Learning to see through the haze: Multi-sensor learning-fusion System for Vulnerable Traffic Participant Detection in Fog

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F21%3A00347067" target="_blank" >RIV/68407700:21230/21:00347067 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1016/j.robot.2020.103687" target="_blank" >https://doi.org/10.1016/j.robot.2020.103687</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.robot.2020.103687" target="_blank" >10.1016/j.robot.2020.103687</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Learning to see through the haze: Multi-sensor learning-fusion System for Vulnerable Traffic Participant Detection in Fog

Popis výsledku v původním jazyce

We present an experimental investigation of a multi-sensor fusion-learning system for detecting pedestrians in foggy weather conditions. The method combines two pipelines for people detection running on two different sensors commonly found on moving vehicles: lidar and radar. The two pipelines are not only combined by sensor fusion, but information from one pipeline is used to train the other. We build upon our previous work, where we showed that a lidar pipeline can be used to train a Support Vector Machine (SVM)-based pipeline to interpret radar data, which is useful when conditions then become unfavourable to the original lidar pipeline. In this paper, we test the method on a wider range of conditions, such as from a moving vehicle, and with multiple people present. Additionally, we also compare how the traditional SVM performs interpreting the radar data versus a modern deep neural network on these experiments. Our experiments indicate that either of the approaches results in progressive improvement in the performance during normal operation. Further, our experiments indicate that in the event of the loss of information from a sensor, pedestrian detection and position estimation is still effective.

Název v anglickém jazyce

Learning to see through the haze: Multi-sensor learning-fusion System for Vulnerable Traffic Participant Detection in Fog

Popis výsledku anglicky

We present an experimental investigation of a multi-sensor fusion-learning system for detecting pedestrians in foggy weather conditions. The method combines two pipelines for people detection running on two different sensors commonly found on moving vehicles: lidar and radar. The two pipelines are not only combined by sensor fusion, but information from one pipeline is used to train the other. We build upon our previous work, where we showed that a lidar pipeline can be used to train a Support Vector Machine (SVM)-based pipeline to interpret radar data, which is useful when conditions then become unfavourable to the original lidar pipeline. In this paper, we test the method on a wider range of conditions, such as from a moving vehicle, and with multiple people present. Additionally, we also compare how the traditional SVM performs interpreting the radar data versus a modern deep neural network on these experiments. Our experiments indicate that either of the approaches results in progressive improvement in the performance during normal operation. Further, our experiments indicate that in the event of the loss of information from a sensor, pedestrian detection and position estimation is still effective.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Robotics and Autonomous Systems

ISSN

0921-8890

e-ISSN

1872-793X

Svazek periodika

136

Číslo periodika v rámci svazku

February

Stát vydavatele periodika

NL - Nizozemsko

Počet stran výsledku

13

Strana od-do

—

Kód UT WoS článku

000609453100002

EID výsledku v databázi Scopus

2-s2.0-85096678939