Challenge #1: 3D Model for Camera Mount
Udacity Self-Driving Car Challenge #1
The Udacity Self-Driving Car
As detailed in this post, a critical part of our process in launching the Self-Driving Car Nanodegree program is to build our own self-driving vehicle. To achieve this, we formed a core Self-Driving Car Team with Google Self-Driving Car founder and Udacity President Sebastian Thrun. One of the first decisions we made together? Open source code, written by hundreds of students from across the globe!
Now, we’re breaking down the problem of making the car autonomous into Udacity Challenges, and this was our first one!
Our Point Grey Blackfly cameras, while amazing, only provided simple tripod mounts, which don’t provide support for the lens when mounted in a car. Cars are often bumpy, unpredictable, and the data we record must be consistent, otherwise hours of driving are rendered useless. We also noticed (via YouTube videos) that both Nvidia and Comma use 3D printed hardware for mounting their cameras, which inspired the first beta challenge!
The challenge: design a mount to support the lens and camera body that can be mounted using standard GoPro hardware.
We received a ton of amazing submissions, and had an intriguing problem. How could we test that the submissions actually worked, and worked well? Unlike our code submissions for other challenges, there wasn’t necessarily a script or benchmark we could provide, so we did it the old fashioned way: we printed and tested each one!
The contributors to this challenge went above and beyond, sharing ideas and schematics and more with one another, which ultimately led to an amazing result.
After much iteration and printing, the following design—contributed by amazing student Haden Wasserbaech—proved to be the most stable:
Note from Haden
I am Haden Wasserbaech, I am an undergrad student that loves designing, building, and programming autonomous robots. I am currently leading a team of students to make a robot to compete in the Intelligent Ground Vehicle Competition (IGVC). In this competition a robot must navigate between GPS waypoints while avoiding obstacles and staying within white lines. Robots typically use a range of sensors including LiDAR, GPS, IMU, cameras, and wheel encoders.
For this challenge I wanted to make a camera mount that was somewhat minimal yet also very rigid, this is why I added large cutouts to the design while also keeping it fairly thick. I also added support for the heavy lens of the camera, but I had to make sure that the lens would still rotate. In order to make everything fit I made the CAD model slightly larger than the camera to allow for 3D printing tolerances. I also had some trouble finding good documentation on some of the dimensions of the camera and lens, but with the help of the other participants we managed to figure it out!
This challenge was awesome because I got to use my skills with CAD to design a mount that will firmly hold the camera. This will help Udacity to produce high quality data for students to use in this Nanodegree! Additionally since the CAD model is open source anyone can print or modify the design if they would like.
The results of this beta challenge only reaffirmed our belief that we could build an open source autonomous car, powered by students from around the world. We can’t wait to see the results of Challenge #2!