Peculiarity Classification of Flat Finishing Skill Training by using Torus Type Self-Organizing Maps with Cluster Maps

The paper proposes an unsupervised classification method for peculiarities of flat finishing motion with an iron file, measured by a 3D stylus. The classified peculiarities are used to correct learner's finishing motions effectively for skill training. In the case of such skill training, the number of classes of peculiarity is unknown. A torus type Self-Organizing Maps(torus SOM) is effectively used to classify such unknown number of classes of peculiarity patterns. An automatic clustering method is applied to the torus SOM results based on cluster map value. Experimental results of the classification with measured data of an expert and sixteen learners show effectiveness of the proposed method. The effectiveness of the cluster map is also evaluated.