, Laugier Christian, Scaramuzza Davide (2014), 2-Point-based Outlier Rejection for Camera-IMU Systems with applications to Micro Aerial Vehicles, in IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014., IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014..
A Monocular Pose Estimation System based on Infrared LEDs
Faessler Matthias, Mueggler Elias, Schwabe Karl, Scaramuzza Davide (2014), A Monocular Pose Estimation System based on Infrared LEDs, in IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014., IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014..
Event-based, 6-DOF Pose Tracking for High-Speed Maneuvers
Mueggler Elias, Huber Basil, Scaramuzza Davide (2014), Event-based, 6-DOF Pose Tracking for High-Speed Maneuvers, in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Chicago, 2014., IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Chicago, 2014..
Low-Latency Event-Based Visual Odometry
Censi Andrea, Scaramuzza Davide (2014), Low-Latency Event-Based Visual Odometry, in IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014., IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014..
Micro Air Vehicle Localization and Position Tracking from Textured 3D Cadastral Models
Majdik Andras, Verda Damiano, Albers-Schoenberg Yves, Scaramuzza Davide (2014), Micro Air Vehicle Localization and Position Tracking from Textured 3D Cadastral Models, in IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014., IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014..
REMODE: Probabilistic, Monocular Dense Reconstruction in Real Time
Pizzoli Matia, Forster Christian, Scaramuzza Davide (2014), REMODE: Probabilistic, Monocular Dense Reconstruction in Real Time, in IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014., IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014..
SVO: Fast Semi-Direct Monocular Visual Odometry
Forster Christian, Pizzoli Matia, Scaramuzza Davide (2014), SVO: Fast Semi-Direct Monocular Visual Odometry, in IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014., IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014..
Air-Ground Localization and Map Augmentation Using Monocular Dense Reconstruction
Forster Christian, Pizzoli Matia, Scaramuzza Davide (2013), Air-Ground Localization and Map Augmentation Using Monocular Dense Reconstruction, in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013., IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013..
Collaborative Monocular SLAM with Multiple Micro Aerial Vehicles
Forster Christian, Lynen Simon, Kneip Laurent, Scaramuza Davide (2013), Collaborative Monocular SLAM with Multiple Micro Aerial Vehicles, in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013., IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013..
Low-latency localization by Active LED Markers tracking using a Dynamic Vision Sensor
Censi Andrea, Scaramuzza Davide (2013), Low-latency localization by Active LED Markers tracking using a Dynamic Vision Sensor, in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013., IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013..
MAV Urban Localization from Google Street View Data
Majdik Andras, Albers-Schoenberg Yves, Scaramuzza Davide (2013), MAV Urban Localization from Google Street View Data, in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013..
Appearance-based Active, Monocular, Dense Reconstruction for Micro Aerial Vehicles
Forster Christian, Pizzoli Matia, Scaramuzza Davide, Appearance-based Active, Monocular, Dense Reconstruction for Micro Aerial Vehicles, in Robotics: Science and Systems (RSS), Berkeley, 2014., Robotics: Science and Systems (RSS), Berkeley, 2014..
Perception-aware Path Planning
Costante Gabriele, Forster Christian, Delmerico Jeffrey, Valigi Paolo, Scaramuzza Davide, Perception-aware Path Planning, in IEEE Transactions on Robotics.
SVO: Semi-Direct Visual Odometry for Monocular and Multi-Camera Systems
Forster Christian, Zhang Zichao, Gassner Michael, Werlberger Manuel, Scaramuzza Davide, SVO: Semi-Direct Visual Odometry for Monocular and Multi-Camera Systems , in IEEE Transactions on Robotics.
The Event-Camera Dataset and Simulator: Event-based Data for Pose Estimation, Visual Odometry, and SLAM
Mueggler Elias, Rebecq Henri, Gallego Guillermo, Delbruck Tobi, Scaramuzza Davide, The Event-Camera Dataset and Simulator: Event-based Data for Pose Estimation, Visual Odometry, and SLAM, in International Journal on Robotics Research.
The Zurich Urban Micro Aerial Vehicle Dataset
Majdik Andras, Till Charles, Scaramuzza Davide, The Zurich Urban Micro Aerial Vehicle Dataset, in International Journal of Robotics Research.

