Self-driving cars are rapidly moving from a prototype to an everyday reality. Upwards of 90% of all car accidents continue to be caused in some way by a human error. The absence of a driver in production consumer vehicles have shown an decrease of collisions and fatalities. Leading companies in autonomous vehicles like Tesla and Google claim that Autopilot-enabled cars reduce car collisions by approximately 80%, and have covered 130 million miles without a fatality, compared to a US national average of one fatality every 94 million miles. In addition, self-driving cars can provide other benefits like increasing safety, reducing traffic, and lowering carbon dioxide emissions (up to 94% per mile with autonomous taxis). The technology of driverless cars include developing components such as traffic sign and light, lane, vehicle, and pedestrian detection and tracking. These features utilize the an on-board camera as input to perform their task. This project focuses on implementing vehicle recognition and tracking using soft-computing techniques.
To use tqdm progress bars with Keras on Jupyter, enable Javascript widgets for Jupyter.
jupyter nbextension enable --py widgetsnbextension
- Aldrin Balisi
- Kevin Deng
- Raymond Kam