We build autonomous robots for real-world physical work
One robot platform — autonomy, safety and fleet software — carried into the places where the work is heavy, repetitive and outdoors
- #Autonomous-robots
- #Field-robotics
- #Safety-first
- #Fleet-software
- #Prototype-to-production
Software complete and validated in simulation · first prototype in build
Dover, Delaware · Working worldwide
M200 Construction Mule
Autonomous material transport for construction sites: ~200 kg payload, waypoint navigation with follow-me mode in the next phase, all-weather duty, and a safety system designed for working near people. The only program with an engineering pack behind it.
First product · software complete, prototype build starting
See the robots→ACR-1 Quadruped
A four-legged research platform for the terrain wheels cannot handle — stairs, debris, unfinished floors. Simulation-first by design: no physical build until measured milestones are passed.
Research · simulation only
See the robots→Exploration · concept stage — not products, not an announced roadmap
Material-placing manipulator — construction
A future arm-equipped machine that places material rather than only transporting it — blocks onto a course, panels onto a frame. Dexterous manipulation on an unstructured site is an open research problem industry-wide, which is why this is a long-horizon exploration and not a roadmap item.
Autonomous water taxi — marine
A small solar-assisted electric craft running on-demand routes across bays and along coastal cities, sized in concept for a handful of passengers — four at most. Water replaces the road; the autonomy, safety and fleet layers are the same ones under the M200.
Field platform — agriculture
An autonomous platform working along crop rows: transport, irrigation and spraying first. Selective harvesting — judging what is ripe and picking it without damage — is the harder question and would come later as research, never as a promised feature.
Search and assist — disaster response
Locating people and animals in rubble, floods, avalanche and maritime incidents, then carrying first-aid payloads and breathable air to them. Life-safety work demands a higher reliability and certification bar than material transport — which is exactly why this follows rather than leads.
Construction first, because it is the hardest useful place to start
Sites are outdoors, unstructured and full of people. A platform that earns its evidence there has earned the right to be carried elsewhere — which is why the M200 Construction Mule is the first product and every other sector is labelled exploration.
The machine
A wheeled material carrier that moves blocks, mortar and tools between site zones so skilled people do not push wheelbarrows.
The status
Software complete and validated in simulation. First prototype in build. No physical robot exists yet, and every figure we publish says which it is.
The model
Robots rented as outcomes: a site pays for material moved and reported, not for a machine to own and maintain.
One stack under every machine we build
A construction carrier, a legged platform, a water taxi, a field platform, a search-and-assist machine — different bodies over the same foundations. Building the autonomy, safety and fleet layers once, and proving them on the hardest site first, is what makes every further domain an engineering question rather than a rebuild — and what keeps each one labelled exploration until it has earned more.
Autonomy & perception
Mapping, localisation, obstacle handling and task planning that run on the machine itself — evaluated in simulation against thousands of runs before any hardware moves.
Safety architecture
An independent, hardware-level stop chain that never depends on AI. People always have right of way; the certification path is designed in from day one, not bolted on.
Fleet & operations software
Task allocation across many robots, live monitoring, offline-tolerant field apps and automatic daily reports — the layer that makes a robot useful to a business.
Mechanics & embedded systems
Parametric mechanical design, drivetrain and power architecture, and certifiable embedded control — the physical half, engineered by the same team as the software.
From simulation to a supervised site
Each stage ends with something that can be inspected — a result archive, a machine, a signed scoreboard — and nothing is claimed before its stage has produced it.
- [01]
Simulation
Mission logic, fleet coordination and tasking validated across thousands of simulation runs — complete, and labelled simulation until the yard says otherwise.
- [02]
Prototype build
The first physical M200: chassis, drivetrain, sensors and the safety chain. Procurement and detailed design under way; no hardware built yet.
- [03]
Yard validation
A fenced test program before any robot works near people. The safety chain is proven before the motors get power.
- [04]
Supervised pilots
Partner sites, supervised and measured against targets signed before the trial starts. Robots rented as outcomes, never sold as machines.
Straight answers
Still unclear on something? Ask directly — [email protected].
What does AtaForge do?
AtaForge is a robotics company. We build autonomous robots for real-world physical work — the mechanics, the autonomy software, and the fleet systems that operate them. Our first application is material transport on construction sites.
Do you have a robot working today?
Not yet, and we say so plainly. The software platform is complete and validated across thousands of simulation runs; the first physical prototype is being built now. Every figure we publish is labelled simulation until the yard says otherwise.
Why can one company address several robotics domains?
Because the hard part is shared. Autonomy, safety architecture and fleet software are the same underneath a construction carrier, a water taxi, a field platform or a search-and-assist robot. Building that platform once is what lets us name several sectors — and every sector beyond construction stays labelled exploration until it has earned its own evidence.
What is the M200 Construction Mule?
Our first product: an autonomous material-transport robot for construction sites, ~200 kg payload, tasked in plain language and coordinated as a fleet. It is in the prototype build phase, on a CE certification path.
Where is AtaForge based?
AtaForge Inc. is registered in Dover, Delaware, United States, with AtaForge Europe Ltd. in Malta — one engineering organisation, working worldwide.
How does a conversation start?
With a technical call. Tell us the site, the material flow or the sector you have in mind; we will say honestly whether it fits the M200 program, a pilot, or a domain we are still exploring.
Do you sell robots?
Not as machines. The M200 is planned as robots-as-a-service: a site rents the outcome — material moved, reported every evening — rather than buying and maintaining hardware.