The InCoRAP project investigates human-robot interaction in factory environments, taking into account human intention. This allows for a smooth interaction between humans and robots, since the robot can already integrate recognized human intentions into its actions. Existing approaches to worker assistance leverage comparatively coarse-grained information such as the worker’s trajectory in the environment. This often leads to interruptions in the interaction, for example if the robot still performs actions that would no longer be necessary in the current context. In contrast, we base any assistive functionality (including the actions of the robot) on high-level intentions. To infer such high-level intentions, we regard sensor data in relation to the worker’s current task. This requires a comprehensive, detailed model of the factory environment, integrating information from a rich variety of sensors as well as sophisticated process information stemming, e.g., from an ERP system. The environment model must be accessible for various purposes and on different levels of interpretation and detail. The detection of human intention from available sensor data, the generation and maintenance of such a multi-sensor-based hierarchical semantic model of the current environment, and their usage for action planning and situation-adequate assistance comprise the R&D focus of the project.