An augmented geo-tagging racing challenge platform.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
AERO_HackTM
docker
speedfeed
.dockerignore
.gitignore
Makefile
README.md
data.json
manage.py
requirements.dev.pip
requirements.pip

README.md

SpeedFeed

An augmented geo-tagging racing challenge platform.

People on the project: @jrocketfingers, @slymaximus

Category: Robotics/Automotive

Table No. 6

Inspiration

Ain't seen nothin like this yet

What it does

Using vehicle dynamics data gathered from an OBD2 dongle, and continental's long range radar and a camera, we can figure out high scores on 0-100 runs in an area, top speeds in an area, and times on predefined routes (technically races). Due to raw data provided by continental we could also figure out the congestion during the run, and grade the timed routes with a score depending on the congestion encountered (and hence the driver skill). As this wouldn't be widely available, it's a thing we'll play with during development but might not be considered as serious developmental effort.

How it's built

It's not yet. But we're gonna be using a bunch of spatial db stuff as soon as we finish hacking the data over to the aggregation server. Basically instead of an OBD2 dongle and a smartphone app, right now we'll use a laptop and a mobile app, with laptop acting as an OBD2 data transciever.

AERO_HackTM - aero_hacktm_runtask.dll The aero_hacktm_runtask.dll library collects ADAS information within a Continental's ADAS Test Car and it's uploading it asynchronous to an server via HTTP. The DLL uses Casablanca REST API version 1.1.0 (older one, because of Visual Studio 2012 compatibilities request).

Challengers we ran into

Besides crappy internet?