Skip to content

CogitoNTNU/cogitron

Repository files navigation

Cogito Cogitron

GitHub Workflow Status (with event) GitHub top language License: GPLv3 Project Version

Cogito Project Logo
📋 Table of contents

Description

Longterm robot project created by the members of Cogito Cogitron.

The project is largely inspired by the work done by HuggingFace with lerobot.

Getting started

Prerequisits:

  • Ensure that Git and Python is installed on your machine.

Download the repository by running the following:

git clone https://github.com/CogitoNTNU/cogitron

After cloning the repository you can change into the directory by running:

cd ./cogitron

Connect to existing enviroment

You can search for devices in year network by using the nmap script run by:

./cogitron/scripts/find_server.sh

It is assumed that nmap is installed and that you are running the script in a linux enviroment. If you are on windows you can install WSL and run it from there.

If you get a Permission denied error after trying to execute a script you can use the following command to allow execution of the script:

sudo chmod +x ./cogitron/scripts/find_server.sh

After executing the script you will obtain a list of IP-addresses with port 22 open. These will likely have an active SSH-service.

You can connect to a SSH server by running

ssh cogitron@<ip-address>

where you replace the "<ip-address>" with the actual ip address you are trying to connect to. You can then type in the password to the SSH session. This gives you a Bash shell to the machine.

Setup enviroment on a new machine

Note: It is required to run the code on Linux.

To set up a Python enviroment run the following in the terminal:

python3 -m venv venv

Then activate it:

. venv/bin/activate

Install the requirements with:

pip install -r requirements.txt

If you want to connect to connect to the machine with SSH you have to setup a SSH server on the machine.

Compile

Installing the following packages might be required if they are not present on your machine:

apt install libcap-dev libsystemd-dev python3-dev

You can run the following command to build and install the project. Replace <wheel-name> with the generated wheel filename:

python3 -m build --outdir dist; pip install dist/<wheel-name>.whl --force-reinstall

Scripts

Script for training and running the AI models can be found under the folder:

./cogitron/scripts/

For more information see the documentation for lerobot

License


Distributed under the GPLv3 License. See LICENSE for more information.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •