ASC vs RTG: telling yard cranes apart in satellite imagery
If you're building out a yard crane count from imagery instead of a terminal visit, the first fork in the road is automation status. An automated stacking crane (ASC) and a rubber-tyred gantry (RTG) can look like the same gray rectangle from altitude until you know what to check. Here's what actually separates them at sub-meter resolution.
Start with the running surface, not the crane
The single most reliable tell has nothing to do with the crane body. It's what it's sitting on.
ASCs run on rails embedded in the block, usually two parallel steel tracks spanning the full length of the stack, sometimes with a visible gap or seam where the rail bed meets the pavement. Because the rails fix the span, every ASC in a block will have near-identical gauge width, and the crane legs sit squarely over the tracks with no lateral play. Zoom into the block ends and you'll often see the rail extend a few meters past the last bay, sometimes with an end stop or buffer.
RTGs, as the name says, ride on rubber tires on open pavement. No rail lines to spot, but look for tire scrub marks on the concrete, faint parallel striping where the wheels have worn a path over repeated passes down the block. RTGs also need enough clearance to steer between blocks, so you'll sometimes catch one mid-transit in open yard space between stacks, something an ASC physically cannot do since it's locked to its rail run.
If the imagery is sharp enough to make out the crane's leg structure, RTGs typically show an A-frame or trapezoidal leg profile, wider at the base to carry the tire bogies. ASC legs read as straighter, boxier verticals sitting flush on the rail. At 0.5 m GSD this shows up more as a shadow shape than a crisp outline, but the shadow is often enough.
Span, orientation, and parking tell you the rest
Automation shows up in behavior as much as in hardware, and behavior leaves a visual footprint.
ASC blocks tend to park in a straight, evenly spaced line, nose to tail, all oriented the same direction, because the automation system doesn't tolerate the kind of casual positioning a human operator would leave a crane in overnight. When you see a row of yard cranes lined up with suspicious precision, that uniformity is itself a clue, not just a style choice.
RTGs are messier. Drivers park wherever the shift ends, sometimes angled slightly off the block centerline, sometimes clustered near the block end closest to the office or fuel point rather than spread evenly. If you're scanning a terminal and one yard looks regimented and another looks lived-in, that difference in parking discipline is often automation status showing through, before you've even confirmed rail versus tire.
One more cue worth checking: many ASC blocks are built with a cantilevered beam on one side, overhanging a truck lane so haulage trucks can pull in beneath the crane without entering the automated zone. That overhang reads as an asymmetric crane profile, wider on one side than the other, which an RTG rarely has since trucks typically enter from the block end under driver line of sight.
Rail-mounted gantries (RMGs) that aren't automated exist too, and they'll throw the rail test off if you stop there. For those, orientation and parking regularity are the tiebreaker since the running surface alone won't separate automated from manually operated rail cranes.
None of this replaces attributing the crane to a manufacturer, which needs boom shape, cab placement, and sometimes a visible OEM marking if the pass angle cooperates. But getting the ASC/RTG call right first is what makes the rest of a terminal's yard inventory usable, since automation retrofit status is one of the first things a competitive OEM read wants out of a crane count. Crane Fleet Mapping builds that terminal-by-terminal inventory once a year from VHR imagery, cranes counted and typed block by block.
If your current count is a spreadsheet of guesses from old terminal brochures, it might be time to replace it with one built from the imagery itself.