direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Diese Seite listet Software, Web Services und Tutorials, die im Rahmen der Forschung am RBO Lab entwickelt werden und hier öffentlich zugänglich gemacht werden.

Code

Online Interactive Perception

Our code of our perceptual system for Online Interactive Perception of articulated objects. It includes our methods presented in at IROS14 [1] and ICRA16 [2]. The system extracts patterns of motion at different levels (point feature motion, rigid body motion, kinematic structure motion) and infers the kinematic structure and state of the interacted articulated objects. It uses an RGB-D stream of an interaction as input. Optionally, it can reconstruct the shape of the moving parts and use it to improve tracking. More info at omip wiki [3].

Repository: https://github.com/tu-rbo/omip [4] and https://github.com/tu-rbo/omip_msgs [5]

Accompanies papers: Online Interactive Perception of Articulated Objects with Multi-Level Recursive Estimation Based on Task-Specific Priors [6] and An Integrated Approach to Visual Perception of Articulated Objects [7]

Maintainers: Roberto Martín-Martín

Project: Interactive Perception [8]

Learning State Representations with Robotic Priors

This simple python script contains the essence of our work on learning state representations with robotic priors and complements the paper to provide sufficient detail for reproducing our results and for reusing the method in other research while minimizing code overhead.

Repository: github.com/tu-rbo/learning-state-representations-with-robotic-priors [9]

Accompanies paper: Learning State Representations with Robotic Priors [10]

Maintainers: Rico Jonschkowski

Project: Representation Learning [11]

Probabilistic Multi-Class Segmentation for the Amazon Picking Challenge

This repository includes the code and data for the object perception method of our winning entry to the 2015 Amazon Picking Challenge. The object segmentation method and the analysis from the accompaning paper are implemented in python. The data can be used to develop and benchmark new perception methods in this setting.

Repository: gitlab.tubit.tu-berlin.de/rbo-lab/rbo-apc-object-segmentation [12]

Accompanying paper: Probabilistic Multi-Class Segmentation for the Amazon Picking Challenge [13]

Original author: Rico Jonschkowski

Contact person: Vincent Wall [14]

Project: Amazon Picking Challenge 2015 [15]

concarne: Learning with Side Information

concarne is a lightweight python framework for learning with side information (aka privileged information). concarne implements a variety of different patterns that enable to apply side information. As it builds upon Theano [16] and lasagne [17], you can use neural network structures that you have developed yourself and easily combine them with the side information learning task.

Repository: https://github.com/tu-rbo/concarne [18]

Accompanies paper: Patterns for Learning with Side Information [19]

Maintainers: Sebastian Höfer, Rico Jonschkowski

Research project: Representation Learning [20]

 

 

Webservices

RBO Aleph: Ein Webservice für Proteinstrukturvorhersage

RBO Aleph ist eine Pipeline für die Strukturvorhersage von Proteinen. Der Fokus liegt auf ab initio Strukturvorhersage.

RBO Aleph war unter den besten ab initio Strukturvorhersagemethoden in CASP11.

Zugriff gibt es über das Webinterface:

http://compbio.robotics.tu-berlin.de/rbo_aleph/ [21]

Contact: Mahmoud Mabrouk [22]

 

 

EPC-map: Kombination von evolutionären und physicochemischen Informationen in der Kontaktvorhersage

EPC-map (using Evolutionary and Physicochemical information to predict Contact maps) vereint decoy-basierte Kontaktvorhersage mit maschinellem Lernen und evolutionären Informationen. 

In CASP11 hat EPC-map den 2. Platz belegt in der Kategorie long+medium range Kontakte und den 5. Platz bei long-range Kontakten.

Zugriff gibt es über das Webinterface:

http://compbio.robotics.tu-berlin.de/epc-map/ [23]

Contact: Kolja Stahl [24]

EPSILON-CP v2: Vorhersage von Residue-Residue Kontakten durch die Kombination von Informationen aus Sequenz und Physikalischer Chemie

EPSILON-CP v2 benutzt deep learning um verschiedene Informationsquellen (Sequenzbasiert, physicochemisch, evolutionaer) effektiv zu kombinieren. Im Gegensatz zu v1 ist das Netzwerk fully convolutional. EPSILON-CP hat den 5. Platz im finalen Ranking von CASP12 in der Kategorie Contact Prediction belegt (Gruppenname RBO-EPSILON).

http://compbio.robotics.tu-berlin.de/epsilon [25]


EPSILON-CP v2 ist Teil der Structure Prediction Pipeline RBO Aleph die stetig in CAMEO evaluiert wird.

Contact: Kolja Stahl [26]

Tutorials

Tutorial on building PneuFlex actuators

PneuFlex actuators are a version of so called soft pneumatic continuum actuators. Their production process is relatively simple, and can accomodate complex actuator shapes. This tutorial [27] explains how to build them.

 

Contact: Raphael Deimel, Vincent Wall

 

 

Pneumaticbox Control System

The Pneumaticbox is a specific hardware combined with an embedded software system to provide a complete control solution for PneuFlex actuators:

  • The Hardware [28] consists of valves, pressure sensors, common I/O interfaces (USB, SPI, I2C), and an embedded computer.
  • The pneumaticbox-airserver [29] runs on the Pneumaticbox to abstract hardware I/O and runs real-time controllers on behalf of networked clients. It includes controllers for straight forward control of valve opening times [30], but also more advanced controllers for air mass and air mass flow [31] control. Additionally, safety measures in the form of a pressure limiter [32] and a pressure watchdog (emergency stop) [33] are implemented.
  • The python-pneumaticbox [34] is the Python client library. It can be used for easy integration into existing systems or direct scripting of experiments.

 

Related publications:

  • Raphael Deimel and Oliver Brock. Soft Hands for Reliable Grasping Strategies [35]. Soft Robotics, Soft Hands, Grasping. Springer-Verlag, chap. Soft Hands for Reliable Grasping Strategies, 211–221, 2015.
  • Raphael Deimel and Oliver Brock. A Novel Type of Compliant and Underactuated Robotic Hand for Dexterous Grasping [36]. The International Journal of Robotics Research 35(1-3):161-185, 2016.
  • Raphael Deimel and Marcel Radke and Oliver Brock. Mass Control of Pneumatic Soft Continuum Actuators with Commodity Components [37]. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 774–779, 2016.


Contact: Raphael Deimel, Vincent Wall [38]

------ Links: ------

Zusatzinformationen / Extras

Direktzugang:

Schnellnavigation zur Seite über Nummerneingabe

Copyright TU Berlin 2008