Direkt zum Inhalt springen
Intelligent Autonomous Systems
TUM School of Computation, Information and Technology
Technical University of Munich

Technical University of Munich



Topic Prepositions

  • Re-design of programming of the robots (re-usable code, lisp tutorials, have robots do the demos _every_ day to generate the data)
  • UIMA perception and UIMA knowledge (Nico and Moritz know more)
  • Semantic mapping system that will run automatically and on the fly
  • Natural language processing
  • 2nd generation of our main software stacks (KnowRob, CRAM, ProbCog, Mapping)

Results of Discussion

Perception (Dejan, Martin, Ferenc, Florian, Hartmut, Hendrik)
  • Hough-voting based object recognition using furniture_classification framework - Dejan, Martin, Vlad
  • Binarisation of 3D features (e.g. PFH, SHOT) in order to speed up the feature build-up step - Dejan, Martin, Ferenc, Zoltan
  • Interactive segmentation and estimation of push point formulated as constraint satisfaction problem (e.g. Waltz algorithm) - Hartmut needs data
  • Interactive segmentation and estimation of push point using machine learning algorithms - Hartmut needs data
KnowRob (Moritz, ...)
ProbCog (Dominik, ...)
UIMA (Nico, Thomas W. ...)
CRAM (Lorenz, ...)

Rechte Seite