LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots

Topi Miekkala; Pasi Pyykönen; Matti Kutila; Arto Kyytinen

doi:10.1109/ICCP53602.2021.9733448

LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots

TTS Kehitys Oy

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

2 Citations (Scopus)

252 Downloads (Pure)

Abstract

This article is related to using modern LiDARs and neural networks based algorithms for vulnerable road users (VRU) detection. The problem is obvious especially when considering blind spots of heavy goods vehicles. LiDARs have developed a lot recently and the results indicate that adults can be detected up-to 75 m distance from the sensor even thought that pattern recognition requires sufficient point cloud-resolution. Two different LiDAR brands have been compared to understand costbenefits between LiDAR technologies. The results have been conducted using an automated passenger car while considering feasibility to big trucks. Due to automotive requirements, the processing rate is also considered since usually, the main bottleneck is computation power, which is limited in automotive products. The used neural network algorithm is Yolo based and has been designed for VRU detection.

Original language	English
Title of host publication	2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP)
Editors	Sergiu Nedevschi, Rodica Potolea, Radu Razvan Slavescu
Publisher	IEEE Institute of Electrical and Electronic Engineers
Pages	299-303
ISBN (Electronic)	978-1-6654-0976-6
ISBN (Print)	978-1-6654-0977-3
DOIs	https://doi.org/10.1109/ICCP53602.2021.9733448
Publication status	Published - 16 Mar 2022
MoE publication type	A4 Article in a conference publication
Event	IEEE 17th International Conference on Intelligent Computer Communication and Processing - Technical University of Cluj-Napoca, Cluj-Napoca, Romania Duration: 28 Oct 2021 → 30 Oct 2021 https://iccp.ro/iccp2021/

Conference

Conference	IEEE 17th International Conference on Intelligent Computer Communication and Processing
Country/Territory	Romania
City	Cluj-Napoca
Period	28/10/21 → 30/10/21
Internet address	https://iccp.ro/iccp2021/

Funding

We would like to express our appreciation to the European Commission co-funding the NextPerception (ID: 876487) . We are also grateful to the project consortium for collaboration and fruitful ideas and discussions. NextPerception project has received funding from the ECSEL Joint Underaking (JU) under grant agreement No 876487. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and Finland, Spain, Italy, Germany, Czech Republic, Belgium and The Neherlands.

Keywords

3D Object Detection
neural network
LiDAR
heavy goods vehicle
VRU
blind spot safety

Access to Document

10.1109/ICCP53602.2021.9733448

paper_8_ICCP2021Accepted author manuscript, 1.21 MB

Cite this

Miekkala, T., Pyykönen, P., Kutila, M., & Kyytinen, A. (2022). LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots. In S. Nedevschi, R. Potolea, & R. R. Slavescu (Eds.), 2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP) (pp. 299-303). IEEE Institute of Electrical and Electronic Engineers. https://doi.org/10.1109/ICCP53602.2021.9733448

Miekkala, Topi ; Pyykönen, Pasi ; Kutila, Matti et al. / LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots. 2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP). editor / Sergiu Nedevschi ; Rodica Potolea ; Radu Razvan Slavescu. IEEE Institute of Electrical and Electronic Engineers, 2022. pp. 299-303

@inproceedings{0f0f4fd3c5b34ecea52dc7f5ed0ebeef,

title = "LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots",

abstract = "This article is related to using modern LiDARs and neural networks based algorithms for vulnerable road users (VRU) detection. The problem is obvious especially when considering blind spots of heavy goods vehicles. LiDARs have developed a lot recently and the results indicate that adults can be detected up-to 75 m distance from the sensor even thought that pattern recognition requires sufficient point cloud-resolution. Two different LiDAR brands have been compared to understand costbenefits between LiDAR technologies. The results have been conducted using an automated passenger car while considering feasibility to big trucks. Due to automotive requirements, the processing rate is also considered since usually, the main bottleneck is computation power, which is limited in automotive products. The used neural network algorithm is Yolo based and has been designed for VRU detection. ",

keywords = "3D Object Detection, neural network, LiDAR, heavy goods vehicle, VRU, blind spot safety",

author = "Topi Miekkala and Pasi Pyyk{\"o}nen and Matti Kutila and Arto Kyytinen",

