Research Area R: Generative models for mastering everyday activity and their embodiment
An advantage of Computer Science and Artificial Intelligence approaches is the ability to build computational models that test theories and hypotheses as a whole and observe the effects empirically. CS and AI methods are used to design, implement, and apply information processing principles to autonomous control and investigate how changes in the information processing mechanisms affect the capability of mastering everyday activity.
The goal of Area R is to investigate the information processing infrastructure necessary for robotic agents that can master human-scale everyday manipulation activities. This system will enable robotic agents to take vague task descriptions and use information about the task, situational context, and object context to perform the task appropriately. Inference of the appropriate action parametrizations has to be done without delaying plan execution. The information processing infrastructure will improve with experience (in the form of NEEMs) and by exploiting the routine and mundane character of everyday activity by making use of the PEAMs investigated in the Research Areas H and P.