Manipulation of Deformable Linear Objects Using Knot Invariants to Classify the Object Condition Based on Image Sensor Information

摘要

Using a topological model and knot theory, we propose a method for describing the condition of a rope. We also propose a recognition method based on the image information obtained from the charge-coupled device cameras to obtain the structure of the rope when manipulated by a robot. This method will help solve the difficulties of robots manipulating deformable objects by providing a theoretical framework of error recovery for deformable object manipulation. We confirm the effectiveness of the methods through experiments.