로봇학회논문지 제14권 제2호 (통권 제52호) (p.122-130)

유니티 실시간 엔진과 End-to-End CNN 접근법을 이용한 자율주행차 학습환경

Autonomous-Driving Vehicle Learning Environments using Unity Real-time Engine and End-to-End CNN Approach
키워드 :
Autonomous Shuttle Vehicle,Artificial Intelligence,Virtual Environment,Behavior Learning

목차

Abstract
1. Introduction
2. Open-Source Simulators
  2.1 GAZEBO
  2.2 TORCS
  2.3 CARLA
  2.4 AirSim
3. Autonomous Vehicle Simulator
  3.1 The process of importing Shuttle cart in Unity
  3.2 Environments Numbers
  3.3. Weather Conditions and Different Road Layout
  3.4 Types of Sensors
  3.5 User Interface of Simulator and Operation
4. End-To-End CNN based Self-Driving
  4.1 Data Collection and Preprocessing
  4.2 Architecture of CNN Training
5. Results
6. Conclusion
References

초록

Collecting a rich but meaningful training data plays a key role in machine learning and deep learning researches for a self-driving vehicle. This paper introduces a detailed overview of existing open-source simulators which could be used for training self-driving vehicles. After reviewing the simulators, we propose a new effective approach to make a synthetic autonomous vehicle simulation platform suitable for learning and training artificial intelligence algorithms. Specially, we develop a synthetic simulator with various realistic situations and weather conditions which make the autonomous shuttle to learn more realistic situations and handle some unexpected events. The virtual environment is the mimics of the activity of a genuine shuttle vehicle on a physical world. Instead of doing the whole experiment of training in the real physical world, scenarios in 3D virtual worlds are made to calculate the parameters and training the model. From the simulator, the user can obtain data for the various situation and utilize it for the training purpose. Flexible options are available to choose sensors, monitor the output and implement any autonomous driving algorithm. Finally, we verify the effectiveness of the developed simulator by implementing an end-to-end CNN algorithm for training a self-driving shuttle.