Incorporating AI into current autonomous vehicle design

Abstract

Many hundreds of thousands of decisions are made by a driver during their regular commute to work. Determining the speed of the vehicle, merging with traffic, watching for road signs and traffic lights, etc, are just a few. Autonomous aspects are starting to be implemented into the present-day market, but enhancing the level of automation comes along with considerable challenges. One is dealing with different environmental conditions, such as rain, snow, fog, etc, each having distinct characteristics that make it tricky for sensors to do their proper job. Another is dealing with radar interference due to the size of precipitation that can completely scatter signals and discombobulate a vehicle. The last is upgrading the capabilities of autonomous vehicles presently being sold in the market by changing their perceptions and motion planning. Incorporating the many methods of artificial intelligence, such as machine learning, deep learning, and reinforcement learning, can make achieving these challenges a lot more efficient.