Automation of nanoindentation targeting control using machine vision

Area of Science:

• Materials Science
• Mechanical Engineering
• Nanotechnology

Background:

• Nanoindentation is crucial for nanoscale mechanical property evaluation.
• Conventional methods struggle with targeted analysis of microstructural features.
• A need exists for automated, precise nanoindentation workflows.

Purpose of the Study:

• To develop an automated nanoindentation framework supporting machine learning.
• To enhance precision and efficiency in characterizing complex materials.
• To enable feature-based and large-scale automated indentation.

Main Methods:

• Implemented a three-mode automated system: standard, feature-based (image-to-coordinate mapping), and large-scale.
• Utilized direct sample-to-indenter alignment to minimize travel errors.
• Employed pixel-to-micron calibration for accurate navigation and Self-Organizing Feature Maps for guided indentation.

Main Results:

• Achieved precise, automated indentation with reduced user intervention.
• Demonstrated phase-specific and orientation-guided indentation capabilities.
• Mitigated initial travel-distance errors by 2.5-6 μm through direct alignment.

Conclusions:

• The developed framework significantly enhances precision and efficiency in nanoindentation.
• It enables targeted characterization of microstructurally complex materials.
• The adaptable interface supports integration with existing nanoindenter platforms for autonomous testing.