Embedded Files
Abstract
This paper introduces H-MaP, a hybrid sequential manipulation planner that addresses complex tasks requiring both sequential actions and dynamic contact mode switches. Our approach reduces configuration space dimensionality by decoupling object trajectory planning from manipulation planning through object-based waypoint generation, informed contact sampling, and optimization-based motion planning. This architecture enables handling of challenging scenarios involving tool use, auxiliary object manipulation, and bimanual coordination. Experimental results across seven diverse tasks demonstrate H-MaP's superior performance compared to existing methods, particularly in highly constrained environments where traditional approaches fail due to local minima or scalability issues. The planner's effectiveness is validated through both simulation and real-robot experiments.
System flowchart of H-MaP's three-phase architecture:
(I) Bi-RRT-based waypoint generation with obstacle handling, (II) hybrid contact point determination through learning and sampling, and (III) optimization-based motion planning. The system supports both standard end-effector (ID_{manip_0}) and tool-based manipulations (ID_{manip_n}), enabling dynamic replanning and recursive obstacle handling for single and multi-tool scenarios.
Comparison of success rates (out of 10 random trials) and planning times (seconds) for successful trials. (-) indicates complete failure across all trials. (*) LGP times reflect motion planning with provided task plan skeletons, excluding task planning time, thus representing a lower bound for total LGP planning time.
Comparison of total object trajectory lengths (in meters).
Bi-RRT, CMGMP, and our method often produce shorter or comparable paths due to their sampling-based generation, while LGP and KOMO yield longer yet smoother trajectories by fully optimizing robot motions. Our approach first performs an object-centric search and then couples the object and robot configurations in a single optimization phase, resulting in short but reasonably smooth paths.
Page updated
Google Sites
Report abuse