Sensor-based robotics for fully automated fibre-mat stitching

Today, automobile and aeronautics industry uses more and more robust and light-weight parts made of carbon fibres. To achieve high structural properties the carbon fibre mats are stitched in the preforming process using a sewing robot. Current systems rely on manual programming and are neither flexible nor productive. The goal of this work is to develop a robotic system that autonomously follows and stitches the seams. To cope with the black and reflective carbon fibres a laser scan sensor is used, though online processing of the range data for seam-tracking remains a challenging task. The paper presents a fast and robust two-out-of-three voting image processing method for tracking the seam and for controlling the robot motion in real-time. The experimental results demonstrate the feasibility of the fully automated, sensor-guided robotic sewing process.