Integration of Cloud Computing and Artificial Intelligence in Driverless Car Using High Performance Behavioural Cloning

Abstract

Research on autonomous vehicle system has risen in the past couple of years. It has posed challenges to the researchers to develop an understanding about the real time scenarios. Deep Learning has demonstrated its extraordinary computational potential by transcending its abilities into more complex areas, where pattern matching, image recognition and behavioral cloning plays a vital role.

The system consists of Image Processing and analyzing of the training data into behavioral cloning of the vehicle in a simulated environment. A control algorithm responsible for consolidating the sub systems calculations of the correct steering angle is used to keep the vehicle within the lane markings of the road. The solution proposed requires a better data collection and data interpretation. Addition of cloud computing fastens the data calculation and hence improves the performance of the system.