MJAR: A novel joint generalization-based diagnosis method for industrial robots with compound faults.
Yiming HeChao ZhaoXing ZhouWeiming ShenPublished in: Robotics Comput. Integr. Manuf. (2024)
Keyphrases
- preprocessing
- experimental evaluation
- fault diagnosis
- significant improvement
- objective function
- high precision
- synthetic data
- computational cost
- detection method
- high accuracy
- theoretical analysis
- mutual information
- industrial robots
- error rate
- computationally efficient
- neural network
- support vector machine
- probabilistic model
- control system
- pairwise
- cost function
- test cases
- classification method
- fault detection
- joint estimation
- feature selection