With the increased use of small robots in human households like vacuum robots the next logical step is the introduction of more complex robots which are capable of manipulating objects. Introducing such robots into human household poses a multitude of new challenges, they need to operate in dynamically changing environments while being aware of their surrounding to avoid accidents. Furthermore, they need to be able to reason about the outcome of actions and change their behavior accordingly, to name a few. PyCRAM is a framework to design high-level cognition enabled robot behavior, designed to address the aforementioned challenges of deploying robots in human households. This is done by providing tools to define a robot control plan, which is written as high-level human instructions like "go to the table" or "pick up the cup". PyCRAM is then able to translate these instructions to instructions for a robot while taking the current context the robot is operating in into account. Resulting in robot behavior which is constantly adapting to the current state of the environment and context.

You can find the framework on GitHub: PyCRAM