Related Experiment Video
Updated: Nov 21, 2025

Evaluation of an Exclusive Spur Dike U-Turn Design with Radar-Collected Data and Simulation
Published on: February 1, 2020
Walking Secure: Safe Routing Planning Algorithm and Pedestrian's Crossing Intention Detector Based on Fuzzy Logic
José Manuel Lozano Domínguez1, Tomás de J Mateo Sanguino1
1Department of Electronic Engineering, Computer Systems and Automatics, University of Huelva, Av. de las Artes s/n, 21007 Huelva, Spain.
This study presents a mobile app using smartphone sensors and fuzzy logic to detect pedestrian crossing intentions, enhancing smart city road safety. The app also generates safer routes, prioritizing pedestrian areas over traditional navigation.
Area of Science:
- Artificial Intelligence
- Smart City Technologies
- Road Safety Engineering
Background:
- Improving road safety is critical for developing secure smart cities.
- Current navigation systems often do not prioritize pedestrian safety or utilize dedicated pedestrian infrastructure.
- Detecting pedestrian intent at crosswalks remains a challenge for automated systems.
Purpose of the Study:
- To develop and evaluate a mobile application that uses smartphone sensors and fuzzy logic to determine pedestrian crossing intentions.
- To create an algorithm for generating safer navigation routes that incorporate pedestrian-exclusive areas.
- To compare the safety and efficiency of algorithm-generated routes against existing navigation services like Google Maps.
Main Methods:
- Integration of smartphone sensors with a fuzzy logic strategy to analyze pedestrian behavior near crosswalks.
- Development of an optimization algorithm utilizing a cloud database of pedestrian areas (crosswalks, walkways) for route generation.
- Experimentation involving 31 volunteers and 3120 data samples to test the fuzzy logic crossing intention detection.
- Comparative analysis of 30 generated routes against Google Maps, evaluating time, distance, and safety metrics.
Main Results:
- The smartphone-based fuzzy logic system achieved 98.63% accuracy in detecting pedestrian crossing intention.
- True positive rate reached 98.27% and specificity was 99.39% based on ROC analysis.
- Algorithm-generated routes demonstrated significantly higher safety, increasing the use of safe pedestrian areas by at least 183% compared to Google Maps.
Conclusions:
- A smartphone can effectively function as a reliable pedestrian crossing intention detection system.
- The proposed algorithm successfully generates safer navigation routes by prioritizing pedestrian infrastructure.
- This technology offers a promising advancement for enhancing pedestrian safety within smart city environments.
Related Concept Videos
Design Example: Alignment of a Road Line Using GIS
Survey Safety
Root-Locus Method
This system can be represented by a block...
Manipulation and Analysis
PD Controller: Design
Designing a continuous-data controller requires selecting and linking components like adders and integrators, which are fundamental in Proportional,...

