Intelligent Multi-Destination Navigation Using Hybrid Machine Learning–Based Traffic Prediction and Dynamic Route Optimization

Abstract

Traveling both domestic and abroad for different purposes such as business, studies, and fun has become more popular because of the increasing interest for tourism. This tourism is inspired by different digital media and advertising. On the other hand, people are becoming more aware of the environment and how much does the car affect it and how much energy does it need. The single-source to single-destination routes are the main focus of the navigation systems such as Google Maps. In this paper, we propose an intelligent navigation system which uses machine learning algorithms and GPS to efficiently guide the people to different kinds of destinations. The system aims to optimize the trip, reduce the gasoline, and increase the efficiency of the travel by integrating the traffic, real-time data cost effectiveness, shortest path, fuel economy, and ideal route suggestions for multiple destinations. Grid-based route planning, collection of data, route optimization and traffic prediction using LSTM and GRU models are included in the system. This framework explains how to develop an environmentally friendly, user, and intelligent navigation system that can provide the best path in real time for various destinations.