Autonomous Driving through Chaotic Traffic in India via Reinforcement Learning
This video showcases autonomous driving through chaotic traffic-scene in an open environment, meaning there's no traffic-rule being followed here, which makes prediction of obstacle's behaviour a very difficult task.
This demo highlights integration of our autonomous campus driving and parking software with reinforcement learning based learned policy for autonomous navigation in unknown and unseen environments (a research front initiated by Sanjeev Sharma, Founder and CEO) to allow autonomous vehicle to navigate through the stochastic clutter on road, and reach the parking spot with near end-effector constraints execution.