Bimanual Regrasp Planning and Control for Active Reduction of Object Pose Uncertainty
Ryuta Nagahama, Weiwei Wan, Zhengtao Hu, Kensuke Harada
AI summary
Problem
Precisely grasping objects is hindered by pose uncertainties that cause task failures, while existing correction methods rely on cumbersome external cameras or custom fixtures.
Approach
The method sequentially executes three approximately orthogonal grasps using two robotic arms with flat finger pads, combined with admittance control to safely conform to the object's actual pose during handovers.
Key results
- Developed a regrasp planning algorithm that selects approximately orthogonal grasp triplets
- Integrated admittance control to safely adapt grasping poses to uncertain object positions
- Demonstrated high repeatability across varying initial uncertainties in physical experiments
- Achieved remaining pose deviation levels comparable to optical tracking systems
Why it matters
Enables precise, flexible object handling in diverse robotic manipulation tasks without requiring external sensors or environment modifications.
Abstract
Precisely grasping an object is a challenging task due to pose uncertainties. Conventional methods have used cameras and fixtures to reduce object uncertainty. They are effective but require intensive preparation, such as designing jigs based on the object geometry and calibrating cameras with high-precision tools fabricatedusinglasers. Inthisstudy,weproposeamethodtoreduce the uncertainty of the position and orientation of a grasped object without using a fixture or a camera. Our method is based on the concept that the flat finger pads of a parallel gripper can reduce uncertainty along its opening/closing direction through flat surface contact. Three approximately orthogonal grasps by parallel grip- pers with flat finger pads collectively constrain an object’s position and orientation to a unique state. Guided by the concepts, we develop a regrasp planning and admittance control approach that sequentially finds and leverages three approximately orthogonal grasps of two robotic arms to actively reduce uncertainties in the object pose. We evaluated the proposed method on different initial object uncertainties and verified that it had good repeatability. The deviation levels of the experimental trials were on the same order of magnitude as those of an optical tracking system, demonstrating strong relative inference performance.