Collision detection of humanoid robot arm under model uncertainties for handling of unknown object

In recent years human-robot collision safety has received considerable attention. Thus, various collision detection algorithms have been proposed to ensure human-robot collision safety, and these algorithms are usually model-based. However, the dynamic model of a robot arm is uncertain or unknown in cases where the arm performs a task with various objects or tools. In this paper, we propose a collision detection method for a robot arm that changes its tools or pick up various objects. For this purpose, a novel collision detection index, which is decoupled from the inevitable external force generated by the object being handled by the robot, is developed. The proposed index is verified through various simulations using a 7-DOF robot arm, and the corresponding results show that regardless of the object that is being handled, it is possible to detect collisions.