Driving Behavior Classification Using a ConvLSTM

This work explores the classification of driving behaviors using a hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks (ConvLSTM). Sensor data are collected from a smartphone application and undergo a preprocessing pipeline, inclu...

Full description

Saved in:
Bibliographic Details
Main Authors: Alberto Pingo, João Castro, Paulo Loureiro, Sílvio Mendes, Anabela Bernardino, Rolando Miragaia, Iryna Husyeva
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Future Transportation
Subjects:
Online Access:https://www.mdpi.com/2673-7590/5/2/52
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849432096309772288
author Alberto Pingo
João Castro
Paulo Loureiro
Sílvio Mendes
Anabela Bernardino
Rolando Miragaia
Iryna Husyeva
author_facet Alberto Pingo
João Castro
Paulo Loureiro
Sílvio Mendes
Anabela Bernardino
Rolando Miragaia
Iryna Husyeva
author_sort Alberto Pingo
collection DOAJ
description This work explores the classification of driving behaviors using a hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks (ConvLSTM). Sensor data are collected from a smartphone application and undergo a preprocessing pipeline, including data normalization, labeling, and feature extraction, to enhance the model’s performance. By capturing temporal and spatial dependencies within driving patterns, the proposed ConvLSTM model effectively differentiates between normal and aggressive driving behaviors. The model is trained and evaluated against traditional stacked LSTM and Bidirectional LSTM (BiLSTM) architectures, demonstrating superior accuracy and robustness. Experimental results confirm that the preprocessing techniques improve classification performance, ensuring high reliability in driving behavior recognition. The novelty of this work lies in a simple data preprocessing methodology combined with the specific application scenario. By enhancing data quality before feeding it into the AI model, we improve classification accuracy and robustness. The proposed framework not only optimizes model performance but also demonstrates practical feasibility, making it a strong candidate for real-world deployment.
format Article
id doaj-art-e936221535974e67834bbc1c0200e202
institution Kabale University
issn 2673-7590
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Future Transportation
spelling doaj-art-e936221535974e67834bbc1c0200e2022025-08-20T03:27:26ZengMDPI AGFuture Transportation2673-75902025-05-01525210.3390/futuretransp5020052Driving Behavior Classification Using a ConvLSTMAlberto Pingo0João Castro1Paulo Loureiro2Sílvio Mendes3Anabela Bernardino4Rolando Miragaia5Iryna Husyeva6School of Technology and Management, Polytechnic University of Leiria, 2411-901 Leiria, PortugalSchool of Technology and Management, Polytechnic University of Leiria, 2411-901 Leiria, PortugalComputer Science and Communication Research Centre, School of Technology and Management, Polytechnic University of Leiria, 2411-901 Leiria, PortugalComputer Science and Communication Research Centre, School of Technology and Management, Polytechnic University of Leiria, 2411-901 Leiria, PortugalComputer Science and Communication Research Centre, School of Technology and Management, Polytechnic University of Leiria, 2411-901 Leiria, PortugalComputer Science and Communication Research Centre, School of Technology and Management, Polytechnic University of Leiria, 2411-901 Leiria, PortugalComputer Science and Communication Research Centre, National Technical University of Ukraine Igor Sikorsky Kyiv Polytechnic Institute, 03056 Kyiv, UkraineThis work explores the classification of driving behaviors using a hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks (ConvLSTM). Sensor data are collected from a smartphone application and undergo a preprocessing pipeline, including data normalization, labeling, and feature extraction, to enhance the model’s performance. By capturing temporal and spatial dependencies within driving patterns, the proposed ConvLSTM model effectively differentiates between normal and aggressive driving behaviors. The model is trained and evaluated against traditional stacked LSTM and Bidirectional LSTM (BiLSTM) architectures, demonstrating superior accuracy and robustness. Experimental results confirm that the preprocessing techniques improve classification performance, ensuring high reliability in driving behavior recognition. The novelty of this work lies in a simple data preprocessing methodology combined with the specific application scenario. By enhancing data quality before feeding it into the AI model, we improve classification accuracy and robustness. The proposed framework not only optimizes model performance but also demonstrates practical feasibility, making it a strong candidate for real-world deployment.https://www.mdpi.com/2673-7590/5/2/52artificial intelligenceneural networksLSTMRNNdriving classification
spellingShingle Alberto Pingo
João Castro
Paulo Loureiro
Sílvio Mendes
Anabela Bernardino
Rolando Miragaia
Iryna Husyeva
Driving Behavior Classification Using a ConvLSTM
Future Transportation
artificial intelligence
neural networks
LSTM
RNN
driving classification
title Driving Behavior Classification Using a ConvLSTM
title_full Driving Behavior Classification Using a ConvLSTM
title_fullStr Driving Behavior Classification Using a ConvLSTM
title_full_unstemmed Driving Behavior Classification Using a ConvLSTM
title_short Driving Behavior Classification Using a ConvLSTM
title_sort driving behavior classification using a convlstm
topic artificial intelligence
neural networks
LSTM
RNN
driving classification
url https://www.mdpi.com/2673-7590/5/2/52
work_keys_str_mv AT albertopingo drivingbehaviorclassificationusingaconvlstm
AT joaocastro drivingbehaviorclassificationusingaconvlstm
AT pauloloureiro drivingbehaviorclassificationusingaconvlstm
AT silviomendes drivingbehaviorclassificationusingaconvlstm
AT anabelabernardino drivingbehaviorclassificationusingaconvlstm
AT rolandomiragaia drivingbehaviorclassificationusingaconvlstm
AT irynahusyeva drivingbehaviorclassificationusingaconvlstm