DeltaX Self-driving competition
This is the main information page of the competition
DeltaX Self-driving competition, is part of the DeltaX student competition. The task in this competition is to create a software solution that can drive a toy car on a small race track, based on visual inputs.
The competition includes prizes for winners. The main prize is 1000 euros.
The rules of the self-driving competition can be found here (updated 24.11): link
1. To participate, first please register by sending an email to ardi.tampuu@ut.ee.
2. As a next step you could take a look at the public dataset we have made available to all teams:
(link to data)
Data is organized by recordings. The most important data for each recording are the frames and the JSON file that contains timestaps for these frames and the timestamped commands given to the car by human driver. To better understand the data format check the documentation in next step. The data is cleaned - despite there being collisions in the video files, these sections are removed from the frames and JSON files.
3. Communication with the car is achieved via software provided by RCSnail, the documentation for controlling the car and a video eplaining the steps are below.
In the lower part of the later file you also see mentioned installation of RCSnail-AI-Lite. You can achieve control over the car by placing your image-to-commands processing in between the
#in https://github.com/martinliivak/RCSnail-AI-lite/blob/master/src/main.py line 39: frame, data = await recv_array_with_json(queue=data_queue)
and line 50: controls_queue.send_json(next_controls)
BONUS 1 Example video of driving and camera feed from the track in Delta: https://drive.google.com/file/d/1ccim7tzmPKkdyU-s4siVsu7TUhcdjSVU/view?usp=sharing
BONUS 2 - you might find useful to take inspiration from the work done with these cars by Martin Liivak.
Example video from the master's thesis of Martin Liivak :