Depth image-based deformation estimation of deformable objects for collaborative mobile transportation

<p>Human-Robot collaborative transportation is a promising technology that combines the strength of humans and robots.<br> The most common approaches rely on methodologies that exploit force-sensing. However, the drawbacks are multiple. First, the magnitude of force applied might be limited to avoid damages. Then, force measurements might be unidirectional according to the material properties; e.g., compression forces are not measurable for fabrics.<br> This paper proposes an approach based on estimating the deformation state of the manipulated object from depth images. Specifically, the segmented depth images of the manipulated object are fed to a Convolutional Neural Network (CNN) model to estimate the current deformation status. Compared with the desired deformation, the current deformation status is used to generate the robot&#39;s twist command. The methodology is proved in a mobile robot application, where carbon-fiber fabrics are transported. A comparison with the state-of-the-art is reported proving that the proposed method is more accurate and more repeatable.</p>