Real-time Vehicle Monitoring and Route Prediction using Geolocation through LSTM

Abstract

Traveling by vehicle like bus, car, etc. is a popular form of public transportation because these are relatively affordable, convenient, accessible, and an environmentally friendly option. But sometimes it becomes difficult to manage the passengers in the journey, avoiding traffic jams, providing proper service and thus the passengers get bored. In this study, we propose a real-time geolocation Android application, integrated with Google Firebase, to facilitate predictive analytics for a vehicle company's operational efficiency by collecting real-time geolocation data (longitude and latitude), date, time and speed. The data from the database will be exported as a JSON file and then it has to be converted into a CSV file. This database will be used to train a LSTM (Long short-term memory) model, a type of sequential neural network to enable the company to predict a vehicle's future location on a predefined route based on date, time and speed. Thus it will be helpful to determine the vehicle's estimated arrival time, manage passenger allocation, and forecast traffic conditions to avoid congestion. By harnessing this technology, a vehicle company can enhance its monitoring capabilities, optimize scheduling, and improve overall service delivery.