Pose interpolation for industrial manipulators under manual guidance

摘要

Human Robot Collaboration (HRC) has shown great promise across a wide range of disciplines. The role of robots often ranges from being an assistant to even a co-worker. This paper discusses a collaborative framework, whereby a robot assists the human in achieving smooth tool pose transitions during the course of a tooling task. The human operator manually teaches the robot a set of orientations at specific key locations on the workpiece, prior to the commencement of the task. Subsequently, a spatial field of virtual frames is generated through Inverse Distance Weighted interpolation, which in turn governs the tool orientation at each instance of robot motion. Afterward, a guided curve tracing task on a three dimensional surface is performed to analyze the effectiveness of the approach.