TJAssist : Car Autopilot System to Assist Driver on Traffic Congestion

This project is about a car autopilot system which take over driving during traffic jam condition. As traffic jam is very serious in Malaysia especially in major cities like Kuala Lumpur, this system will be able to help the driver to utilize the time during traffic jam to conduct other works. The r...

Full description

Saved in:
Bibliographic Details
Main Author: Chong , Vun Vui
Format: Final Year Project
Language:English
Published: IRC 2015
Subjects:
Online Access:http://utpedia.utp.edu.my/15866/1/Chong%20Vun%20Vui_16523.pdf
http://utpedia.utp.edu.my/15866/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This project is about a car autopilot system which take over driving during traffic jam condition. As traffic jam is very serious in Malaysia especially in major cities like Kuala Lumpur, this system will be able to help the driver to utilize the time during traffic jam to conduct other works. The reason for building this system even though autopilot system already existed in the market is because the existing autopilot system in the market are too complex to be implemented into daily use car due to the high number of sensor used in the vehicle and high computing power required. To prove the validity of this project, a prototype of the system will be created with Lego Mindstorms EV3 Education Set. Increment methodology is used to build this prototype as it provide the ability to do rapid analysis and development. The prototype are equipped with the function to follow the car in front at straight road and slight turns. In case of sharp turns or any interference by the driver, the system will stop and return the driving control to the driver. Throughout the autopilot, the driver will be notified continuously on the decision made by the autopilot system via voice notification system. In case the car in front reverses, the system will stop the system and warn the driver about it. After a testing has been done, this prototype managed to achieve an average accuracy level of 89%