I was selected for term 1 of self-driving cars nanodegree through KPIT scholarship. In the nanodegree, we implemented several projects like lane line detection, traffic signal classifier, behavioral cloning, and sensor fusion via extended kalman filters. Following is the GitHub link for the codes that I implemented during the course.
Of particular interest for me was the behavioral cloning project. In this project, our goal was to train a network using Keras which can successfully emulate human driving behavior autonomously in a simulated environment. I built a modified CNN based on NVIDIA’s End to End Learning for Self-Driving Cars paper.