Improving Unfolding Success Rate Through Cloth Pose Correction

Cloth unfolding is a prerequisite action for most cloth manipulation tasks. Its importance has led to the development of a benchmark and a competition of cloth unfolding through re-grasping in the air, which took place at the 2024 IEEE International Conference on Robotics and Automation (ICRA). While the competition took place on a common dual-arm robot configuration, recreating it with different robot arms to develop and test algorithms has led to hardware-specific issues and solutions. In this paper, we present our implementation of the dual-arm configuration based on the one used in the competition. We also demonstrate how even a relatively rudimentary exploitation of the robot’s capability to move has led to a substantial increase in the number of possible grasps.