Praktikant: Deep Learning / Computer Vision for Bin Picking

*Ausschreibung für die Fachrichtungen*

  • Automatisierungstechnik
  • Elektrotechnik
  • Informatik
  • Mathematik
  • Mechatronik
  • Regelungstechnik
  • Softwaredesign
  • Softwareengineering

Was Sie mitbringen

  • Self-reliance, motivation, eager to accept challenges
  • Very good English language skills
  • Programming skills in Python and C++
  • Knowledge in the fields of machine learning and 3D image processing helpful
  • For internship applicants: certificate of mandatory intern-ship from university

Was Sie erwarten können

Bin Picking has most of the industrial applications. In most of the situations, the work pieces are placed in chaotic boxes and they need to be picked up by robot. For correctly picking of work piece, the robot must know the exact position and orientation (6D Pose) of the work piece in box. Here, the vision system comes into play: A laser scanner generates the 3D point cloud of the region under investigation. A vision algorithm then identifies the region of point cloud belonging to the box and the regions belonging to the work piece (semantic segmentation). Another vision algorithm then estimates the 6D pose of work piece that is used to pickup the workpiece by the robot.

Your tasks:

- Literature review of state-of-the-art approaches
- Generation of 3D point cloud data from real robot cell.
- Labelling and preparation of data.
- Generation of data from simulations (V-rep).
- Implementation of deep learning / computer vision algorithm for semantic segmentation, object detection, pose estimation.
- Evaluation of experiments on real and simulated data.
Die Vergütung richtet sich nach den Richtlinien des Bundes über Praktikantenvergütungen.

Fraunhofer ist die größte Organisation für anwendungsorientierte Forschung in Europa. Unsere Forschungsfelder richten sich nach den Bedürfnissen der Menschen: Gesundheit, Sicherheit, Kommunikation, Mobilität, Energie und Umwelt. Wir sind kreativ, wir gestalten Technik, wir entwerfen Produkte, wir verbessern Verfahren, wir eröffnen neue Wege.

Herr Dominik Kicherer
Telefon: +49 711 970-1130

Kennziffer: IPA-2018-232 Bewerbungsfrist: -
counter-image