Related Experiment Video
Updated: Oct 30, 2025
10:52
Simulation of Human-induced Vibrations Based on the Characterized In-field Pedestrian Behavior
Published on: April 13, 2016
8.9K
Feasibility of Using a MEMS Microphone Array for Pedestrian Detection in an Autonomous Emergency Braking System
Alberto Izquierdo1, Lara Del Val2, Juan J Villacorta1
1Department of Signal Theory and Communications and Telematics Engineering, University of Valladolid, Paseo de Belén, 15, 47011 Valladolid, Spain.
Sensors (Basel, Switzerland)
|July 2, 2021
Summary
A novel active sound sensor system offers reliable pedestrian detection for vehicles, overcoming limitations of light-based and radar systems. This cost-effective solution ensures high detection probability, crucial for preventing collisions.
Area of Science:
- Engineering
- Sensor Technology
- Automotive Safety
Background:
- Current pedestrian detection systems (camera, LIDAR, RADAR) have limitations in poor visibility or are costly with low detection rates.
- There is a need for non-light-based, cost-effective pedestrian detection systems with high reliability.
Purpose of the Study:
- To develop and validate a new active sound sensor system for enhanced pedestrian detection.
- To address the limitations of existing automotive pedestrian detection technologies.
Main Methods:
- Development of an active sound system utilizing a 2D array of MEMS microphones (14 kHz–21 kHz band).
- Real-time processing architecture based on FPGA and multicore processor.
- Algorithms include beamformer, range/lane filters, and Constant False Alarm Rate (CFAR) detector.
Main Results:
- The developed system accurately detects and estimates pedestrian positions.
- Experimental tests calculated Detection Probability and False Alarm Probability across various conditions.
- The system ensures vehicle safety up to 50 km/h, enabling collision avoidance.
Conclusions:
- The novel active sound sensor system meets the requirements for cost-effectiveness and high detection probability.
- This technology offers a viable alternative for robust pedestrian detection in adverse conditions.
- The system's performance validates its capability to significantly enhance automotive safety.
