A One Month update on my Learning Path
As I’m writing this, I’m waiting to Meet Sebestian Thrun and then we’d discuss my cool ideas with Larry Page regarding the Kitty Hawk Flyer.
I might get to do a Test Flight of it as well.
Edit: Earlier the Flying Car Nanodegree was a Two-Term Nanodegree. It has been changed to a Single Term Nanodegree, which means: I’m Officially a Flying Car Engineer now. 😎
If you’re thinking this is what I’ve achieved none of this is true. This is my Dream from Last Night nothing remotely close to my reality.
It has been One Month since I publicly shared my Machine Learning Path and this post is to share a huge Milestone.
This Valentine’s day I got an accepted into the Flying Car Nanodegree’s Pioneer batch and the last few months which have just flown past.
This is reflection of my Journey:
The Dream
- I’ve been very vocal about my Dream of Flying Cars and I’m a fan of Udacity. So I had to sign up for the ND. I just had to.
- My expectations were realistic-there are no flying cars. For a while there will not be any either.
- But, if there’s even one possible simulation of this universe where they will exist. I’m definitely not going to miss out on that.
- TL;DR: I expected to learn more about How things fly rather than just how to build a flying car
Challenge Accepted!
The Challenge
-
The main challenge for me was Time.I was pursuing the Self Driving Car Nanodegree, the Google India Scholarship, and now the Flying Car ND too!Needless to say, I’m a Full Time University Student too.
- I was also trying to do a Fast AI marathon during the same time.
- And I had hoped to start practising on Kaggle.
- Did I mention I was the youngest student in the Complete Batch?
- I was hoping to put in all my waking hours into studies if not more. I did that, but still I was short on time.
Path
The Path was filled with way too many ups and Downs. I had taken quite a challenge and No Excuses, I failed at many of the goals.
- Failed to Sleep a few nights, quite often.
- I started Blogging about my Flying Car Path.
- I was even offered to write on the Udacity Blog since My Blogs were doing great
-
I put out a few posts that I’m very proud of:- Diary of a Flying Car Engineer #0- Quadcopter as the Test Platform- Quadcopter Components Explained- Quadcopter Physics Explained- ReaderBot: An Event Driven ChatBot
- However, Due to a personal problem, I couldn’t watch the lectures for 3 weeks, I couldn’t keep up and I lost that chance.
- I failed to finish my Fast AI marathon on time. (I’m still behind)
- I failed to finish my PyTorch Blog posts on time. (Also behind on this, No excuses)
- No Excuses, I took the challenge, got back with double the speed.
- And Finally, I graduated the Term on 30th June, 2018.
Outcome
- 4 Months since I joined, I graduated the Term! Still retaining the Pioneer batch badge.
- I have learned many things that I will surely get to.
- Thanks to the Amazing Community that was built, I met many many amazing people. I was the youngest noob, so I had everything to learn.
- Picture this, 4 weeks ago: I knew nothing about Aviation or Flight Controllers or Drones.
- Now, I’ve been considered fit enough to be an Admin for the Slack/Flying Car Community- A Peer Moderator.
- I’m close to completing my Self Driving Car ND too!
The Flying Car Nanodegree (Earlier Term 1)
The real deal.
Here’s a 50 feet view of the Complete Course:
The first term consists of seventeen lectures and four projects. Personally, I’m a fan of the Production quality. I think they are the MKBHDs of Education when it comes to the Video Production.
MATH! The Nanodegree is very very Math Heavy. If you’re a Math Person, you’d be much more comfortable on here. The Programming too, is very challenging.
Udacity mentions the Nanodegree to be Advanced, I believe that is the toughest Nanodegree. The Code includes implementing Algorithms, and Math! and much learning of about physics. I really wish I had been a better student during school.
Term:
- Introduction: Introduction to Autonomous Flight and the Basics.
- Path planning: Path planning introduction and beyond: from A* on a grid to a real-world 3D planning.
- Control: Vehicle dynamics, control architecture(cascade PID controller) and full 3D control.
- Sensor Fusion and Estimation: Introduction to estimation, sensors(GPS, IMU, etc.) and extended/unscented Kalman filter.
Projects:
Link to My Github: https://github.com/init27/
Backyard Flyer Project
We used Event Driven programming to create a basic Flight Controller.
This was very basic, I assumed the ND is a piece of cake- I was very wrong.
My Solution: https://github.com/init27/Flying-Car-ND-Backyard-Flyer
Path Planning
We start from A* Search Algorithm in the 2D world to 3D motion planning using random sampling, graphs.
This project is a continuation of the Backyard Flyer project, we coded path planning through an urban environment in 3D!