Collaboration

Group / person Country
Types of collaboration
GeorgiaTech United States of America (North America)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel
DACUDA AG Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Research Infrastructure
- Exchange of personnel
- Industry/business/other use-inspired collaboration
Hungarian Academy of Sciences Hungary (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel
MIT United States of America (North America)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel
SENSEFLY Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Research Infrastructure
- Exchange of personnel
- Industry/business/other use-inspired collaboration
University of Perugia Italy (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel
Max Planck Institute Tuebingen Germany (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure
- Exchange of personnel
INRIA Grenoble France (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel

Scientific events

Active participation

Title Type of contribution Title of article or contribution Date Place Persons involved
IEEE IROS’16 Tutorial on Visual SLAM (keynote speaker) Talk given at a conference Visual SLAM tutorial 01.10.2016 Daejeon, Korean Republic (South Korea) Scaramuzza Davide;
IEEE Colombian Conference on Robotics and Automation, Bogota, September, 2016 (keynote speaker) Talk given at a conference Vision based navigation of flying robots 01.09.2016 Bogota, Colombia Scaramuzza Davide;
Universidad Pontificia de Bogota, September, 2016. Individual talk Vision based navigation of flying robots 01.09.2016 Bogota, Colombia Scaramuzza Davide;
20th Robocup Symposium, Leipzig, July 2016 (keynote speaker) Talk given at a conference Vision based navigation of flying robots 01.07.2016 Lepzig, Germany Scaramuzza Davide;
14th International Conference on Intelligent Systems, Shanghai, July, 2016 (plenary speaker) Talk given at a conference Vision based navigation of flying robots 01.07.2016 Shanghai, China Scaramuzza Davide;
SHERPA summer school, June 2016 Individual talk Visual SLAM tutorial 01.06.2016 Trento, Italy Scaramuzza Davide;
Machine Intelligence Summit, Berlin, June 2016 (keynote speaker) Talk given at a conference Vision based navigation of flying robots 01.06.2016 Berlin, Germany Scaramuzza Davide;
Tutorial on Micro Aerial Vehicles, Int. Conference on Robotics and Automation, ICRA16, Stockholm Talk given at a conference Visual SLAM tutorial 01.05.2016 Stockholm, Sweden Scaramuzza Davide;
SwissNex Sao Paulo (inviter Prof. Dr. Marcelo Becker), May 2016 Individual talk Vision based navigation of flying robots 01.05.2016 Sao Paulo, Brazil Scaramuzza Davide;
University of Sao Paulo, Brazilian Symposium on Robotics and Mechatronics, May, 2016 (plenary speaker) Talk given at a conference Vision based navigation of flying robots 01.05.2016 Sao Carlos, Brazil Scaramuzza Davide;
Dagstuhl School on Vision for Autonomous Vehicles, November, 2015 Individual talk Vision for robotics 01.11.2015 Dagstuhl, Germany Scaramuzza Davide;
European Conference on Mobile Robots (ECMR), Lincoln, 2015. Talk given at a conference Towards Evasive Maneuvers with Quadrotors using Dynamic Vision Sensors 01.10.2015 Lincoln, Great Britain and Northern Ireland Müggler Elias; Scaramuzza Davide; Fontana Flavio;
Qualcomm, Vienna, October. 2015 Individual talk Vision based navigation of flying robots 01.10.2015 Vienna, Austria Scaramuzza Davide;
International Workshop on Software Engineering for Resilient Systems, Paris, September, 2015 (keynote speaker) Talk given at a conference Vision based navigation of flying robots 01.09.2015 Paris, France Scaramuzza Davide;
Int. Conference on Intelligent Robotics and Applications, Portsmouth, UK, August, 2015 (plenary speaker) Talk given at a conference Vision-based navigation of flying robots 01.08.2015 Portsmouth, Great Britain and Northern Ireland Scaramuzza Davide;
Armasuisse UAV workshop, Thun, July, 2015 Talk given at a conference Swarm of vision controlled flying robots 01.07.2015 Thun, Switzerland Scaramuzza Davide;
Innorobo Workshop, Lyon, July, 2015. (keynote speaker) Talk given at a conference Towards Robust and Safe Autonomous Drones 01.07.2015 Lyon, France Scaramuzza Davide;
Robotics: Science and Systems (RSS), Rome, 2015. Talk given at a conference Continuous-Time Trajectory Estimation for Event-based Vision Sensors 01.06.2015 Rome, Italy Müggler Elias; Scaramuzza Davide;
