United States Department of Transportation
ROSA P serves as an archival repository of USDOT-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by USDOT or funded partners.
As a repository, ROSA P retains documents in their original published format to ensure public access to scientific information.
i
Machine Learning Methods for Track Condition Assessment Using Repeated Inspection Data
-
2020-10-01
[PDF-3.91 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report VT-4
-
Corporate Publisher:
-
Abstract:The project was split into two primary tasks. Task 1 concentrates on analyzing existing data from past field testing of a Multifunction Doppler LIDAR system. Virginia Tech has conducted several studies with the prototype Multifunction Doppler LIDAR instrument as far back as 2006. Much of the data analyzed in this program comes from a past study using Doppler LIDAR velocimetry to evaluate down track distance and track geometry. Another more recent set of data was collected during a limited-scope track stability exploratory study with grant funding in 2019. The progress to-date has included a significant amount of software development in creating semi-automated Graphical User Interface (GUI) tools to assist in and accelerate the process for filtering and analyzing raw LIDAR data. The resulting suite of tools is intended to greatly reduce the amount of time needed between data collection and the analysis required to identify points of interest on the track. Rapid feedback on test data is essential for being able to return to specific locations on track and investigate the cause of rail motion.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
