The paper proposes a new education software system for flat finishing skill with an iron file based on classification of personal peculiarities. The software measures learner's flat finishing motion by using a 3D stylus, and displays classified personal peculiarities effectively to correct learner's finishing motions. The system also extracts peculiarity feature pattern based on the trajectory of the iron file. A torus type Self-Organizing Map is used to classify such unknown number of classes of peculiarity patterns. To let the trainers make viewing entire students' peculiarities easily, the resulted feature maps of the torus type SOM are classified into appropriate peculiarity classes based on the cluster map values. Early developing results of the peculiarity computation classification parts with measured data of an expert and sixteen learners show effectiveness of the proposed system.
[1]
Shinpei Matsumoto,et al.
Personal peculiarity classification of flat finishing tool motion by using self-organizing maps for skill training
,
2015,
2015 10th Asian Control Conference (ASCC).
[2]
Shimpei Matsumoto,et al.
A brush coating skill training system for manufacturing education at Japanese elementary and junior high schools
,
2015,
Artificial Life and Robotics.
[3]
Teuvo Kohonen,et al.
Self-Organizing Maps
,
2010
.
[4]
Shinpei Matsumoto,et al.
Personal Peculiarity Classification of Flat Finishing Skill Training by using Torus type Self-Organizing Maps
,
2017,
DCAI.