Web-Books
im Austria-Forum
Austria-Forum
Web-Books
Informatik
Joint Austrian Computer Vision and Robotics Workshop 2020
Seite - 157 -
  • Benutzer
  • Version
    • Vollversion
    • Textversion
  • Sprache
    • Deutsch
    • English - Englisch

Seite - 157 - in Joint Austrian Computer Vision and Robotics Workshop 2020

Bild der Seite - 157 -

Bild der Seite - 157 - in Joint Austrian Computer Vision and Robotics Workshop 2020

Text der Seite - 157 -

360°Monitoring forRobotsUsingTime-of-FlightSensors Thomas Maier,Birgit Hasenberger BECOM SystemsGmbH {thomas.maier,birgit.hasenberger}@becom-group.com Abstract. In this paper, we present a system based onmultiple Time-of-Flight (ToF) 3D sensors paired with a central processing hub for integration into robotsormobilemachines. This systemcanproduce a360°viewfromtherobot’sperspectiveandenables tasks ranging from navigation and obstacle avoid- ance tohuman-robot collaboration. 1. Introduction Today’s e-commerce growth and the paradigms of Industry 4.0 in the manufacturing space present new challenges for robotic systems [1]. In order to increase mobility and autonomy of such systems they need to gather and interpret as much informa- tionaspossiblefromtheirsurroundings. Approaches for collision avoidance with 1D time-of-flight sen- sors have been explored [2], the use of several high- resolutionsensorswouldenableapplications likehu- man pose estimation and gesture recognition as well asautomationtaskssuchashandlingofgoods. Close collaborationandadditional functionalitiesaremade possible with the setup proposed in this paper which consists of intrinsic 3D Time-of-Flight sensors that can cover360°arounda robot’s armor chassis. 2.Systemarchitecture Themulti-ToFplatform1 consistsofacentralpro- cessing module based on a NVIDIA Tegra TX2 pro- cessor (the hub) and multiple camera modules that work inparallel (the frontends). Thisplatformarchi- tecture allows for the integration of various sensor and camera types. TheToFsensorfrontendfor theplatformisacom- pact module designed for close-range detection. Ta- ble1summarises the technicaldataandperformance of two frontend variants. The frontend is connected 1https://www.becom-group.com/goto/multi-tof-platform Figure 1. Image of a front-end including descriptions of its components (left)anda ringof four front-ends (right). QVGA VGA frontend frontend Resolution 304pxx240px 640pxx480px Fieldofview 110°x82° 110°x82° Distance range 0.1–1.5m 0.1–2.0m Operating wavelength 850nm 940nm Framerate 40Hz 30Hz Table1.Frontend specifications. to the hub via FPD-Link III, using a cable that also provides thepowersupply. Four frontendscanbear- rangedasa ring toallowa360°coverage. The hub controls the frontends, performs calibra- tion and correction operations on the incoming data, andultimatelycalculatesdepthmapsorpointclouds. The data is transmitted from the hub via a Gigabit Ethernet connection and supports ROS to gather the individual data streams. The following operations areperformedon thehub: 1. Synchronizationand triggering 2. Acquisitionanddepthmapcalculation 3. Corrections (temperature, FPPN, distance off- set, intrinsicandextrinsic) 4. Filtering (spatial and temporal) 5. 3Dpoint cloudcalculation 6. Registrationand transformation 157
zurück zum  Buch Joint Austrian Computer Vision and Robotics Workshop 2020"
Joint Austrian Computer Vision and Robotics Workshop 2020
Titel
Joint Austrian Computer Vision and Robotics Workshop 2020
Herausgeber
Graz University of Technology
Ort
Graz
Datum
2020
Sprache
englisch
Lizenz
CC BY 4.0
ISBN
978-3-85125-752-6
Abmessungen
21.0 x 29.7 cm
Seiten
188
Kategorien
Informatik
Technik
Web-Books
Bibliothek
Datenschutz
Impressum
Austria-Forum
Austria-Forum
Web-Books
Joint Austrian Computer Vision and Robotics Workshop 2020