Unitree Go2 Setup¶

Full autonomous navigation, mapping, and agentic control on a real Go2. No ROS required.

Requirements¶

Unitree Go2 Pro or Air (stock firmware 1.1.7+, no jailbreak needed)
Ubuntu 22.04/24.04 with CUDA GPU (recommended), or macOS (experimental)
Python 3.12

Install¶

First, install system dependencies for your platform: - Ubuntu - macOS - Nix

Then install dimOS:

uv venv --python "3.12"
source .venv/bin/activate
uv pip install 'dimos[base,unitree]'

Run on Your Go2¶

First-time setup, connecting to wifi, finding robot IP¶

Use dimos go2tool to provision wifi and find the robot's IP. Skip if the robot is already on your network and you know its IP.

Power on the Go2. It advertises over BLE immediately.
Provision wifi (one-time per network):

optionally use discover to make sure robot is detected

dimos go2tool discover

configure wifi

dimos go2tool connect-wifi --ssid <wifi> --password <password>

Scans BLE and connects to the only robot it finds, or prompts you to pick if there are several.

Find the robot's IP:

dimos go2tool discover

Prints SOURCE NAME IP MAC SERIAL for every robot it sees over BLE and LAN. Export the IP:

export ROBOT_IP=<discovered_ip>

Pre-flight checks¶

Robot is reachable and low latency <10ms, 0% packet loss
```
ping $ROBOT_IP
```
Built-in obstacle avoidance is on. (dimOS handles path planning, but the onboard obstacle avoidance provides an extra safety layer around tight spots)

Ready to run dimOS¶

export ROBOT_IP=<YOUR_GO2_IP>
dimos run unitree-go2

That's it. dimOS connects via WebRTC (no jailbreak required), starts the full navigation stack, and opens the command center in your browser.

What's Running¶

Module	What It Does
GO2Connection	WebRTC connection to the robot. Streams LiDAR, video, odometry
VoxelGridMapper	Builds a 3D voxel map using column-carving (CUDA accelerated)
CostMapper	Converts 3D map → 2D costmap via terrain slope analysis
ReplanningAStarPlanner	Continuous A* path planning with dynamic replanning
WavefrontFrontierExplorer	Autonomous exploration of unmapped areas
RerunBridge	3D visualization in browser
WebsocketVis	Command center at localhost:7779

Send Goals¶

From the command center (localhost:7779): - Click on the map to set navigation goals - Toggle autonomous exploration - Monitor robot pose, costmap, and planned path

Agentic Control¶

Natural language control with an LLM agent that understands physical space:

export OPENAI_API_KEY=<YOUR_KEY>
export ROBOT_IP=<YOUR_GO2_IP>
dimos run unitree-go2-agentic

Then use the human CLI to talk to the agent:

humancli
> explore the space

The agent subscribes to camera, LiDAR, and spatial memory streams. It sees what the robot sees.