A Robot Task Planner that Merges Symbolic and Geometric Reasoning

International audienceWe have developed an original planner, aSyMov, that has beenspecially designed to address intricate robot planning problemswhere geometric constraints cannot be simply “abstracted” in a waythat has no influence on the symbolic plan. This paper presents theingredients that allowed us to establish an effective link between therepresentations used by a symbolic task planner and the represen-tations used by a realistic motion and manipulation planning library.The architecture and the main plan search strategies are presented to-gether with an illustrative example solved by a prototype implemen-tation of aSyMov. At each step of the planning process both sym-bolic and geometric constraints are considered. Besides, the plan-ning process tries to arbitrate between finding a plan with the level ofknowledge it has already acquired, or “investing” more in a deeperknowledge of the topology of the different configuration spaces itmanipulates