Deep Learning Model for Intelligent Location Prediction in VANETs: LSTM-CNN A Hybrid Approach

In Cooperative Intelligent Transport Systems (C-ITS), vehicles and/or RSUs exchange various messages to improve traffic management, promote collision prevention, and ensure safety on our roads. These messages are broadcast within the vehicle’s vicinity. Each message must contain the accurate position of the vehicle, along with other relevant information such as the velocity and the vehicle identifier. Accurate information on the vehicle’s position significantly affects the majority of sensitive applications. Finding an approach to predict timely and accurately the exact position of a vehicle has become a great challenge in CITS. This study addresses the problem of Vehicle Location Prediction (VLP) using a deep learning approach. A hybrid solution based on Convolutional Neural Networks (CNNs) and Long-Short-Term-Memory (LSTM) is proposed. While CNN excels at extracting spatial patterns from positions sequences, LSTM captures long-term temporal dependencies, providing a comprehensive spatiotemporal representation. A comparative analysis has been conducted to evaluate the performance of our approach; the results showed that the CNN-LSTM model outperforms other models in terms of localisation error.