Fixture-Free Automated Sewing System Using Dual-Arm Manipulator and High-Speed Fabric Edge Detection
KAI TANG, Xuzhao Huang, Akira Seino, Fuyuki Tokuda, Akinari Kobayashi, Norman Tien, Kazuhiro Kosuge
AI summary
Problem
Automating sewing along fabric edges without fixtures remains difficult due to soft material manipulation challenges and the lack of real-time, high-accuracy vision systems capable of keeping pace with industrial sewing speeds.
Approach
The system decomposes sewing into fundamental operations coordinated by a dual-arm Petri net controller, while a new perception module reformulates edge detection as a classification task using distributed anchors modeled by a Gaussian Uniform Mixture Model.
Key results
- Five-layer architecture decomposing sewing into fundamental operations
- Hi-FEDS achieves 120 FPS detection with ~1-pixel error
- Successful high-quality sewing across varied fabric shapes and materials
- Experimental platform validates real-time coordination of dual arms and sewing machine
Why it matters
Provides a scalable, jig-free automation pathway for garment manufacturing, addressing a major bottleneck in textile production costs and flexibility.
Abstract
Inspiredbyhumanworkerswhoperformcomplicated sewing tasks by repeating relatively simple operations, this letter proposes a fixture-free automated sewing system using a dual-arm manipulator and an ordinary sewing machine to sew two aligned fabrics along the edges, a common task in garment production. The proposed sewing system has a five-layer architecture: percep- tion, dual-arm sewing Petri net, fundamental operations, control primitives, and hardware layers. This architecture decomposes various complex sewing tasks into sequences of fundamental op- erations. To meet the real-time requirement of automated sewing, a High-speed Fabric Edge Detection System (Hi-FEDS) is further proposed for the perception layer, which formulates the fabric edge detection problem for sewing as a classification problem of predefined distributed anchors. The anchor distribution is modeled by the Gaussian Uniform Mixture Model (GUMM). This method achieves high-speed fabric edge detection at an average of 120 FPS, with an average error of about one pixel. An experimental robotic sewing platform is developed, and the sewing results show that our system achieves high-quality sewing across fabrics of various shapes and materials.