On Cognitive Reasoning for Compliant Manipulation Tasks in Smart Production Environments

Abstract

Highly automated smart production environments require robots with autonomous planning mechanisms as well as effect-based performance inference methods. This report discusses the possibilities of cognitive reasoning for compliant manipulation tasks to satisfy these demands. The article builds on the representations for compliant wiping actions and their effects which are fundamental to many tasks in industrial manufacturing. It is described how these actions can be planned, executed, and interpreted by means of generic action descriptions and qualitative models.