Every construction robotics company worth knowing.
Construction robotics sorts by trade, not by company. Each serious vendor has picked one painful, repetitive job and built a machine for exactly that, which means the useful question is never who makes the best robot. It is who owns the trade where your schedule is bleeding.
The map changed in 2026. Two trades that had a single credible vendor now have two, the equipment and technology giants have started buying their way in rather than watching, and the most interesting money has moved from the machines to the software that drives them.
Jaibot
Hilti - overhead MEP drilling
The tool giant's robotics arm, and the most field-proven robot in construction by operating history. In use since 2020 with contractors including Skanska and Bouygues, distributed through Hilti's existing fleet management channel rather than sold outright.
FieldPrinter
Dusty Robotics - layout
The US leader in BIM layout printing and standard on many large general-contractor interiors. Priced on daily utilisation rather than a list price, which means you pay for the days you actually print.
SitePrint
HP - layout, the second option
A printer company doing what a printer company would do. Smaller and more portable than a gantry-scale unit, prints text and arcs as well as wall lines, works on concrete, plywood, vinyl and tarmac. Sold through construction technology dealers as a purchase plus usage.
TyBOT 3.0
Advanced Construction Robotics - rebar
TyBOT has been on jobsites since 2017, longer than almost anything else in the category, and since September 2024 it can simply be bought: $425,000 standard, $455,000 fully configured. One of very few construction robots with a published purchase price.
Hadrian X
FBR - robotic masonry
The Australian pioneer of truck-mounted bricklaying, now commercially available in the US as walls-as-a-service. You are buying laid walls at a price per area, not a machine, which removes the adoption problem entirely for a developer.
Spot
Boston Dynamics - inspection and capture
The de facto site-scanning platform, with an ecosystem of scanner and camera payload partners around it. From $74,500 before payloads, and the only machine on this list a contractor can straightforwardly buy and own.
Ozmo
Skyline Robotics - facade work
High-rise window cleaning sold as a service, deployed on Manhattan towers and expanding to London with partners including Principle Cleaning Services. The clearest case of a robot taking a job people genuinely should not be doing by hand.
Bravo
PaintJet - coatings
Industrial-scale exterior painting worked from standard lifts, sold as a service for warehouses and large commercial facades. A good example of a vendor who kept the machine and sold the result.
The trade map matters more than the company
Nobody in construction robotics is trying to build a general-purpose machine, and the ones who tried are mostly gone. What survives is narrow: a robot that ties rebar, a robot that prints layout, a robot that drills ceilings. That narrowness is the business model, not a limitation of it.
The practical consequence is that choosing a vendor is really choosing a trade. If your schedule keeps slipping on overhead MEP, there is one serious answer and it is Hilti. If it is layout rework, there are now two. If it is rebar on flat decks, there is one. Working backwards from the machine to the problem almost never produces a deployment that lasts.
It also means the category has no leader in any meaningful sense. A ranking of construction robotics companies is a ranking of unrelated products, which is why this page is organised by what each one does rather than by who is biggest.
Who has shipped, and who has only announced
The division that matters most is not technical. It is whether a company has put machines on paying jobsites and kept them there. Hilti has done it since 2020. Advanced Construction Robotics has done it since 2017. Dusty is standard equipment on large commercial interiors. Boston Dynamics sells a machine you can buy today and has done for years.
Against that, a good deal of what gets announced in this category never reaches a jobsite, and the gap between a convincing demonstration and a machine that survives a winter on a live project is where most construction robotics companies have ended.
This is the single question worth asking any vendor, and it is answerable: how many projects, over how many years, and can you name a contractor who would take the call. A company that can answer it crisply is in a different category from one that answers with a video.
The giants have arrived, and that changes procurement
The most consequential development in 2026 is not a new robot. It is who now owns them. Hilti is a tool manufacturer with a robotics arm and a fleet channel contractors already buy through. HP is a printer company that correctly identified layout as a printing problem. JLG, an access equipment giant, acquired Canvas and its drywall finishing robot early in 2026.
For a contractor this is straightforwardly good news, and it is not about the technology. It means robots arrive through procurement routes that already exist, on terms a finance department already understands, with a service organisation already in place. The hardest part of adopting a construction robot has never been operating it; it has been getting it onto the books and keeping it running.
It also implies a prediction worth stating plainly: the independent single-product robotics company is likely to become the exception rather than the norm in this category, either acquired or distributing through somebody else's channel. Hilti's decade of investment and the JLG-Canvas deal point the same direction.
Two trades now have two credible vendors each
Layout was Dusty's alone for years. HP SitePrint now competes directly, and the two are genuinely different products rather than the same idea twice. Dusty prices on days used and suits high-volume repetitive floor plates. SitePrint is a purchase plus usage through a dealer, is more portable, prints on four surface types including plywood and tarmac, and handles text and arcs. A contractor with varied smaller jobs and one with a tower of identical floors should not buy the same machine.
Rebar is the other. TyBOT held the trade essentially alone, and the change there is not a competitor but a channel: it moved from service-only to outright purchase in September 2024, at a published price. A trade with a buyable machine behaves differently from one where every conversation starts with a sales process.
Competition is the thing this category has needed most. A trade with one vendor has no price discovery and no pressure to publish anything. Two is not many, but it is the difference between a market and a monopoly.
The software layer is where the money moved
The newer companies worth knowing are not building machines at all. They are building the autonomy that drives machines a contractor already owns. Gravis Robotics ran an autonomous excavator trial at Manchester Airport on that basis. FieldAI works on the perception and autonomy layer rather than the chassis.
This matters commercially because a retrofit is the lowest-risk way into the category that exists. The underlying machine keeps working as a normal machine if the autonomy is switched off, the capital is already spent, and the operator already knows the equipment. Nothing about a bolted-on autonomy kit requires a contractor to bet on a startup surviving.
It also explains why earthmoving has no equivalent of a Jaibot or a TyBOT. The excavators and dozers are already there and already good. What was missing was never the machine.
Who to call first, by the gap you actually have
Overhead MEP drilling eating your schedule and your crews' shoulders: Hilti, through its fleet programme. Layout rework bleeding budget on repetitive commercial interiors: Dusty. The same problem on varied smaller jobs or on plywood and tarmac: HP SitePrint. Rebar tying setting the pace on bridge decks or large flat mats: Advanced Construction Robotics.
Masonry labour you cannot hire: FBR, and you are buying walls rather than a robot. Documentation disputes turning into change orders: Boston Dynamics Spot with a scanning payload. Facade maintenance you would rather not put people on ropes for: Skyline. Large-area industrial coatings: PaintJet. Earthmoving and piling at volume: an autonomy retrofit on the machines you own.
Notice that most of those are services or leases rather than purchases. That is deliberate on the vendors' part and it works in your favour on a first deployment, because the question of whether this works on your sites gets answered before the capital question does.
Still worth tracking, still not in our index
Canvas does drywall finishing and was acquired by JLG in early 2026. Built Robotics sells autonomous retrofits for excavators and piling rigs, in the region of $150,000 to $300,000 per machine. Construction Robotics makes the SAM100 semi-automated mason at around $500,000. Brokk has been doing remote demolition longer than anyone has called it robotics. Monumental is working on bricklaying out of the Netherlands.
Two names that used to sit on this list have come off it. HP SitePrint and TyBOT both now have full records and profile pages here, which is what happens when a machine accumulates enough verifiable deployment and published specification to be worth indexing rather than watching.
We index a machine when there is something checkable to index: a price or a stated pricing model, real deployments, and a specification its maker will publish. Until then it goes on this list, which is a statement about the available evidence rather than about the company.