Accurate Garment Surface Analysis using an Active Stereo Robot Head with Application to Dual-Arm Flattening

We present a visually guided, dual-arm, industrial robot system that is capable of autonomously flattening gar- ments by means of a novel visual perception pipeline that fully interprets high-quality RGB-D images of the clothing scene based on an active stereo robot head. A segmented clothing range map is B-Spline smoothed prior to being parsed by means of shape and topology into ‘wrinkle’ structures. The wrinkle length, width and height are used to quantify the topology of wrinkles and thereby rank the size of wrinkles such that a greedy algorithm can identify the largest wrinkle present. A flattening plan optimised for this specific wrinkle is formulated based on dual-arm manipulation. Validation of the reported autonomous flattening behaviour has been undertaken and has demonstrated that dual-arm flattening requires significantly fewer manipulation iterations than single-arm flattening. The experimental results also revel that the flattening process is heavily influenced by the quality of the RGB-D sensor, use of a custom off-the-shelf high-resolution stereo-based sensor system outperforming a commercial low-resolution kinect-like camera in terms of required flattening iterations