Robotics: Science and Systems (RSS), Rome, 2015. Talk given at a conference IMU Preintegration on Manifold for Efficient Visual-Inertial Maximum-a-Posteriori Estimation 01.06.2015 Rome, Italy Scaramuzza Davide;
IEEE Italy Section Summer School, Perugia, June, 2015. Individual talk Visual SLAM Tutorial 01.06.2015 Perugia, Italy Scaramuzza Davide;
MIT, (inviter Andrea Censi) May, 2015 Individual talk Vision based navigation of flying robots 01.05.2015 Boston, United States of America Scaramuzza Davide;
Armasuisse Workshop on Multi-agent systems, Thun, May 2015. Talk given at a conference Swarm of vision controlled flying robots 01.05.2015 Thun, Switzerland Scaramuzza Davide;
International Workshop on “Scaling up active perception”, ICRA’15, Seattle, May, 2015 Talk given at a conference Active vision 01.05.2015 Seattle, United States of America Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015. Talk given at a conference Lifetime Estimation of Events from Dynamic Vision Sensors 01.05.2015 Seattle, United States of America Müggler Elias; Scaramuzza Davide;
Int. Workshop on “The problem of mobile sensors: setting future goals and indicators for SLAM", RSS’15, May, 2015 Talk given at a conference Direct methods for Visual SLAM 01.05.2015 Rome, Italy Scaramuzza Davide;
International Workshop on Taxonomy of Interconnected Systems, ICRA’15, Seattle, May, 2015 Talk given at a conference Swarm of vision controlled flying robots 01.05.2015 Seattle, United States of America Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015. Talk given at a conference Continuous On-Board Monocular-Vision-based Elevation Mapping Applied to Autonomous Landing of Micro Aerial Vehicles 01.05.2015 Seattle, United States of America Scaramuzza Davide; Fässler Matthias;
IEEE International Conference on Robotics and Automation (ICRA), Seattle, 2015. Talk given at a conference Automatic Re-Initialization and Failure Recovery for Aggressive Flight with a Monocular Vision-Based Quadrotor 01.05.2015 Seattle, United States of America Fässler Matthias; Scaramuzza Davide;
Indian Institute of Technology at New Delhi, Allahabad, and Kanpur (inviter SwissNex India), March 2015. Individual talk Vision based navigation of flying robots 01.03.2015 Kanpur, India Scaramuzza Davide;
Computer vision Winter workshop, Graz, February 2015 (keynote speaker) Talk given at a conference Vision-based navigation of flying robots 01.02.2015 Graz, Austria Scaramuzza Davide;
Workshop on Future UAS Technologies, Thun, Switzerland, Jan. 2015 Talk given at a conference Swarm of vision controlled flying robots 01.01.2015 Thun, Switzerland Scaramuzza Davide;
University of Zaragoza (inviter Prof. Jose Neira), Dec. 2014 Individual talk Vision based navigation of flying robots 01.12.2014 Zaragoza, Spain Scaramuzza Davide;
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Chicago, 2014. Talk given at a conference Event-based, 6-DOF Pose Tracking for High-Speed Maneuvers 01.10.2014 Chicago, United States of America Scaramuzza Davide; Müggler Elias;
Int. Workshop on Aerial Open Source Robotics (IROS’14), Chicago, Sep. 2014 Talk given at a conference Open sourse softare for vision based navigation of flying robots 01.10.2014 Chicago, United States of America Müggler Elias;
Int. Summer School on Cooperation of Robots and Sensor Networks for Emergency Response, TU Darmstadt, Sep. 2014 (keynote speaker) Individual talk Computer vision algorithms for mobile robotics 01.09.2014 Darmstadt, Germany Scaramuzza Davide;
3rd Int. Workshop on Visual Control of Mobile Robots (IROS’14), Chicago, Sep. 2014 Talk given at a conference Vision-based navigation of flying robots 01.09.2014 Chicago, United States of America Müggler Elias;
Int. Workshop on Transducers for robot autonomous navigation, Benevento, Italy, Sep. 2014 (keynote speaker) Talk given at a conference Vision-based navigation of flying robots 01.09.2014 Benevento, Italy Scaramuzza Davide;
University of Naples Federico II (inviter Prof. Bruno Siciliano), Sep, 2014 Individual talk Vision based navigation of flying robots 01.09.2014 Naples, Italy Scaramuzza Davide;
Advanced Center for Aerospace Technologies and University of Seville (inviter Prof. Anibal Ollero), September 2013. Individual talk Vision based navigation of flying robots 01.09.2014 Seville, Spain Scaramuzza Davide;
