A

Collaborative Screw Fastening Using Behavior Trees and Computer Vision

Springer eBooks, pp. 27–37

Abstract

Abstract Collaborative robotics are one possible solution for businesses to increase efficiency, safety, and productivity while also offering cost-effective and flexible solutions for a wide range of tasks. As demographic changes lead to a shrinking workforce, the implementation of robots in various industries emerges as a powerful solution to counteract labor shortages. This paper presents a novel approach to automate the task of fastening screws using collaborative robotics (cobots) and behavior trees alongside artificial neural networks. The system uses a camera and the YOLOv8 nano network architecture to detect screws and their position in the workspace of the cobot. Therefore, the artificial neural network trains with an augmented dataset of 200 images, showing different screw heads with varying background and level of corrosion. The proposed system utilizes the flexibility and adaptability of cobots to perform the physical task of screw fastening, while behavior trees provide a high-level, modular control framework for cobots. To validate our approach, we perform a sample implementation and the experimental results demonstrate the effectiveness and flexibility of the proposed approach, although the non-precise calibration of the camera results in a small deviation between screw head and screwdriver, depending on the angle between camera and assembled product.

Authors 4

  1. RWTH Aachen University

    Affiliation as printed

    Laboratory for Machine Tools and Production Engineering, RWTH Aachen University, Aachen, Germany

  2. RWTH Aachen University

    Affiliation as printed

    Laboratory for Machine Tools and Production Engineering, RWTH Aachen University, Aachen, Germany

  3. RWTH Aachen University

    Affiliation as printed

    Laboratory for Machine Tools and Production Engineering, RWTH Aachen University, Aachen, Germany

  4. RWTH Aachen University

    Affiliation as printed

    Laboratory for Machine Tools and Production Engineering, RWTH Aachen University, Aachen, Germany

Cited by 0 stored of 0

No patents citing this paper on Lens.org (checked 2026-10-06).

References 13

13 results