Deep Neural Networks for Understanding Autonomous Vehicle Behavior and Control

Abstract

In recent years, technological advancements have made a promising impact on the development of autonomous vehicles. The evolution of electric vehicles, development of state-of-the-art sensors, and advances in artificial intelligence have provided necessary tools for the academia and industry to develop the prototypes of autonomous vehicles that enhance the road safety and traffic efficiency. The increase in the deployment of sensors for the autonomous vehicle, make it less cost-effective to be utilized by the consumer. This thesis focuses on the development of full-stack autonomous vehicle using the limited amount of sensors suite. The architecture aspect of the autonomous vehicle is categorized into four layers that include sensor layer, perception layer, planning layer and control layer. For the verification of the proposed system, the autonomous vehicle is tested in an unconstrained environment. The experimentation results show the efficacy of each module, including localization, object detection, mission and motion planning, obstacle avoidance, velocity replanning, lateral and longitudinal control. Further, in order to demonstrate the experimental validation and the application aspect of the autonomous vehicle, the proposed system is tested as an autonomous taxi service. A neural network-based controller (N2C) using behavioral cloning is designed as a surrogate to the classical controller that predicts throttle, brake, and torque command when trained with the manual driving data acquired from the controller-area-network (CAN) bus. In addition to complement N2C, an end-to-end neural network is designed for predicting the steering angle and speed using image data as an input as a replacement of a deep learning framework for motion planning. The experimental evaluation shows the efficacy of proposed frameworks in real-time and on the Udacity dataset showing better metric scores in the former and reliable prediction in the later case when compared with the state-of-the-art methods.