ETH Computer Vision Lab (inviter Prof. Luc Van Gool), August, 2014 Individual talk Vision based navigation of flying robots 01.08.2014 Zurich, Switzerland Scaramuzza Davide;
International Data Fusion and Big Data Workshop, Thun, ArmaSuisse, Switzerland, August 2014 Talk given at a conference Vision-based navigation of flying robots 01.08.2014 Thun, Switzerland Scaramuzza Davide;
2nd International workshop on Multi-Unmanned Vehicle Systems (keynote speaker), Compiegne, France, July, 2014 Talk given at a conference Vision-based navigation of flying robots 01.07.2014 Compiegne, France Scaramuzza Davide;
Carnegie Mellon University (inviter Prof. Sanjiv Singh), July 2014. Individual talk Vision based navigation of flying robots 01.07.2014 Pittsburgh, United States of America Scaramuzza Davide;
Polytechnique of Torino (inviter Prof. Marcello Chiamberge), July 2014 Individual talk Vision based navigation of flying robots 01.07.2014 Torino, Italy Scaramuzza Davide;
3rd Workshop on Resource-efficient Integration of Planning and Perception for true autonomous operation of Micro Air Vehicles (MAVs) (RSS’14), Berkley, July, 2014 (keynote speaker) Talk given at a conference Vision-based navigation of flying robots 01.06.2014 Berkley, United States of America Scaramuzza Davide;
Robotics: Science and Systems (RSS), Berkeley, 2014. Talk given at a conference Appearance-based Active, Monocular, Dense Reconstruction for Micro Aerial Vehicles 01.06.2014 Hong Kong, China Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014. Talk given at a conference A Monocular Pose Estimation System based on Infrared LEDs 01.05.2014 Hong Kong, China Müggler Elias; Fässler Matthias; Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014. Talk given at a conference Low-Latency Event-Based Visual Odometry 01.05.2014 Hong Kong, China Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014. Talk given at a conference SVO: Fast Semi-Direct Monocular Visual Odometry 01.05.2014 Hong Kong, China Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014. Talk given at a conference REMODE: Probabilistic, Monocular Dense Reconstruction in Real Time 01.05.2014 Hong Kong, China Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014. Talk given at a conference Micro Air Vehicle Localization and Position Tracking from Textured 3D Cadastral Models 01.05.2014 Hong Kong, China Scaramuzza Davide;
IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014. Talk given at a conference 2-Point-based Outlier Rejection for Camera-IMU Systems with applications to Micro Aerial Vehicles 01.05.2014 Hong Kong, China Scaramuzza Davide;
International Workshop on Robotic Sensor Networks (keynote speaker), Berlin, April, 2014 Talk given at a conference Vision-based navigation of flying robots 01.04.2014 Berlin, Germany Scaramuzza Davide;
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013. Talk given at a conference MAV Urban Localization from Google Street View Data 03.11.2013 Tokyo, Japan Scaramuzza Davide;
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013. Talk given at a conference Low-latency localization by Active LED Markers tracking using a Dynamic Vision Sensor 01.10.2013 Tokyo, Japan Scaramuzza Davide;
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013. Talk given at a conference Collaborative Monocular SLAM with Multiple Micro Aerial Vehicles 01.10.2013 Tokyo, Japan Scaramuzza Davide;
5th Workshop on Planning, Perception and Navigation for Intelligent Vehicles (IROS’13), (keynote speaker) Talk given at a conference Vision-based navigation of flying robots 01.10.2013 Tokyo, Japan Scaramuzza Davide;
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, 2013. Talk given at a conference Air-Ground Localization and Map Augmentation Using Monocular Dense Reconstruction 01.10.2013 Tokyo, Japan Scaramuzza Davide;
French National Robotics Days (Journees Nationales de la Recherche en Robotique), Annecy, Oct., 2013 Talk given at a conference Vision-based navigation of flying robots 01.10.2013 annecy, France Scaramuzza Davide;
4th JRC ECML Workshop on Unmanned Aerial Systems for Rapid Mapping, Geneva, September, 2013 Talk given at a conference Vision-based navigation of flying robots 01.09.2013 Geneva, Switzerland Scaramuzza Davide;
Workshop on “Towards a Swiss Robotic Rescue Team” (keynote speaker), (ICRA’13), 2013 Talk given at a conference Vision-based navigation of ground and aerial robots 01.05.2013 Karlsruhe, Germany Scaramuzza Davide;


