AIM Rail AI Measurement of Railroad Ballast

Railroad ballast is crushed stone featuring sharp edges that interlock with each other and is packed below the rails, and between and around the ties. Ballast is critical because it bears the load from the railroad ties and holds the track in place as heavy trains pass.

Ballast becomes fouled when the rock begins to wear, and smaller pieces grind and shear off, eventually becoming “fines.” When track become badly fouled, fines clog and reduce drainage capability and eventually lead to track instability and geometry issues.

Assessing ballast condition is a significant challenge for railroad operators and inspectors: they must often rely on visual evaluations to judge fouling, without being able to effectively access ballast below the surface.

AIM Rail is a portable technology which characterizes ballast particle sizes in real-time, providing a quantifiable fouling index on site. Run on a smartphone app and powered by machine vision AI, AIM Rail uses a probe and multi-camera system to capture images of subsurface ballast. The captured images are stitched, processed by AIM Rail’s edge AI software, and transferred to the app. The app enables inspectors to characterize areas of good rock, void spaces, and fine particles, outputting a fouling index estimate while tracking site locations and depths.

The powerful AI image segmentation identifies areas with large, clean particles and contrasts those with fine particles and voids. The key output is the AIM Rail fouling index, which rates ballast from clean to highly fouled. Fouling data is pushed to the cloud and can integrate with data from multiple inspectors and sites. A web interface gives teams access to advanced GIS visualization and full site analysis.

AIM Rail’s data consolidation and media logging platform was built on Oceanit’s MERCI system backbone.