Development of real-time location of shuttle bus with time arrival estimation using machine learning

Ahmad Jais, Muhammad Zulhusni (2024) Development of real-time location of shuttle bus with time arrival estimation using machine learning. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full Text) Development of real-time location of shuttle bus with time arrival estimation using machine learning.pdf - Submitted Version Download (3MB)

Abstract

This paper presents a study on the implementation of a smart shuttle tracking system that incorporates IoT devices, specifically the Raspberry Pi4b, and leverages machine learning techniques for accurate estimated time of arrival (ETA) estimation. The objective of this research is to address the problem of long waiting times and uncertainty faced by shuttle passengers. The literature review provides an overview of smart shuttle tracking systems, highlighting the importance of real-time location tracking and ETA estimation for improving the waiting experience. It explores existing technologies such as GPS-based tracking systems and the role of IoT devices in collecting and transmitting real-time shuttle data. Additionally, it investigates machine learning and artificial neural networks, showcasing their potential in analyzing complex data and enhancing ETA estimation accuracy. The study proposes a system architecture that integrates IoT devices, such as Raspberry Pi4b, with a cloud-based platform firebse for real-time data collection and processing. A machine learning model, based on artificial neural networks, is trained using historical shuttle data, traffic patterns, and other relevant factors to estimate the shuttle's ETA accurately. The model is deployed on the cloud platform to provide real-time ETA updates to shuttle passengers via a mobile application built using user-centric design principles. The project's scope includes the development of the hardware infrastructure, the implementation of the machine learning model, and the design of the mobile application. The safety and security aspects of the system are also considered to ensure passenger safety. The research contributes to the field by offering an integrated solution that combines IoT, machine learning, and user-centric design principles to enhance shuttle tracking and improve the waiting experience for passengers. The evaluation of the system's performance, user feedback, and lessons learned from the implementation will be discussed.

Actions (login required)

View Item