文摘
Direct teaching of an industrial robot is a novel technique to easily teach manipulators. However, the human-hand generated teaching data can have significant noise errors ranging from low to high frequency. To use the teaching data, a post-processing method to correct the teaching trajectory is required. Currently there is no suitable way to correct the direct teaching trajectory other than conventional simple-line smoothing methods. Here, a novel shape-based trajectory correction method to rebuild the teaching data with the curvature and velocity information is proposed. The proposed method is tested on three real objects using a physical direct-teaching manipulator.