The project will enable Network Rail to automatically identify and measure railway structures from LiDAR data, saving valuable time and resources, while also improving planning and operations across the rail network. The current, manual process takes analysts months or even years due to the size of the data and the labor-intensive tasks involved.
“The combination of cross-sectional area, shape, length and speed all place a space requirement on today’s railway,” said James Sweeney, Senior Engineer at Network Rail. “We anticipate this project will offer us a more efficient way to capture, analyse and measure railway features along 20,000 miles of track, which is important to railway safety and the growth and capacity of our network.”
Network Rail collects detailed information about its track and the surrounding features, such as bridges and tunnels. The data is then analyzed to assess clearances between trains and the infrastructure around them, which is key to safety. The new project aims to automate the extraction and calculation of railway features from sensor data, leveraging AI to automatically analyse point cloud data, identify different structure types and perform measurements on the structures. The data will be collected from reality capture solutions from Hexagon’s Geosystems division.
“Network Rail, supported by Innovate UK, is leading the way in the use of AI to automate rail structure identification and measurement,” said Mladen Stojic, president of Hexagon’s Geospatial division. “We are excited to be part of a project that can help transform the gauging process for UK railways.”
