Visual shape and position sensing algorithm for a continuum robot

Abstract Continuum robots represent an actively developing and fast-growing technology in robotics. To successfully implement control path planning of continuum it is important to develop accurate three-dimensional shape position sensing algorithm. In this paper, we propose algorithm for the reconstruction robot shape. The performed during several steps. Initially, images from two cameras are processed by applying pre-processing segmentation techniques. Then, gradient descent method applied compare two-dimensional skeleton points both masks. Having compared these points, finds a form. Additionally, proposed able define key using distance base along center line. latter allows controlling interest defined user. As result, developed achieved relatively high level accuracy speed.