Experiments in Autonomous Driving Through Imitation Learning

TL;DR

This paper explores various supervised learning methods to develop an autonomous vehicle using RGBD camera data, focusing on imitation learning despite challenges with data imbalance and initial adversarial attack research.

Contribution

It demonstrates the application of imitation learning techniques for autonomous driving and discusses the difficulties faced with data imbalance and model training.

Findings

Limited effectiveness of approaches due to unbalanced data

Challenges in training the vehicle with initial methods

Focus shifted from adversarial attacks to imitation learning

Abstract

This report demonstrates several methods used to make a self-driving vehicle using a supervised learning algorithm and a forward-facing RGBD camera. The project originally involved research in creating an adversarial attack on the vehicle's model, but due to difficulties with the initial training of the car, the plans were discarded in favor of completing the imitation learning portion of the project. Many approaches were explored, but due to challenges introduced by an unbalanced data set, the approaches had limited effectiveness.