🤖 AI Summary
This study addresses the challenge of discrete-continuous coupled planning in contact-rich manipulation by proposing a unified framework based on geometric manifolds. The core innovation lies in modeling contact modes as strata within the configuration space, thereby reformulating planning as inter-stratum transitions and intra-stratum geodesic searches without requiring predefined mode sequences. By integrating differential geometry with manifold optimization algorithms, this approach enables seamless hybrid planning. Experimental validation on block pushing and dexterous hand reorientation tasks demonstrates that the method generates high-quality trajectories within seconds. Consequently, this framework significantly enhances both planning efficiency and adaptability for complex contact-rich manipulation tasks, offering a robust solution to the inherent discontinuities in hybrid system planning.
📝 Abstract
Contact-rich manipulation poses a discrete question and a continuous one at once, namely which contacts are active and how to move while they hold. The two are coupled by a change of dimension, since each contact that a robot maintains confines its motion to a lower-dimensional manifold. We make that coupling the explicit object of planning by observing that a contact mode is not merely analogous to a stratum of the configuration space; it is one. A plan is then a walk over strata whose within-stratum segments are geodesics. On two contact-rich manipulation tasks in simulation, pushing a T-shaped block around obstacles and reorienting a cube in a dexterous hand, our planner returns solutions within seconds with no mode, contact sequence, or stratum given in advance.