Overnight floor scrubbing wasn't automatable until a machine could map the building and stay out of the way

October 1, 2026 · 3 min read

Ride-on scrubbers have existed for decades. What did not exist was a scrubber that could learn a building, run the same route every night without a driver, stop for a pallet left in the aisle, and report what it cleaned. Autonomous navigation did that, and it changed who does the floor at two in the morning.

The short answer

The hard part of cleaning a large floor was never the scrubbing. It was that somebody had to drive the machine for hours, usually at night, usually at a rate that made turnover a permanent problem. Autonomous scrubbers map a building once, run assigned routes on their own, detect people, carts and spills and route around or stop, and report coverage to a dashboard. That turned a driving job into a supervising job, and it is why airports, warehouses and supermarkets now run them nightly.

Why this job resisted automation for so long

Early automated floor machines followed buried wires or painted lines and stopped dead when anything was in the way. They could not be redeployed to a new area without infrastructure, and they could not share a floor with people. The sensing and computing needed to localize a machine in a changing building and plan around obstacles was expensive and fragile. Cleaning contractors also had little incentive to invest, because the labor was billed by the hour.

What changed

Lidar and cameras became cheap and reliable enough to put on a floor machine. Simultaneous localization and mapping software matured, so a machine can build a map of a building during a teach run and then find itself on that map every night. Obstacle detection got good enough to see low objects and moving people. And makers wrapped the whole thing in a software plan with a dashboard, so a facilities manager can see which areas were cleaned, when, and what the machine skipped. Disclosure: Steel Collar is an independent dealer for Avidbots, one of the makers in this category, and the matrix ranks every scrubber in it on fit.

The casebook on this site includes two published accounts as the vendor published them: a Cincinnati airport that put one machine on about 200,000 square feet a week and more than 5 million square feet in its first year, and a Las Vegas lighting warehouse that doubled its floor area and reports its cleaning team's productivity doubled with one robot running within a week of delivery. The airport's own payback estimate assumed eight hours of running a day, which is a fleet-style duty cycle. A robot that runs two hours a night pays back far more slowly.

What it looks like now

A robot scrubber sized to the building, mapped by the vendor or a trained operator, running routes on a schedule or on demand. A person still handles edges, spills during the day, restrooms and anything the robot flags. Every machine in this category runs on a software plan, so the honest total cost is the machine plus the annual plan plus service, and the concierge says that plainly. Deployment time runs from days to a few weeks depending on how complicated the building is.

What still disqualifies a building

Floors that are mostly narrow aisles, tight back rooms and thresholds, where a robot spends its time turning around. Heavy daytime traffic with no overnight window and no appetite for a robot moving through customers. Fewer square feet than it takes to use up a machine's nightly runtime, which is the arithmetic question the concierge asks first. And a cleaning contract where the contractor, not you, decides what equipment runs on your floor.

How to find out if yours is automatable

Measure the open hard floor you clean, how often you clean it, and how many driver hours that takes per week. Those three numbers do most of the work. Describe the building to Joe, the concierge on this site. It will size the machine class against your square footage, tell you which models fit open floors and which fit tighter ones, and say plainly where a walk-behind still wins.

Related on this site

The robot replaces the driving, not the cleaning team. Count the hours your people spend steering and the square feet you clean less often than you should.

Put it to the test on your own job

Tell Joe what you are running and he will apply this to your numbers, including telling you when a robot is the wrong answer.

More notes