CRAM describes the way in which the generative model and generalized actionplan framework are realized in EASE. Robot agents need to be equipped with software that enables them to perceive their environments and produce competent actions in order to physically perform manipulation activities. CRAM integrates perception, motion control, plan-based control, and other cognitive capabilities in a coherent software framework that facilitates its embodiment in robot agents.
GISKARD is a software framework for constraint- and optimization-based cognitive motion control in robot agents.
KnowRob is a knowledge processing system that combines knowledge representation and reasoning methods with techniques for knowledge acquisition and for grounding the knowledge in a physical system. It serves as a common semantic framework for integrating information from different sources.
The NEEM-HUB acts as a central data storage and management system for the CRC EASE. Researchers and cooperating researchers can upload their NEEMs, share them with the community, and work with the data on OPENEASE. The NEEM handbook describes how to use the NEEM-HUB.
openCollBench is an open benchmark that enables a fair analysis of different collision detection & proximity query algorithms. A simple yet interactive web interface allows both expert and non-expert users to easily evaluate different collision detection algorithms’ performance in standardised or optionally user-definable scenarios and identify possible bottlenecks. In contrast to typically used simple charts or histograms to show the results, openCollBench proposes a heatmap visualisation directly on the benchmarked objects that allows the identification of critical regions on a sub-object level. An anonymous login system, in combination with a server-side scheduling algorithm, guarantees security as well as the reproducibility and comparability of the results. This makes openCollBench useful for end-users who want to choose the optimal collision detection method or optimise their objects with respect to collision detection but also for researchers who want to compare their new algorithms with existing solutions
PRAC is an interpreter for natural-language instructions for robot applications. The PRAC system makes knowledge about everyday activities available for service robots so they can autonomously acquire new high-level skills by browsing the internet (like wikiHow). PRAC addresses the problem of natural language being inherently vague and unspecific. PRAC maintains probabilistic first-order knowledge bases over semantic networks represented in Markov logic networks.
RobCoG acquires commonsense and intuitive physics knowledge for robot agents using games with a purpose. In the games users are asked to execute various tasks in different scenarios. The games are equipped with a semantic logging system which captures and stores symbolic and subsymbolic data during the gameplay.
RoboSherlock is a common framework for cognitive perception, based on the principle of unstructured information management (UIM). UIM has proven itself to be a powerful paradigm for scaling intelligent information and question answering systems towards real-world complexity