F-ARM: Fault Diagnosis for Robotic Arm

Research

About The Project

Deep learning-based fault diagnosis system for robotic arm anomaly detection using sensor data analysis

Key Achievements

—Developed deep learning-based fault diagnosis system for robotic arm anomaly detection using multi-sensor data

—Implemented CNN and LSTM models for real-time fault classification achieving high accuracy in bearing defect detection

—Collected and preprocessed vibration sensor data from robotic arm joints for training machine learning models

—Won Grand Prize in Mechanical Engineering Capstone Design Competition (2021)

—Published graduation thesis applying deep learning techniques to mechanical engineering fault diagnosis

—Created automated data collection pipeline and feature extraction system for continuous monitoring

Project Files

More Projects