Scholarlink

Amidst the rapidly advancing landscape of robotics, AI, and IoT, collaborative robots (cobots) play a pivotal role in the transition to Industry 5.0. The precise positioning and performance of cobots are crucial for ensuring product quality and worker safety. This paper examines system and subsystem-level Prognostics and Health Management (PHM) by leveraging Tool Centre Point (TCP) position and orientation data, along with joints’ currents measurements of cobots, to detect anomalies caused by functional deviations. The proposed framework employs multiple Long Short-Term Memory AutoEncoder (LSTM-AE) models, and the results obtained demonstrate that the framework is highly effective in detecting and isolating cobot anomalies, with high potential to be developed into a smart solution for identifying real faults.

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

Multi-Model-Based Condition Monitoring Framework for the Health Assessment of the Tool Center Point and Joints of Industrial Collaborative Robots

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Samuel Ayankoso,
  • Yang Guan,
  • Fengshou Gu,
  • Hassna Louadah,
  • Hamidreza Fahham,
  • Andrew Ball

摘要

Amidst the rapidly advancing landscape of robotics, AI, and IoT, collaborative robots (cobots) play a pivotal role in the transition to Industry 5.0. The precise positioning and performance of cobots are crucial for ensuring product quality and worker safety. This paper examines system and subsystem-level Prognostics and Health Management (PHM) by leveraging Tool Centre Point (TCP) position and orientation data, along with joints’ currents measurements of cobots, to detect anomalies caused by functional deviations. The proposed framework employs multiple Long Short-Term Memory AutoEncoder (LSTM-AE) models, and the results obtained demonstrate that the framework is highly effective in detecting and isolating cobot anomalies, with high potential to be developed into a smart solution for identifying real faults.