Robot-handling fabrics with curved edges based on visual servoing and polygonal approximation

Abstract

The main scope of this work is the development of a robot controller for the manipulation of fabrics lying on a work table towards the sewing process. A fuzzy visual servoing manipulator controller is developed to guide the fabric along the sewing line. The paper focuses on handling fabrics with curved edges by locally approximating the curve section with straight-line segments. An innovative algorithm based on the dominant point detection method and the genetic-based method is incorporated in the robot visual servoing system to achieve sewing of the curved edges. Simulation and experimental tests are conducted to evaluate the performance of the proposed algorithm. The results demonstrate that this approach is efficient and effective for the polygonal approximation, and the proposed robotic system is proved to be robust and efficient, achieving acceptable seam accuracy in the minimum time.