# Drone Autonomy Crash Course | Carlos Gonzalez
Source: [https://www.cggonzalez.com/blog/index.html](https://www.cggonzalez.com/blog/index.html)
By: Carlos Gonzalez
*A review of modern drone modeling, state estimation, motion planning, and control\.*
## F\.A\.Q\.
##### What is this?
A short introduction to doing modeling, state estimation, motion planning, and control for quadcopters\. My goal with this is not to be an authoritative source but instead to serve as a “jumping off point” for people who are new to the topic\. Once you finish this you’ll have a good understanding of how a quadcopter can be made autonomous\. I’ve also included references in every post for those that want a more rigorous look at the topics that I discuss\. Some sections are still being completed, denoted \(WIP\), but they'll be done soon\.
##### Why did you make this?
Because something like this should exist… Seriously… When I was an undergrad I became obsessed with quadcopters and that led me to wanting to know how to make them autonomous\. However, even after reading all the research papers I could find I was just as confused as when I started\. Partly because academic research papers can be succinct to a fault, but also because few of them were self\-contained so trying to see the big picture was nearly impossible \(for me at least\)\. I just wanted a short, simple, and straightforward overview of the parts that go into making a quadcopter autonomous\. After a couple of years of putting the pieces together I decided that I should make a short guide that does just that so that newcomers to the topic don’t have as tough of a time as I did\.
##### Why didn't you just write a paper instead?
I actually did start writing a paper but it turned out to be over 80 pages long\. As a result I decided that breaking it down into pieces would make it a bit more digestible\. This also led to my writing being more casual which worked out well since my goal isn’t to be rigorous but to give more of an intuitive feel for the topics\. All of the posts include references where the topics are explored more rigorously\.
##### Who are you?
Hello\! I’m[Carlos](https://www.cggonzalez.com/)\. I’m a guy who’s really into quadcopters\.
A detailed technical article explaining drone physics, including coordinate systems, equations of motion, forces, and control, with references to the multirotor simulation framework.
Building a fault-tolerant RL octocopter from scratch, using MuJoCo simulation and PPO training with domain randomization to handle motor failures. The project aims to directly command motors via RL policy without PID loops, focusing on six failure classes.
DiffAero is a GPU-accelerated, fully differentiable simulation framework for quadrotor control policy learning that supports environment- and agent-level parallelism, multiple dynamics models, and customizable sensors. It enables robust flight policy learning in hours on consumer-grade hardware and is released as open-source.
This paper presents a two-stage learning pipeline that uses an attention-augmented gated recurrent network to estimate wind from onboard kinematics and then leverages that estimate in a reinforcement learning controller to improve quadrotor trajectory tracking under turbulent winds, reducing tracking error by 48% compared to a baseline.