Self-Thinking Machines in Autonomous Driving

EasyChair Preprint 1355

Abstract

A truly autonomous machine , therefore, needs to be able to learn and adapt its own models. A machine learning about itself and its environment is in the position of being an active part of the system it is trying to learn about; this situation draws interesting parallels with learning in human infants. A system is presented here that enables a machine to autonomously learn a model with no prior knowledge of its own system or the external environment. The autonomous machine sends out random motor commands to its motor system and receives information back from the vision system. This set of evidence is used to learn the structure and parameters of the machine in an environment which are then used as input to 3D Convolutional Neural Network, a machine learning technique, to create a its own internal model. This model can then be used to enable the autonomous machine to predict movements.

Keyphrases: Autonomous Car, Deep Reinforcement Learning, Self-Thinking

@booklet{EasyChair:1355,
  author    = {Poondru Prithvinath Reddy},
  title     = {Self-Thinking Machines in Autonomous Driving},
  howpublished = {EasyChair Preprint 1355},
  year      = {EasyChair, 2019}}