Platform Autonomy

Ground Robotics Autonomy Box

Bolt-On Autonomy for UGVs & Robotic Dogs

A retrofit autonomy computer that upgrades existing ground robotic platforms with navigation, obstacle avoidance and autonomous mission behaviors.

The Ground Robotics Autonomy Box is a retrofit autonomy computer that mounts onto a UGV or robotic dog already in service and drives it through its existing vehicle interface. It adds perception, localisation, obstacle avoidance, waypoint and route execution, area and building scanning, follow-me behaviour and assisted teleoperation without replacing the platform.

Ground robotics autonomy box mounted on an unmanned ground vehicle

Overview

What the Ground Robotics Autonomy Box does

The Ground Robotics Autonomy Box is a self-contained autonomy computer that mounts onto a UGV or robotic dog already in service. It brings its own perception, localisation, planning and mission execution, and drives the platform through its existing vehicle or robot interface.

Units keep the platforms they have trained on and gain autonomous navigation, obstacle avoidance, traversability assessment, waypoint and route execution, building and area scanning, follow-me and leader/follower behaviour, and assisted teleoperation that prevents collisions when a human is driving.

The box exposes ROS interfaces for integrators, streams video and position back to the operator, and reports mission state to NOMAD Tactical Robotic C2 so a ground robot becomes one asset in a coordinated multi-robot picture.

Capabilities

Core capabilities

  • Autonomous navigation, indoor and outdoor
  • Obstacle detection and avoidance
  • Traversability assessment
  • Route and waypoint execution
  • Building scanning and clearing patterns
  • Area scanning and autonomous patrol
  • Follow-me and leader/follower behaviour
  • Assisted teleoperation with collision prevention
  • GNSS-denied operation
  • ROS / ROS 2 integration
  • Vehicle and robot interface integration

How it works

How the Ground Robotics Autonomy Box works

From sensor input to actionable output in four steps.

Step 01

Mount the Autonomy Box

A self-contained edge computer and sensor payload is installed on the host UGV or robotic dog and connected to its existing vehicle or robot interface.

Step 02

Build a local map and plan

LiDAR, depth cameras and inertial sensing create a local map. The planner generates routes and waypoints without relying on GPS.

Step 03

Avoid obstacles and assess terrain

The robot detects unexpected obstacles and evaluates whether ground ahead is traversable, adjusting the path in real time.

Step 04

Execute and report

The platform runs building scans, patrols, follow-me or standoff inspection missions while streaming video and position back to NOMAD or ATAK.

Use cases

Where it's used

Typical situations this system is deployed in, and what it solves in each one.

Tunnels, buildings and places people should not enter

The robot maps and moves through confined spaces without GPS, so the inside of a building or tunnel can be seen before anyone goes in.

Robot patrols

A wheeled, tracked or legged robot repeats a patrol route on its own and avoids obstacles along the way, freeing people from routine rounds.

Standoff inspection

Suspicious objects and stretches of road can be approached and inspected with the operator at a safe distance.

Carrying loads and following the team

The robot follows a person or another vehicle and carries equipment, with collision prevention on board.

Integration

Interfaces and integration

Built to drop into an existing architecture — standard interfaces, standard hardware, no forced replacement of fielded systems.

  • Control interface: ROS / ROS 2, CAN, serial or vendor SDK per platform
  • Sensors: LiDAR, stereo/depth cameras, IMU, GNSS where available
  • Power and mounting adapted to the host platform
  • Video and telemetry over the Tactical Video & Teleoperation Suite
  • Mission assignment and status reporting to NOMAD and ATAK

FAQ

Common questions about the Ground Robotics Autonomy Box

Plain-language answers for integrators, procurement teams and operators.

What platforms can be upgraded?
Tracked and wheeled UGVs, robotic dogs and other legged platforms, and small tactical robots with CAN, serial or ROS control interfaces.
Does it work indoors without GPS?
Yes. LiDAR, depth cameras and IMU-based SLAM let the robot navigate tunnels, buildings and other GNSS-denied spaces.
What sensors are required?
Typical configurations use LiDAR, stereo or depth cameras and an IMU. GNSS can be used where available but is not required.
Can a human still drive the robot?
Yes. Assisted teleoperation keeps the human in command while the autonomy stack prevents collisions and limits unsafe commands.
How is it integrated into our system?
The box exposes ROS / ROS 2 interfaces and connects over CAN or serial. It reports mission status and video to NOMAD and ATAK.
Is it suitable for EOD or hazardous inspection?
Yes. Standoff inspection, suspicious-object approach and confined-space scanning are common uses, keeping the operator at a safe distance.

Discuss your platform and mission

Tell us the platform, sensor and mission. We will walk you through integration options and arrange a technical demonstration.