Hierarchical Hybrid Planning for Mobile Robots
In KI - Künstliche Intelligenz, German Journal on Artificial Intelligence - Organ des Fachbereiches "Künstliche Intelligenz" der Gesellschaft für Informatik e.V., Springer, volume 31, number 4, pages 373-376, Nov/2017.
To act robustly in complex environments mobile robots should consider various aspects of tasks at once, such as causal preconditions and effects, time, space and resource usage. Hybrid planning is a strategy to treat them jointly. However, this leads to a huge search space for constructing a plan. This article presents a dissertation that discusses how hybrid and hierarchical planning are combined in the new planner CHIMP. The hierarchical planning strategy helps to reduce the search space and makes hybrid planning fast enough to be used on a robot. The thesis further shows how this is useful for plan execution too, and how new tasks can be merged online into a plan that is being executed.
Hybrid planning, Plan-based robot control, HTN planning