Self-organised

Title Date Place
“Vision-based High Speed Autonomous Navigation of UAVs”, IROS, Daejeon, South Korea, October, 2016 01.10.2016 Daejeon, Korean Republic (South Korea)
“Vision-based Control and Navigation of Small, Light-Weight UAVs”, IROS, Hamburg, Germany, October, 2015. 01.10.2015 Hamburg, Germany
International Workshop on Neuromorphic Sensing, ICRA’15, Seattle, May, 2015 01.05.2015 Seattle, United States of America
“Computer Vision in Vehicle Technology with Special Session on Micro Aerial Vehicles”, ECCV, Zurich, Switzerland, 2014 01.09.2014 Zurich, Switzerland
12th Int. Workshop on Non-classical Cameras, Camera Networks and Omnid. Vision (ICRA'14), Hong Kong, June, 2014. 01.06.2014 Hong Kong, China
“Vision-based Control and Navigation of Micro Helicopters in GPS-denied Environments”, IROS, Tokyo, Japan, October, 2013 01.10.2013 Tokyo, Japan

Communication with the public

Communication Title Media Place Year
Talks/events/exhibitions ICRA’15, Seattle, 27.06.2015, 150 visitors International 2016
Talks/events/exhibitions IROS’15, Hamburg, 02.10.2015, 150 visitors International 2016
Talks/events/exhibitions NCCR Industry Day, Nov. 2015, 500 visitors Western Switzerland 2016
Media relations: print media, online media Was können Roboter heute eigentlich alles? Blick German-speaking Switzerland 2016
New media (web, blogs, podcasts, news feeds etc.) Aerial-guided navigation of the youBot KUKA YouBot Newsletter International 2015
Media relations: print media, online media Autonomous Vision-Controlled Flying Robots EB-Navi German-speaking Switzerland 2015
Talks/events/exhibitions CeBIT Expo, Hannover, 16-20.03.2015, 30,000 visitors International 2015
Media relations: radio, television Die autonome Drohne ist da! Technische Rundschau German-speaking Switzerland 2015
Media relations: radio, television Drones Under Control BBC News International 2015
New media (web, blogs, podcasts, news feeds etc.) Embodied Quadrotors Robohub International 2015
New media (web, blogs, podcasts, news feeds etc.) Flying with Drones UZH News German-speaking Switzerland 2015
Media relations: print media, online media Forscher rüsten Drohnen zu «Seiltänzern Blick German-speaking Switzerland 2015
Media relations: print media, online media Forscher rüsten Drohnen zu Luftakrobaten auf 20 minuten German-speaking Switzerland 2015
New media (web, blogs, podcasts, news feeds etc.) Full-safe drone De Ingenieur International 2015
New media (web, blogs, podcasts, news feeds etc.) Ground-Flight collaboration Robohub International 2015
Media relations: print media, online media Just toss it in the air and this drone will take off, land itself Bangalore Mirror International 2015
Media relations: print media, online media Mit der Drohne über die Teppich-Stadt Tages- Anzeiger German-speaking Switzerland 2015
New media (web, blogs, podcasts, news feeds etc.) New technology keeps drones from crashing and burning Engadget International 2015
New media (web, blogs, podcasts, news feeds etc.) New technology makes drones safer, smarter The Times of India International 2015
New media (web, blogs, podcasts, news feeds etc.) New Technology Making Drones Safer and Smarter Scientific Computing International 2015
New media (web, blogs, podcasts, news feeds etc.) Palm-sized machines and self-flying vehicles go on display in Germany DailyMail International 2015
Media relations: radio, television Post aus der Luft SRF 10vor10 German-speaking Switzerland 2015
New media (web, blogs, podcasts, news feeds etc.) Quadrotor automatically recovers from failure or aggressive launch, without GPS Robohub International 2015
Talks/events/exhibitions Scientifica, Zurich, Sep. 2015, 20,000 visitors German-speaking Switzerland 2015
New media (web, blogs, podcasts, news feeds etc.) Swiss Team Develops Self-Stabilizing Drones Robotics Business Review International 2015
Talks/events/exhibitions SwissCore, Brussels, 12.05.2015, 100 visitors International 2015
