Robotic Automation in Apparel Manufacturing: A Novel Approach to Fabric Handling and Sewing

TL;DR

This paper tackles the challenge of flexible robotic sewing in apparel by using a water-soluble posing agent to temporarily stiffen fabrics, thereby removing the need for explicit fabric-property modeling. It introduces a three-stage pipeline—pose estimation, temporary joining via ultrasonic welding, and closed-loop visual servoing for sewing—and shows how to integrate this workflow with conventional sewing machines through a Digital Thread that maps DXF seam descriptions to robot trajectories. A key contribution is the closed-loop sewing control enabled by real-time vision feedback and MoveIt! Servo, achieving improved seam accuracy under disturbances in denim and cotton on real factory-like setups. The work demonstrates practical readiness by validating on production floors with industrial partners, offering a scalable path toward flexible, robot-assisted apparel manufacturing.

Abstract

Sewing garments using robots has consistently posed a research challenge due to the inherent complexities in fabric manipulation. In this paper, we introduce an intelligent robotic automation system designed to address this issue. By employing a patented technique that temporarily stiffens garments, we eliminate the traditional necessity for fabric modeling. Our methodological approach is rooted in a meticulously designed three-stage pipeline: first, an accurate pose estimation of the cut fabric pieces; second, a procedure to temporarily join fabric pieces; and third, a closed-loop visual servoing technique for the sewing process. Demonstrating versatility across various fabric types, our approach has been successfully validated in practical settings, notably with cotton material at the Bluewater Defense production line and denim material at Levi's research facility. The techniques described in this paper integrate robotic mechanisms with traditional sewing machines, devising a real-time sewing algorithm, and providing hands-on validation through a collaborative robot setup.

Figures (5)

• Figure 1: Sewing with Robots: This figure provides a graphical representation of the robotic sewing system. The input for this system is a Drawing Interchange Format (DXF) supplied by the operator. The system's orchestration is managed through the Robot Operating System (ROS). Outputs of the system include various fabrics, including a cotton front panel, a polyurethane back panel, and the final product of finished denim shorts.
• Figure 2: Digital Thread: Digital drawing of back panel of denim shorts that includes a seam designated in red by the designer. The figure also illustrates the extraction of the trajectory from this digital drawing.
• Figure 3: Overview of System Workcell: (a) The first column provides a comprehensive view of the workcell, featuring the UR5 robot, welding table, sewing machine, and the overhead-mounted Realsense D435 camera for pose estimation. (b) The second column presents two images: the upper image showcases the ultrasonic welder mounted on the servo motors, whereas the lower image features the sewing machine utilized in our operation.
• Figure 4: Block Diagram of Closed-Loop Control: The figure presents the schematic of the Integrated Closed-Loop Control framework.
• Figure 5: Closed-loop Results: The top image serves as a reference illustrating the methodology for calculating seam error relative to seam distance and actual trajectory, with the Camera's Field of View (FOV) and the Sewing Needle remaining fixed. The bottom left image presents the outcomes of the open-loop system when subjected to disturbances. The bottom right image depicts the performance of the closed-loop system under identical conditions. (Fixed Seam Distance: 20mm)