https://github.com/tlkh/vision-autonomous-driving
Autonomous Navigation using OpenCV
https://github.com/tlkh/vision-autonomous-driving
arduino jetson-tx1 opencv python raspberry-pi robotics ros
Last synced: 2 months ago
JSON representation
Autonomous Navigation using OpenCV
- Host: GitHub
- URL: https://github.com/tlkh/vision-autonomous-driving
- Owner: tlkh
- License: mit
- Created: 2017-12-29T12:39:06.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2019-01-23T14:49:36.000Z (over 7 years ago)
- Last Synced: 2024-12-29T00:12:21.714Z (over 1 year ago)
- Topics: arduino, jetson-tx1, opencv, python, raspberry-pi, robotics, ros
- Language: Python
- Size: 5.36 MB
- Stars: 2
- Watchers: 3
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Vision-based Autonomous Driving
An attempt to do robust self-driving using vision techniques on a physical platform.
### Platform
* Raspberry Pi 3 (Jetson TX1 - Planned)
* Chassis: Wild Thumper (differential drive)
* Pure Python
### Hardware Setup
* 6 DC motors driven by 2S Lithium-polymer cell
* Sabertooth 2x25A Regenerative Motor Driver ([link](https://www.dimensionengineering.com/products/sabertooth2x25)) (Serial mode)
* Arduino Nano to interface between Raspberry Pi and Sabertooth
### Software Implementation
* OpenCV for vanishing point detection
* (simple walkthrough: [notebook](https://github.com/tlkh/opencv-experiments/blob/master/Hough%20Line%20Transform.ipynb))
[Video](https://www.youtube.com/watch?v=fQrRkU2MiVI)
### In Progress
1. Camera calibration ([notebook](https://github.com/tlkh/opencv-experiments/blob/master/camera_calibration/calibration.ipynb))
### TODO
1. PID control
2. Integrate into ROS
3. Add more robust autonomy with mapping and localisation
4. Acceleration with CUDA on Jetson TX1