Advanced user assistance based on AI planning

Artificial Intelligence technologies enable the implementation of cognitive systems with advanced planning and reasoning capabilities. This article presents an approach to use hybrid planning – a method that combines reasoning about procedural knowledge and causalities – to provide user-centered assistance. Based on a completely declarative description of actions, tasks, and solution methods, hybrid planning allows for the generation of knowledge-rich plans of action. The information those plans comprise includes causal dependencies between actions on both abstract and primitive levels as well as information about their hierarchical and temporal relationships. We present the hybrid planning approach in detail and show its potential by describing the realization of various assistance functionalities based on complex cognitive processes like the generation, repair, and explanation of plans. Advanced user assistance is demonstrated by means of a practical application scenario where an innovative electronic support mechanism helps a user to operate a complex mobile communication device.