Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous Driving

In the context of autonomous driving, predicting pedestrian behavior is a critical component for enhancing road safety. Currently, the focus of such predictions extends beyond accuracy and reliability, placing increasing emphasis on the explainability and interpretability of the models. This researc...

Full description

Saved in:
Bibliographic Details
Main Authors: Angie Nataly Melo Castillo, Carlota Salinas Maldonado, Miguel Ángel Sotelo
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/11/6283
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849330627191504896
author Angie Nataly Melo Castillo
Carlota Salinas Maldonado
Miguel Ángel Sotelo
author_facet Angie Nataly Melo Castillo
Carlota Salinas Maldonado
Miguel Ángel Sotelo
author_sort Angie Nataly Melo Castillo
collection DOAJ
description In the context of autonomous driving, predicting pedestrian behavior is a critical component for enhancing road safety. Currently, the focus of such predictions extends beyond accuracy and reliability, placing increasing emphasis on the explainability and interpretability of the models. This research presents a novel neuro-symbolic approach that integrates deep learning with fuzzy logic to develop a pedestrian behavior predictor. The proposed model leverages a set of explainable features and utilizes a fuzzy inference system to determine whether a pedestrian is likely to cross the street. The pipeline was trained and evaluated using both the Pedestrian Intention Estimation (PIE) and Joint Attention for Autonomous Driving (JAAD) datasets. The results provide experimental insights into achieving greater explainability in pedestrian behavior prediction. Additionally, the proposed method was applied to assess the data selection process through a series of experiments, leading to a set of guidelines and recommendations for data selection, feature engineering, and explainability.
format Article
id doaj-art-fa9e17394db84b4281b9e57df48c926f
institution Kabale University
issn 2076-3417
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-fa9e17394db84b4281b9e57df48c926f2025-08-20T03:46:52ZengMDPI AGApplied Sciences2076-34172025-06-011511628310.3390/app15116283Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous DrivingAngie Nataly Melo Castillo0Carlota Salinas Maldonado1Miguel Ángel Sotelo2Computer Engineering Department, University of Alcalá, Ctra. Madrid-Barcelona km. 33, 28805 Alcalá de Henares, SpainComputer Engineering Department, University of Alcalá, Ctra. Madrid-Barcelona km. 33, 28805 Alcalá de Henares, SpainComputer Engineering Department, University of Alcalá, Ctra. Madrid-Barcelona km. 33, 28805 Alcalá de Henares, SpainIn the context of autonomous driving, predicting pedestrian behavior is a critical component for enhancing road safety. Currently, the focus of such predictions extends beyond accuracy and reliability, placing increasing emphasis on the explainability and interpretability of the models. This research presents a novel neuro-symbolic approach that integrates deep learning with fuzzy logic to develop a pedestrian behavior predictor. The proposed model leverages a set of explainable features and utilizes a fuzzy inference system to determine whether a pedestrian is likely to cross the street. The pipeline was trained and evaluated using both the Pedestrian Intention Estimation (PIE) and Joint Attention for Autonomous Driving (JAAD) datasets. The results provide experimental insights into achieving greater explainability in pedestrian behavior prediction. Additionally, the proposed method was applied to assess the data selection process through a series of experiments, leading to a set of guidelines and recommendations for data selection, feature engineering, and explainability.https://www.mdpi.com/2076-3417/15/11/6283autonomous drivingexplainabilityinterpretabilitypedestrian behavior predictionneuro-symbolicdataset selection
spellingShingle Angie Nataly Melo Castillo
Carlota Salinas Maldonado
Miguel Ángel Sotelo
Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous Driving
Applied Sciences
autonomous driving
explainability
interpretability
pedestrian behavior prediction
neuro-symbolic
dataset selection
title Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous Driving
title_full Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous Driving
title_fullStr Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous Driving
title_full_unstemmed Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous Driving
title_short Towards Explainable Pedestrian Behavior Prediction: A Neuro-Symbolic Framework for Autonomous Driving
title_sort towards explainable pedestrian behavior prediction a neuro symbolic framework for autonomous driving
topic autonomous driving
explainability
interpretability
pedestrian behavior prediction
neuro-symbolic
dataset selection
url https://www.mdpi.com/2076-3417/15/11/6283
work_keys_str_mv AT angienatalymelocastillo towardsexplainablepedestrianbehaviorpredictionaneurosymbolicframeworkforautonomousdriving
AT carlotasalinasmaldonado towardsexplainablepedestrianbehaviorpredictionaneurosymbolicframeworkforautonomousdriving
AT miguelangelsotelo towardsexplainablepedestrianbehaviorpredictionaneurosymbolicframeworkforautonomousdriving