A Vector Construction Method Based on Binocular Vision for Configuration Detection of Concentric Cable-Driven Manipulators

The concentric cable-driven manipulators have attracted considerable attention for its applications in confined spaces. However, a major challenge lies in accurately determining its configuration owing to the absence of robust visual information and configuration detection methods. To address this limitation, this paper proposes a vector construction method based on binocular vision for the configuration detection of concentric cable-driven manipulators. Firstly, the markers that are easily identifiable and do not interfere with the manipulator’s functionality are attached to the concentric cable-driven manipulator. Subsequently, a binocular vision system is designed to obtain real-time images of the concentric cable-driven manipulator. Finally, based on the binocular vision system and markers, the vector construction method is proposed to detect configurations of the concentric cable-driven manipulator. Extensive experiments show that the configuration detection accuracy of the proposed method for the concentric cable-driven manipulator is more than 94%. These results not only confirm the effectiveness of the proposed vector construction method but also highlight its potential for enhancing the reliability of concentric cable-driven manipulators in various applications.