Explore Labs

PyCRAM Laboratory

Explore Labs

Enter Lab Back to overview

Description

The PyCRAM Laboratory is dedicated to leveraging and demonstrating the capabilities of PyCRAM, the plan execution framework of the CRAM cognitive architecture. PyCRAM is designed to facilitate the execution of high-level actions on robots operating in partially observable and dynamic environments, as well as to support the development of symbolic and comprehensive robot plans. These plans enable AI-driven and cognitively-informed control of autonomous robots.

The laboratory aims to advance the field of robotic autonomy by providing a modern, accessible, and extensible platform that fosters innovation in the design and implementation of intelligent robotic control systems. By offering an integrated development environment for sophisticated robot behaviors, the PyCRAM Laboratory supports the research and deployment of autonomous robotic solutions in real-world scenarios.

PyCRAM is particularly tailored for operation in human household environments, which are inherently dynamic and partially observable. In such contexts, it is essential for a robot to reason about failures and possess the ability to re-plan its actions at any point in time in response to unexpected changes or incomplete information.

To achieve these objectives, PyCRAM provides a suite of tools that assist in the development of robust and flexible robotic behaviors. These include modules for geometric reasoning, coordinate transformations, control flow structuring, action simulation, and more. Importantly, robot plans created using PyCRAM are robot-agnostic, capable of seamless execution both in simulation and on physical robots, and can be exchanged across different robotic platforms with minimal adaptation.

Examples

image alt

Software components

  • PyCRAM: A software framework for designing high-level robot behaviour.

  • GISKARD: A framework for constraint- and optimization-based robot motion and planning control.

Course and Tutorials

ACRAMbly: A project where students assemble a small plan in an industrial setting.
SUTURO: A project where students design their own applications to run on real robots.

Author and Contact Details

Publications

Pekarek-Rosin, Theresa, Hassouna, Vanessa, Sun, Xiaowen, Krohm, Luca, Kordt, Henri-Leon, Beetz, Michael and Wermter, Stefan, “A Framework for Adapting Human-Robot Interaction to Diverse User Groups”, In Social Robotics, vol. , no. , pp. 24–38, 2025. doi

Kümpel, M., Dech, J., Hawkin, A. and Beetz, Michael, “Evaluation of Autonomous Shopping Assistants Using Semantic Digital Twin Stores”, In AIC’23: 9th workshop on Artificial Intelligence and Cognition, 2023.

Live Demo

For a live demo in the browser plase look at the Virtual Research Building demos here (From chapter 2 onwards)

You can find Jupyter Notebooks for the different features of PyCRAM here

Back to top