Introduction
Self Driving Car
AVYANA is an open-source project for autonomous self-driving cars. It is based off on ROS2 and is aimed for both dev and production environments on a variety of ground based vehicles.
Start by installing the AVYANA Meta Repository.
The Avyana repository is a meta repository, that interfaces all other required repositories. It is built in this manner, so that it becomes easy for us to build individual sub-systems, and it becomes easier for the end-user to debug. The main reason to choose ROS2 Humble as a starting point is that, we would like to work with NVIDIA suite of softwares such as ISAAC, ISAAC-ROS and Jetson Orin at a later point. We have settled up with Nvidia explicitly, as the tone of development towards Autonomous Vehicles and available open-source software and hardware is well in reach from Nvidia, and without the need for an FPGA Board. We plan of using an Nvidia Orin NX 32Gb board for edge-processing, and rely less on cloud connections except to relay data for Visualisation and training. This is done so that the Autonomous Vehicle, can work even in remote areas.
Get Started
Step-by-step guides to setting up your system and installing the framework.
Step-by-step guides to setting up the model.
Step-by-step guides to setting up the studio.
note
This guide is an active work in progress. The information provided is tentatively correct, and would be updated in regular intervals. If any information is required urgently, feel free to raise an request documentation issue.
Overview
The flowchart below is an overview, on how the system is set to be built.