New media (web, blogs, podcasts, news feeds etc.) System allows hampered drones to regain control and land safely Gizmag International 2015
New media (web, blogs, podcasts, news feeds etc.) Un drone qui se stabilise comme un funambule Futura Sciences International 2015
New media (web, blogs, podcasts, news feeds etc.) Wenn Drohnen mit eigenen Augen fliegen Futurezone International 2015
New media (web, blogs, podcasts, news feeds etc.) You Can Launch This Quadrotor by Throwing It in the Air IEEE Spectrum International 2015
New media (web, blogs, podcasts, news feeds etc.) Zürcher Forscher macht Drohnen zu 'Seiltänzern' Moneycab International 2015
New media (web, blogs, podcasts, news feeds etc.) Zürcher Forscher rüsten Drohnen zu Luftakrobaten auf inside-it.ch German-speaking Switzerland 2015
Talks/events/exhibitions AUTOMATICA, Munich, 3.-6.06.2014, 31,000 visitors [ Link ] [ Video ] International 2014
Media relations: print media, online media Die schlauste Drohne der Welt kommt aus Zürich SonntagsBlick German-speaking Switzerland 2014
Media relations: print media, online media Drohnen: Die Vielfalt scheint unbegrenzt Neue Luzerner Zeitung German-speaking Switzerland 2014
New media (web, blogs, podcasts, news feeds etc.) Dynamic Vision Sensors Enable High-Speed Maneuvers With Robots IEEE Spectrum News International 2014
New media (web, blogs, podcasts, news feeds etc.) Erster Preis für Zürcher Flugroboter UZH News German-speaking Switzerland 2014
New media (web, blogs, podcasts, news feeds etc.) Quadrocopter steuert aus der Luft Robotereinsatz Augsburger Allgemeine International 2014
New media (web, blogs, podcasts, news feeds etc.) Research Group from University of Zurich wins KUKA Innovation Award IEEE Spectrum News International 2014
New media (web, blogs, podcasts, news feeds etc.) Think fast, robot MIT News International 2014
Media relations: print media, online media Der Spion von deinem Fenster WeltWoche German-speaking Switzerland 2013
Media relations: radio, television Des drones pour sauver des vies RTS Western Switzerland 2013
Media relations: print media, online media Die fliegenden Kuriere Die Zeit International 2013
Media relations: radio, television Drohnen - Nützliches Werkzeug oder tödliche Waffe? ARTE International 2013
Media relations: radio, television Für den Kriseneinsatz: Uni Zürich zeigt Roboter-Duo Swiss Radio SRF 1 German-speaking Switzerland 2013
Talks/events/exhibitions Robotic Festival Lausanne, EPFL, 20 April, 2013, 10,000 visitors Western Switzerland 2013
Talks/events/exhibitions Robots on Tour German-speaking Switzerland 2013
New media (web, blogs, podcasts, news feeds etc.) Robots Save Easter After Negligent Bunny Makes A Mess Of Everything Gizmodo International 2013
Talks/events/exhibitions Scientifica, Zurich, 31-01 September, 20,000 visitors, 2013 International 2013
Talks/events/exhibitions TunZurich, Zurich, 23-25 October, 2013, 1,000 visitors German-speaking Switzerland 2013
New media (web, blogs, podcasts, news feeds etc.) UZH Wishes Us All a Happy Robot Easter IEEE Spectrum News International 2013

Awards

Title Year
AAAI Best Video Award Finalist 2016
Ellias Mueggler wins Qualcomm Innovation Award 2016
Best Paper Award Finalist at RSS’15 2015
"IEEE Robotics and Automation Early Career Award (academia) to Davide Scaramuzza For his major contributions to robot vision and visually-guided micro aerial vehicles". 2014
Basil Huber, master student supervised by Elias Mueggler and Davide Scaramuzza wins Fritz-Kutter ETH Prize (10k CHF) – Best Master thesis in Industry-related Computer Science at a Swiss University. Thesis title: High-Speed Pose Estimation using a Dynamic Vision Sensor. Article published in IROS'14. 2014
Google Faculty Research Award 4 awards were assigned by Google worldwide in the area of robotics; mine was the only one outside United States. 2014
KUKA Innovation Award (worldwide competition, 20,000 EUR) 2014

