.png){kind=icon}
.png){kind=icon}
Return
Generalized machine learning model for deformation prediction and compensation in robotic machining
.png){kind=icon}
OA
AI
T
G
S
B
DOI：10.1016/j.rcim.2026.103314.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A generalized ML model predicts robot deformation without retraining under varying workspace and process conditions. • Latin hypercube sampling and a variable load system enable representative data acquisition across joints and wrenches. • The proposed model achieves prediction errors below 10% (MAPE) on all axes without retraining or additional data. • Robotic drilling experiments demonstrate up to 80% reduction in hole-position error using ML-based deformation compensation.
Keywords:
Covariate shift
Supervised learning
Data construction
Robot stiffness
Compliance error
Robotic machining
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
R
IF:
11.4
Papers:
3.3K
Citations:
1.3W
Organization
Cited Papers