Newer
Older
# The Donkey
A small utilitarian self driving vehicle.
* Project goal: The the sidewalk self driving vehicle (auto).
Here are [instructions](get_started.md) and part lists to build your own car.
All of these examples assume you're connected to your donkey via SSH and you've activated the virtual environment.
### Activate the virtual environment
```
cd code/donkey
source env/bin/activate
```
python manage.py record --filerecorder --webcontrol
2. Go to '<pi_ip_address>:8889' in your browser to control the car.
3. End training by typing `Ctrl-c` in your ssh sesion.
### Create a predictor for the route.
```
python manage.py auto --indir <indir path>
```
Where indir is the directory of the images used to train the predictor.
- [ ] Threadsafe image capture (for webserver + recorder) http://www.pyimagesearch.com/2015/12/28/increasing-raspberry-pi-fps-with-python-and-opencv/
- [ ] Update vehicle to drive given manual input.
- [ ] Try loading tensor flow on Raspberry Pi
- [ ] Train Convolution network from numpy arrays
Email Adam, Keven and Jeff about the Jan 22nd Race