The challenge is ensuring you have sufficiently frequent passes on hot days, and therefore sufficient fitted trains, right? That paper seems to be notable beca

The challenge is ensuring you have sufficiently frequent passes on hot days, and therefore sufficient fitted trains, right? That paper seems to be notable because it’s doing it from vision alone (and no LIDAR), whereas similar things (including gauging generally) have been used commercially for a while; see eg http://www.rail-vision.co.uk/ http://www.rail-vision.co.uk/