Scholarlink

Nowadays, Deep Learning applications have spread to various areas of knowledge. However, each application has specific characteristics that prevent them from being modeled with the same neural networks used in other applications. Exploring the different configurations of neural networks allows for a better understanding of their behavior and distinguishing where they can be used to solve a given problem. In this work, two multilayer neural network configurations are explored to predict possible failures in the operation of a teach pendant of a collaborative robot and plan its maintenance or repair. The data are analyzed to determine the type of regression to be used. Among the results, the two neural network configurations offer the same results and are a viable option for predicting teach pendant failures due to the type of data available.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Exploring Multilayer Neural Networks to Predict Possible Failures of a Cobot’s Teach Pendant

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Hector Rafael Morano Okuno,
  • Guillermo Sandoval Benitez

摘要

Nowadays, Deep Learning applications have spread to various areas of knowledge. However, each application has specific characteristics that prevent them from being modeled with the same neural networks used in other applications. Exploring the different configurations of neural networks allows for a better understanding of their behavior and distinguishing where they can be used to solve a given problem. In this work, two multilayer neural network configurations are explored to predict possible failures in the operation of a teach pendant of a collaborative robot and plan its maintenance or repair. The data are analyzed to determine the type of regression to be used. Among the results, the two neural network configurations offer the same results and are a viable option for predicting teach pendant failures due to the type of data available.