PDDLStream: Integrating Symbolic Planners and Blackbox Samplers

Many planning applications involve complex relationships defined on high-dimensional, continuous variables. For example, robotic manipulation requires planning with kinematic, collision, visibility, and motion constraints involving robot configurations, object poses, and robot trajectories. These constraints typically require specialized procedures to sample satisfying values. We extend PDDL to support a generic, declarative specification for these procedures that treats their implementation as black boxes. Our framework supports cost-sensitive planning. We provide domain-independent algorithms that reduce PDDLStream problems to a sequence of finite PDDL planning problems. Our best algorithm is able to locally satisfy and optimize candidate plans, resulting in better performance than existing methods. We evaluate our algorithms on three simulated robotic planning domains as well as several real-world robotic tasks.