The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec)
Online ISSN : 2424-3124
2024
Session ID : 2A1-K01
Conference information

Anomaly detection of robot arm using autoencoder
*Noriyuki KAWARAZAKIYuma ISHIDATadashi YOSHIDOME
Author information
CONFERENCE PROCEEDINGS RESTRICTED ACCESS

Details
Abstract

This paper discussed the anomaly detection of robot arm using autoencoder. In our system, the current of joint motor are measured by the current sensor. We adapted the autoencoder to detect the anomalous value in the motor’s current. The autoencorder is the one method of the unsupervised machine learnings. The effectiveness of our system is demonstrated by several experiments.

Content from these authors
© 2024 The Japan Society of Mechanical Engineers
Previous article Next article
feedback
Top