듀얼암 매니퓰레이터의 협력 운반을 위한 말단 제약 기반 모션 플래닝 알고리즘

초록

Cooperative transport of large objects using dual manipulators in confined and complex workspaces is a challenging manipulation problem. This paper proposes a motion planning algorithm for two 6-DOF manipulators to cooperatively transport target objects safely while avoiding obstacles. To reduce the risk of deformation of the grasped object during transport, strict end-effector distance and orientation constraints are integrated into the motion planning pipeline. Furthermore, to address the joint space jump phenomenon that inevitably occurs during inverse kinematics computation with multiple solutions, the state validation process of the geometric planner is improved. This approach overcomes the critical workspace reduction problem caused by conventional joint range restriction methods. We evaluated the proposed algorithm in simulation through challenging environments, crossing two constrained environments (a narrow cabinet and a box-to-box fence) with two varying object types (a planar panel and a cube). The results demonstrate that, even under ±10 mm of robot base position randomness, the planner stably generates collision-free trajectories that strictly satisfy the constraints. Finally, physical experiments using both heterogeneous and homogeneous dual-arm setups verify the robustness and practical applicability of the proposed framework in real-world environments.

키워드