Artificial Intelligence Planning for Complex Systems
AI Planning is about determining actions before doing them, anticipating the things that will need to be done and preparing for them. Planners use domain-independent heuristics to guide the search in huge state spaces, in order to find a plan that achieves the goal while satisfying numerical and temporal constraints and optimising a given metric. Planning for complex systems (e.g., RAS) requires rich models to capture complex dynamics as well as the uncertain and evolving environment, scalable planning techniques and robust methods of execution. PDDL+ is the formalism used in planning to describe hybrid systems, and allows the modelling of the differential equations governing the continuous behaviour of systems. This talk provides an overview of how PDDL+ can be used to model complex domains; presents a new PDDL+ planner based on SMT and the ROSPlan framework for planning with ROS; highlights some open challenges on the integration between task and motion planning.