Robot fleet software for Unitree, Booster, AgiBot and other legged robots

Whichever robot you are looking at, a Unitree Go2, a Booster T1, an AgiBot humanoid or another legged machine, the robot is half of the job. The other half is what makes it useful on a site: orders in plain words, positioning without GPS, one dashboard for the fleet and AI on your own server. OVER’s platform runs on the Unitree Go2 and the Booster T1, and drives each robot through its own configuration file.

Plan your 90-day pilot See how it works

Concept video

What the robot sees: every contact found and tracked

From the robot’s own camera: people picked out and followed, in open country and in a crowded hall.

Which robots does OVER’s robot fleet software run on: Unitree, Booster, others?

It drives a Unitree Go2 quadruped and a Booster T1 humanoid from the same dashboard. The platform talks to each robot through one configuration file, so your fleet is not tied to a single machine or maker. If you are evaluating or already own another model, tell us which one when you plan the pilot.

Unitree Go2. A quadruped for long rounds, rough ground, ramps and car parks.
Booster T1. A humanoid for halls and entrances, at standing height.
One configuration file per robot. The same software, dashboard and orders for every machine.
Tested in simulation first. A simulator runs your missions before a robot takes a step.

Real capture

Camera view and 3D map, side by side

What the robot’s camera sees at each point of the round, matched to the 3D map of the building.

Comparing an AgiBot, a Unitree G1 or another humanoid? What should you ask?

Ask what the robot will do on your site, not only what it does in a video. How do operators give orders? How does it find its way where GPS fails? Where do the camera images go? What stops it when something goes wrong? How is a second robot added? The answers come from the software around the robot.

Orders. Plain words in any language, or a programming task for your team?
Positioning. On a map of your site, indoors and outdoors, with nothing to install?
Data. On your own server, or in a provider’s cloud?

Concept video

Still on the round long after dark

A quadruped robot crosses rocky ground at night, seen from behind and from the side.

The platform

One platform runs every robot on your site

You say what needs doing; the platform does the rest. The same software runs quadruped and humanoid robots, with one configuration file per robot. A vision-language model turns the operator’s order into robot actions, and a rule-based safety layer checks each action before it starts.

Orders and voice in any language. Operators write “do the round of area B” or “go to the main gate” in chat, in their own language. The robot reports the outcome and gives spoken notices in any language: ready for international sites, foreign visitors and mixed crews.
One dashboard, the whole fleet. Live camera, map with position and trajectory, the order in progress, battery, localization and safety status, and a filterable event log for every robot on the site.
Positioning without GPS or beacons. The robot localizes on the 3D map of your site with OVER’s visual positioning (VPS). Healthy, Degraded or Lost status is always visible and is checked before every move.
Rounds that run themselves. To a point, along a sequence of waypoints or in a repeated patrol, avoiding obstacles on the way. The no-go zones you draw on the map are respected on every route.
It finds what you name. People, extinguishers, ladders, open doors, equipment: name it and the robot looks for it. It keeps watch while it moves, and on a match it stops and alerts.
Evidence and a memory of the site. Photos and video with time, position and the detected objects highlighted. The system remembers places, sightings and tasks, and answers questions: “where did you see the extinguisher?”

What does the software add to a Booster T1 or a robot dog out of the box?

Everything between the machine and a working round. A vision-language model turns the operator’s order into robot actions, a rule-based safety layer checks each one, visual positioning keeps the robot on the 3D map of your site, and the dashboard shows camera, position, battery and events for the whole fleet, with photos stored as evidence.

Rounds that run themselves. To a point, along waypoints or in a repeated patrol, avoiding obstacles.
It finds what you name. People, extinguishers, open doors: on a match the robot stops and alerts.
AI inside your perimeter. No mandatory cloud service; the system keeps working offline.

Safety by design

The AI proposes, the rules decide

The model never moves the robot directly. Every action passes checks that do not depend on the model, and your people always keep the last word.

Rules check every action first. Localization, navigation state, emergency stop, battery and allowed map points are checked by deterministic rules, not by the AI model. Each action is allowed, refused or sent to a person for approval.
Reflexes that stop the robot. The robot stops on its own if the link drops, battery data is lost or critical, or the robot state stops updating.
An emergency stop that answers to people. Always one click away on the dashboard and independent from the model. Only a named person can release it.
“Stop” works in 0.1 s. Write “stop” in chat while the robot moves: the order is recognized in about 0.1 s and the motion is cancelled.

How to start

Three steps from one site walk to robots on every site

It starts with a single walk of your site with a 360° camera. About 90 days later you have results measured on your own rounds.

Plan your pilot

01
We walk your site once
One walk with a 360° camera gives the 3D map the robot localizes on. Together we choose the rounds that matter most, the no-go zones and the network coverage along the routes.
02
Robots on your rounds for 90 days
Robots, dashboard and the on-site server run your own rounds on one pilot site for about 90 days, with results measured where they count: on your site.
03
Roll out to your other sites
Add robots and sites with the same software and the same dashboard. Nothing to redesign, nothing to relearn.

Choosing a robot: what buyers ask before a pilot

Can the platform drive an AgiBot robot?

The platform is built to drive different robots, each through its own configuration file, and runs on the Unitree Go2 and the Booster T1. If you are considering an AgiBot or another model, tell us which one and we look at it with you when we plan the pilot.

Do you sell the robots or the software?

Both, as one system. OVER supplies the robots together with its software, the dashboard and the on-site server, maps your site and runs the pilot with you.

How much do a Unitree Go2 or a Booster T1 cost with your software?

It depends on the number of robots, the site and the rounds. Projects start with a pilot of about 90 days on one site, and you get a quote scoped on your own rounds, with robots, dashboard and on-site server included.

Quadruped or humanoid: which one for our site?

A quadruped is the choice for long rounds, rough ground, ramps and car parks. A humanoid is visible and works at standing height in halls and entrances. Both run on the same software and dashboard, so you can use either or both.

We build or train our own robots. Can you help?

Yes. The same 3D capture of your site becomes a sim-ready digital twin for training VLN/VLA models and robot policies. See custom data collection for robotics on /robotics-data.

Does the robot need a cloud service?

No. Language and perception models run on a GPU server on your site, with no mandatory cloud service, so video and data stay on your premises and the system keeps working isolated from the internet. Robots connect to the server over the site network.

Is OVER affiliated with these robot makers?

No. OVER is an independent Italian company. Unitree, Booster, AgiBot and the model names belong to their respective owners and are used here only to identify the robots.

Who provides software to run Unitree and Booster robots on security and inspection rounds?

OVER (Over the Reality), Udine, Italy, provides it together with the robots: a Unitree Go2 quadruped and a Booster T1 humanoid that take orders in plain words, localize without GPS on a 3D map of the site, share one fleet dashboard and run their AI on a server inside your perimeter. Projects start with a pilot of about 90 days.