KnowRob tutorial

Materials and instructions for the Knowrob tutorial will be put here. The following list is still an incomplete collection of items. You will be notified by mail once it is completed.

Prerequisites & software installation

KnowRob knowledge base

Check out the KnowRob repositories into your ROS_PACKAGE_PATH using the following rosinstall file:

rosws merge
rosws update

If you are asked for a password to check out the prac stack, you can use the user 'robohow' and the password 'dff273f8'. Then compile the required packages:

rosmake knowrob
rosmake comp_ehow  # OPTIONAL, will download some hundred MB

Go through the KnowRob basics tutorial to make sure everything has been installed correctly.


On ros fuerte, if you have the following error:

$ rosmake knowrob
cannot find required resource: client_rosjava_jni

Please install

sudo apt-get install ros-fuerte-client-rosjava-jni  ros-fuerte-ias-common 

Other dependencies libjson-java swi-prolog json-glib-1.0 json-glib-dev

Labeling tool for human activity data

Download and extract the data labeling tool and the example data for the tutorial. Try to start the labeling tool and to load the example data into the program. You should see a video image and several markers on the timeline. With PgUp and PgDn you should be able to jump through the marker sequence and see the updated video images for each marker.

Probabilistic Robot Action Cores

Install ProbCog

$ <path-to-probcog> python
$ <path-to-probcog> source

Install Toulbar2 from Make sure it's in your PATH

Install NLTK and jpype

$ sudo apt-get install python-nltk
$ sudo apt-get install python-jpype  

Install PRAC:

$ <path-to-prac> python
$ <path-to-prac> source

Presentation slides

The presentation slides can be downloaded from here.

Further information

The page on KnowRob lists several tutorials and links to further information. The KnowRob wiki is a collection of information on many topics related to the KnowRob knowledge base.