Cloth grasp point detection based on multiple-view geometric cues with application to robotic towel folding

Abstract

Abstract-We present a novel vision-based grasp point detection algorithm that can reliably detect the corners of a piece of cloth, using only geometric cues that are robust to variation in texture. Furthermore, we demonstrate the effectiveness of our algorithm in the context of folding a towel using a generalpurpose two-armed mobile robotic platform without the use of specialized end-effectors or tools. The robot begins by picking up a randomly dropped towel from a table, goes through a sequence of vision-based re-grasps and manipulationspartially in the air, partially on the table-and finally stacks the folded towel in a target location. The reliability and robustness of our algorithm enables for the first time a robot with general purpose manipulators to reliably and fully-autonomously fold previously unseen towels, demonstrating success on all 50 out of 50 single-towel trials as well as on a pile of 5 towels.

A 3D descriptor to detect task-oriented grasping points in clothing

“…In recent years Maitin-Shepard et al [6] and Cusumano-towner et al [7] demonstrated functional end-to-end systems for automatically handling clothes (albeit in very controlled settings that simplify perception). The proposed systems are able to pick up a piece of laundry and manipulate it until a desired configuration is reached.…”

“…This measure, and specially R@1, is of most relevance in flexible object manipulation, where typically we will only have one chance of correctly grasping the object before altering its state, hence requiring a new detection process. The second metric used is mean Average Precision, which is equivalent to the area below the precision-recall curve 6 , and takes into account both precision and recall. As can be seen, the main elbow in performance of the collar detection system is close to 128 visual words for both the shirt and the polo shirt, where around 60% recall is attained.…”

A Pinching Strategy for Fabrics Using Wiping Deformation

“…In general, bending rigidities of fabrics have hysteresis properties against the flexures. With hysteresis, the bending rigidity may become smaller for stable shapes depending on how to inflect the fabric; hence, based on Equation (14), the creation of a pinching area can be changed by the hysteresis effect. That is, this effect of hysteresis may be considered a dexterous manipulation by a robotic hand.…”

“…Control of a few manipulated points was shown to regulate several positioned points to their desired locations [10,11], and a dual-armed robotic system has been proposed for dressing activity [12]. Furthermore, the folding motion of clothes using a robotic laundry system [13] and a visual perception for grasping/folding messy fabrics [14] has been assessed. Recently, a specialized gripper for garments [15] and a tactile sensor for identification of fabrics [16] have been developed.…”

Learning to Control a Low-Cost Manipulator using Data-Efficient Reinforcement Learning

“…While existing techniques have the potential to solve various household manipulation tasks, they typically rely on extremely expensive robot hardware [12]. The long-term goal of our work is to develop affordable, light-weight manipulator systems that can interactively play with children.…”