Welcome to the Hello GRID tutorial! This guide will walk you through setting up and running your first GRID session using the GRID Enterprise platform.
By default, all commands in this tutorial are executed on the local machine. If you are using a remote machine, you must specify it explicitly using the @<machine_name> syntax.

​
Getting Started

In this tutorial, you will:
  • Initialize the GRID Enterprise platform.
  • Set up and start a GRID session using the AirGen simulator using the CLI.
  • Access the simulation, and write some sample code to capture and visualize data.
  • Stop and terminate the session.

​
Setup and Initialization

​
Step 0: Ensure prerequisites

Before starting this tutorial, ensure that you have GRID Enterprise installed on your local machine. If you haven’t already installed GRID Enterprise, refer to the installation guide for detailed instructions. After installation, please make sure your ‘resource configuration’ file is set up correctly - you should have received this file during onboarding.

​
Step 1: Initialize the GRID Platform

The first step is to log into the GRID platform and initialize the containers. Open your terminal and run the following command: 
grid


 .d8888b.  8888888b.  8888888 8888888b.
d88P  Y88b 888   Y88b   888   888  "Y88b
888    888 888    888   888   888    888
888        888   d88P   888   888    888
888  88888 8888888P"    888   888    888
888    888 888 T88b     888   888    888
Y88b  d88P 888  T88b    888   888  .d88P
 "Y8888P88 888   T88b 8888888 8888888P"

General Robot Intelligence Development Platform - Enterprise version
Developed by General Robotics, Inc. (c) 2025

Loading resource configuration from /home/grid/.grid/resource_config.json...
Started container server on 0.0.0.0:8060
Samples currently at /home/grid/.grid/samples
Type 'help' or 'license' for more info.

GRID #
You can sign in using the provided credentials. Login using the command login and then initialize the containers using init. 
GRID # login
Logging in to General Robotics - GRID registry...Login successful!
GRID # init
Please choose the simulator you wish to use:
1: airgen
2: isaac
Enter your choice: 1
Starting GRID containers for airgen on local...
Spinning up containers with profile airgen...
WARN[0000] The "OPENAI_API_KEY" variable is not set. Defaulting to a blank string.
[+] Running 3/3
 ✔ Container grid_core        Started0.5s
 ✔ Container grid_sim_airgen  Started0.5s
 ✔ Container grid_server      Started0.6s
Checking container statuses...
grid_core: ✓
grid_server: ✓
grid_sim_airgen: ✓
Containers are active.
Setting up assets... Done.
This command spins up the necessary containers on your local machine, preparing the environment for your session.
If the containers do not already exist, the images will be downloaded, and the containers will be started. Please note that this process may take a few minutes.

​
Starting Your First Session

​
Step 2: Start Your First GRID Session

Let’s now start a session that runs a pre-configured simulation. Execute the following command in your terminal: 
session start hello_grid

No session configuration was passed. Using a sample configuration from ~/.grid/sample_session_config_airgen.json...
2025-02-11 23:33:59,542 - INFO - Initializing session
Downloading assets...
File already exists at destination. Skipping download.
2025-02-11 23:33:59,774 - INFO - HTTP Request: POST http://127.0.0.1:8000/start_session "HTTP/1.1 200 OK"
response: Initializing session...
response: Downloading content..
response_end: Session has been started successfully
Session started successfully.
2025-02-11 23:34:10,467 - INFO - Session started successfully.
This starts a session named hello_grid. Because we did not explicitly specify a session configuration, GRID will use a sample configuration file to start the session. As the console output shows, this configuration file resides at ~/.grid/sample_session_config_airgen.json.

​
Step 3: Open the Simulation

After the session starts, open the simulation in your browser with the open command: 
open sim @local

Opening sim for session hello_grid in default browser at http://127.0.0.1:3080
Simulation running successfully in the browser

​
Step 4: Open the Notebook Interface

You can also open the notebook interface for the session with: 
open nb @local

Opening nb for session hello_grid in default browser at http://127.0.0.1:8890
Notebook interface opened in the browser

​
Step 5: Open Visualization

Similarly, to visualize the data using rerun, use the following command: 
open hello_grid viz

Opening viz for session hello_grid in default browser at http://127.0.0.1:9090/?url=ws://127.0.0.1:9877
Visualization interface opened in the browser

​
Step 6: List Active Sessions

To check all active sessions, use the following command: 
session list

grid> session list
2024-09-10 17:10:53,338 - INFO - HTTP Request: GET http://localhost:8000/is_idle?session_id=hello_grid "HTTP/1.1 200 OK"
+--------------+-----------+----------------------------------+
| Session ID   | Node IP   | Last Active Time                 |
+==============+===========+==================================+
| hello_grid   | localhost | 2024-09-10T23:42:51.674976+00:00 |
+--------------+-----------+----------------------------------+

​
Create a notebook

Now that we have the session running, let’s create a notebook to interact with the simulation. To do this, go to the notebook interface and click on New in the top right to create a new notebook.

​
Data Capture and Visualization

With our development environment setup, let’s go ahead and write a simple script to capture data and visualize it on rerun. 
from grid.robot.wheeled.airgen_car import AirGenCar
robot = AirGenCar()

import airgen, rerun as rr 
image = robot.getImage("front_center", "rgb")

from grid.utils.logger import log 
log("rgb_image", image)
The expected output of the above code would be the image captured by the drone’s front center camera and displayed on rerun. Data captured by the drone and visualized on rerun

​
Step 7: Try to use an AI model

You can also use pre-trained AI models for computer vision, vision-language, SLAM, point tracking, and more. To use an AI model, you can import it and use it in your code. Let us try to use an object detection model to detect objects in the image captured by the robot. 
from grid.model.perception.detection.owlv2 import OWLv2

det_model = OWLv2(use_local=True)
det_model.run(image.data, "ball")
AI model output visualized on rerun

​
Step 7: Stop the GRID Session

Once you’re finished with the session, stop it with: 
GRID # session stop hello_grid

Stopping session hello_grid ...
2025-02-12 00:09:49,354 - INFO - Terminating session
2025-02-12 00:09:53,568 - INFO - HTTP Request: POST http://127.0.0.1:8000/terminate_session "HTTP/1.1 200 OK"
Session stopped successfully.
2025-02-12 00:09:53,569 - INFO - Session stopped successfully.
This will stop the hello_grid session without terminating the containers.

​
Step 8: Terminate the GRID Containers

Finally, to clean up your environment, terminate the containers by running: 
GRID # terminate

grid> terminate
Stopping containers using docker-compose...
[+] Running 3/3
 ✔ Container grid_sim_airgen  Removed10.2s
 ✔ Container grid_server      Removed10.2s
 ✔ Container grid_core        Removed10.5s
Checking container statuses...
grid_core: ❌
grid_server: ❌
grid_sim_airgen: ❌
Containers stopped successfully.
This stops all running containers on your local machine.

​
Congratulations!

You’ve completed the Hello GRID tutorial. You’ve successfully logged in, started a session, opened a simulation, and cleaned up your environment. This is the beginning of what you can achieve with GRID Enterprise. Explore the advanced features like running sessions on remote machines, mounting custom storage, and integrating your own AI models.

​
Next Steps

  • Now that we have successfully set up the development environment, you can explore the AirGen, Isaac, and AI Models sections to dive deeper into the capabilities of the GRID Enterprise platform.