Perhaps no invention has influenced the 20th century more than the automobile. In fact, there are presently over one billion cars on the road worldwide. Despite the importance of the automobile today, cars are grossly inefficient when it comes to resources such as energy and human productivity. In our country, traffic jams in major cities alone account for Rs. 1.44 Lakh Crore loss to the economy. There are 370 traffic related deaths in India every day. The societal costs of our wasteful utilization of cars are truly staggering!
Self-driving cars (SDCs) have been pegged as one of the fastest growing car market in the world. These cars are currently undergoing a lot of training in controlled environments. Developed nations have inculcated a culture of following traffic rules. So too are the quality of roads in these countries. With a disciplined road populace, it is conceivably easier to come up with data sets and models needed for SDC deployment. Traffic management in India is a nightmare! Add to that the discipline or the lack thereof, of riders and drivers on the road, the lack of strict enforcement of speed limits, and the rampant lane-jumping, signal-jumping, and helmetless violations.
As robotics researchers and technology enthusiasts, the idea of self-driving car intrigues us immensely, more specifically, the kind of data sets and learning models necessary to train such a machine in a setting as complex and anarchic as Indian roads. We believe that progress in this direction will solve many fundamental challenges faced by SDCs on Indian roads and around the world.