Admissible Abstractions for Near-optimal Task and Motion Planning

Admissible Abstractions for Near-optimal Task and Motion Planning

William Vega-Brown, Nicholas Roy

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence
Main track. Pages 4852-4859. https://doi.org/10.24963/ijcai.2018/674

We define an admissibility condition for abstractions expressed using angelic semantics and show that these conditions allow us to accelerate planning while preserving the ability to find the optimal motion plan.  We then derive admissible abstractions for two motion planning domains with continuous state.  We extract upper and lower bounds on the cost of concrete motion plans using local metric and topological properties of the problem domain.  These bounds guide the search for a plan while maintaining performance guarantees.  We show that abstraction can dramatically reduce the complexity of search relative to a direct motion planner.  Using our abstractions, we find near-optimal motion plans in planning problems involving 10^13 states without using a separate task planner.
Keywords:
Heuristic Search and Game Playing: Combinatorial Search and Optimisation
Planning and Scheduling: Robot Planning
Robotics: Motion and Path Planning