Use-inspired outputs


Start-ups

Name Year
Zurich-Eye 2015

Associated projects

Number Title Start Funding scheme
155841 Low-latency, Vision-based Agile Quadrotor Flight 01.11.2015 SNSF Starting Grants

Abstract

Autonomous micro helicopters are about to play a major role in tasks like search and rescue, environment monitoring, security surveillance, inspection, etc. If they are further realized in small scale, they can also be used in narrow outdoor and indoor environments and represent only a limited risk for people. However, for such operations, navigating based on GPS information only is not sufficient. Fully autonomous operation in urban environments and indoor spaces requires micro helicopters to rely on alternative localization systems. In fact, even in outdoor environments GPS is not a reliable service and is completely unavailable indoors. One way to overcome this problem is to use motion-capture systems, whose robustness and reliability have recently made micro helicopters able to perform fast maneuvers and even build small architectures (e.g., see the works of Kumar at the University of Pennsylvania and D’Andrea at ETH Zurich). However, these systems are limited to confined spaces and need pre-installation of the cameras, making it impossible to navigate autonomously in unknown, yet-unexplored environments. An alternative solution is to use onboard laser scanners to simultaneously build a map of the environment and navigate through it. However, 2D laser scanners are still heavy; this constrains the vehicle’s minimum weight and size, increasing the risk for people. Therefore, for very small lightweight helicopters, the only viable solution is to use a combination of onboard cameras and inertial sensors. Previous work of the applicant has demonstrated successful autonomous basic maneuvers of a micro helicopter using only a single camera and an inertial sensor onboard. Although today’s cameras can have high frame rate, image processing is still limited by the onboard-computer capabilities. As a result, vision-based autonomous maneuvers of micro helicopters are still very slow compared to those attainable with motion capture systems. Additionally, previous works have investigated visual navigation with a single helicopter only. However, there is an interest in using multiple platforms: (i) they can explore, monitor, and map an environment faster and more efficiently than with a single robot; (ii) each robot can share its capabilities or benefit from those of the other team members. However, most works on multi-robot navigation with micro helicopters have relied on motion-capture systems, which allow localizing the robots with millimeter precision and in a common, global coordinate frame. These conditions cannot be achieved with onboard sensors and, notably, with vision, where the robot-localization accuracy is strongly linked to the structure of the perceived environment.The purpose of this project is to focus on the navigation of multiple micro helicopters in GPS-denied environments (both indoors and outdoors) and on allowing fast, dynamic maneuvers using onboard monocular vision as the main sensor modality. The choice of monocular vision over stereo vision is dictated by the ultra-light weight of the platform that will be used. Due to the use of cameras as the main sensors, to the reduced size of the helicopters, and to the emphasis on multi-robot navigation, we call this system “a swarm of flying cameras.” The main challenges deal with the limited calculation power, the limitations of monocular images, and the highly dynamic motion of the helicopters. The overall goal of the project will be achieved through the interplay of the following subgoals:1.Fast visual navigation of micro helicopters.2.Multi-robot vision-based localization and mapping.3.Vision-based cooperative exploration.Task 1 aims at enabling fast maneuvers of micro helicopters with limited onboard computing capabilities. To this end, we will investigate methods for fast feature extraction and tracking and local relative-motion estimation to be done on lightweight, embedded computers (e.g., Gumstix or cell-phone GPUs). In task 2, we will investigate methods to estimate both the 3D map of the environment and the poses of each helicopter in a global reference. Most works in multi-robot localization and mapping have been done using laser scanners and for ground mobile robots (i.e., moving in a 2D environments). With flying robots this is more challenging due to the unconstrained motion of the vehicles and due to the lack of range information with monocular vision. Task 3 will study the problem of vision-based cooperative exploration. Again, most works in cooperative exploration have been done for ground mobile robots using laser rangefinders. We will combine the results achieved in the first two tasks and will investigate how to simultaneously compute the next robots’ poses in order to best reconstruct the environment.The achievement of these goals require expertise in computer vision, robotics, and micro-aerial-vehicle control that the applicant has earned throughout his past research both as a researcher (at EPFL, ETH and the University of Pennsylvania), as a team leader, and as the coordinator of a large European project on micro-aerial vehicles (SFLY). This project will also benefit from the ongoing collaboration that the applicant maintains with the Autonomous Systems Lab at ETH (where the applicant did his PhD and first post-doc) and with the GRASP Lab at the University of Pennsylvania (where the applicant did his second post-doc).
-