My Solution: https://github.com/init27/Flying-Car-Nanodegree-Project-2-3D-Motion-Planning
Control
MATH!
The Math in these lectures literally made gave me nightmares! This part had the toughtest lessons and Project.
The lectures start with vehicle dynamics.
The Key things taught were:- PID Controller. - Cascade PID Controller applied to a Drone.
This lecture was so difficult that they eventually relaxed the requirements and made us submit just the C++ implementation and removed the need for the Python submission.
Project:
The project consists of the implementation and tuning of a cascade PID controller for drone trajectory tracking.
My Solution: https://github.com/init27/Flying-Car-Nanodegree-Project-3-Control-of-a-3D-Quadrotor
Sensor fusion and Estimation
This was divided into:
- Intro to Estimation.
- Sensors.
- Kalman Filters.
- GPS.
Project:
We develop the estimation portion of the controller used in the CPP simulator.
There are various scenarios that we had to pass:- Implement a better rate gyro attitude integration scheme- The prediction step should include a correct calculation of the Rgb prime matrix, and a proper update of the state covariance.- Implement the magnetometer update.- Implement the GPS update.
My Solution: https://github.com/init27/Flying-Car-Nanodegree-Project-4-Building-an-Estimator
Community
I was in the Pioneer batch. Needless to say I was going to be surrounded by many amazing people.
I need to take a little minute to mention all of the amazing people that had very much inspired me and allowed me to be exposed to thinking and mindsets that are simply not to be seen anywhere on my Teenaged college campus.
Many of these were people with experience more than my age!
Ankur Kaul: He has a brilliant website on his cool AUV hacks: http://www.allthatido.com
Aysin Tasdelen: She is a lecturer at Istanbul University, we are competing against each other for Nanodegree batches. (I’m losing, terribly). She had been an amazing help and mentor to me during the course.
A Master Applied Physics & Aerial Robotics Engineer: Ten winning teams were announced for the first round of the $2 million Boeing Gofly competition to develop a compact personal VTOL aircraft for a single person capable of traveling up to 20 miles. Hist team was one of the winners headed to the second round!!!
Brian Ringley:Another Researcher! I was obviously feeling completely out of the place. Be sure to check out his amazing work.
Darien Martinez Torres:A Professional Software Engineer. He was is of the same league as all the other mentioned people. Also, one of the first few to graduate. (He has experience of years more than since when I got interested in Code!)
A Stanford Scholar! I had a classmate from Stanford. Xinjie is a Deep learning data scientist and he speaks Math and Code! His discussions really had an extra amazing element to them.
Also, A special and big thanks to Tucker Dunn: Our Student Experience Lead. Thanks to his Tweets and constant updates and involvement in the discussions we were not just on speed with the curriculum but also on the Industry News, the hiring landscape and the events too!
Final Mention: My Roommate, Rishi K Bhalodia. I’m known to be quite the introvert and geek and I socialize with limited people. However, we’ve been a great team. And this guy has convinced me that we’ll build a Flying Car in India someday, if there isn’t a single hiring company-we’ll be the leaders in India.
What’s Next For Me?
I need to fix all the goals that I have missed
- Complete the Fast AI marathon.
- Complete my PyTorch Blogs.
- Complete the SDCND Term (I’m close). I’m also Cheating with the Computer Vision Nanodegree on the Side.
I will not be pursuing the Term 2 right away, I’m losing the Pioneer badge but I feel I want to explore more about DL, ML. I will come back to it, later.
Edit: (Re:) The Term 2 has been cancelled by Udacity and even though I was hoping to take it later, it will not take place. I’m now a Proud Flying Car Engineer.
Once I’m a little more knowledgeable and I’ve “built” something or made a submission to kaggle that actually did well.
But I promise you this, I’ll fly a Flying Car before I die. Yes, I hope to live long 😎
Note:
- During my Freshman and Sophomore Year I was part of an AUV: Autonomous Underwater Vehicle Team.
- I spent my Third Year of studies learning about Self Driving Cars and Autonomous Aerial Vehicle.
Air, Water and Land is done. I’ll keep my journey updated about what comes next.
Again, I’m not an Expert. I’ve barely scratched the Surface but I’m enthusiastic and optimistic and this my journey, shared with you; Unfiltered.
Also, As I’m writing this-I’m waiting for my Flight to Taiwan. I’ve been selected for a fellowship for a two weeks AI Training programme at Asia University.
Oh, How much I wish I had my De-Lorean!
Your, Flying Car Engineer and Neighbourhood Noob.
If you found this interesting and would like to be a part of My Learning Path, you can find me on Twitter here.