note = "Funding Information: NextPerception project has received funding from the ECSEL Joint Underaking (JU) under grant agreement No 876487. The JU receives support from the European Union{\textquoteright}s Horizon 2020 research and innovation programme and Finland, Spain, Italy, Germany, Czech Republic, Belgium and The Neherlands. Publisher Copyright: {\textcopyright} 2021 IEEE.; IEEE 17th International Conference on Intelligent Computer Communication and Processing ; Conference date: 28-10-2021 Through 30-10-2021",

year = "2022",

month = mar,

day = "16",

doi = "10.1109/ICCP53602.2021.9733448",

language = "English",

isbn = "978-1-6654-0977-3",

pages = "299--303",

editor = "Sergiu Nedevschi and Rodica Potolea and Slavescu, \{Radu Razvan\}",

booktitle = "2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP)",

publisher = "IEEE Institute of Electrical and Electronic Engineers",

address = "United States",

url = "https://iccp.ro/iccp2021/",

}

Miekkala, T , Pyykönen, P , Kutila, M & Kyytinen, A 2022, LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots. in S Nedevschi, R Potolea & RR Slavescu (eds), 2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP). IEEE Institute of Electrical and Electronic Engineers, pp. 299-303, IEEE 17th International Conference on Intelligent Computer Communication and Processing, Cluj-Napoca, Romania, 28/10/21. https://doi.org/10.1109/ICCP53602.2021.9733448

LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots. / Miekkala, Topi ; Pyykönen, Pasi ; Kutila, Matti et al.
2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP). ed. / Sergiu Nedevschi; Rodica Potolea; Radu Razvan Slavescu. IEEE Institute of Electrical and Electronic Engineers, 2022. p. 299-303.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots

AU - Miekkala, Topi

AU - Pyykönen, Pasi

AU - Kutila, Matti

AU - Kyytinen, Arto

N1 - Funding Information: NextPerception project has received funding from the ECSEL Joint Underaking (JU) under grant agreement No 876487. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and Finland, Spain, Italy, Germany, Czech Republic, Belgium and The Neherlands. Publisher Copyright: © 2021 IEEE.

PY - 2022/3/16

Y1 - 2022/3/16

N2 - This article is related to using modern LiDARs and neural networks based algorithms for vulnerable road users (VRU) detection. The problem is obvious especially when considering blind spots of heavy goods vehicles. LiDARs have developed a lot recently and the results indicate that adults can be detected up-to 75 m distance from the sensor even thought that pattern recognition requires sufficient point cloud-resolution. Two different LiDAR brands have been compared to understand costbenefits between LiDAR technologies. The results have been conducted using an automated passenger car while considering feasibility to big trucks. Due to automotive requirements, the processing rate is also considered since usually, the main bottleneck is computation power, which is limited in automotive products. The used neural network algorithm is Yolo based and has been designed for VRU detection.

AB - This article is related to using modern LiDARs and neural networks based algorithms for vulnerable road users (VRU) detection. The problem is obvious especially when considering blind spots of heavy goods vehicles. LiDARs have developed a lot recently and the results indicate that adults can be detected up-to 75 m distance from the sensor even thought that pattern recognition requires sufficient point cloud-resolution. Two different LiDAR brands have been compared to understand costbenefits between LiDAR technologies. The results have been conducted using an automated passenger car while considering feasibility to big trucks. Due to automotive requirements, the processing rate is also considered since usually, the main bottleneck is computation power, which is limited in automotive products. The used neural network algorithm is Yolo based and has been designed for VRU detection.

KW - 3D Object Detection

KW - neural network

KW - LiDAR

KW - heavy goods vehicle

KW - VRU

KW - blind spot safety

UR - https://www.scopus.com/pages/publications/85127374713

U2 - 10.1109/ICCP53602.2021.9733448

DO - 10.1109/ICCP53602.2021.9733448

M3 - Conference article in proceedings

SN - 978-1-6654-0977-3

SP - 299

EP - 303

BT - 2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP)

A2 - Nedevschi, Sergiu

A2 - Potolea, Rodica

A2 - Slavescu, Radu Razvan

PB - IEEE Institute of Electrical and Electronic Engineers

T2 - IEEE 17th International Conference on Intelligent Computer Communication and Processing

Y2 - 28 October 2021 through 30 October 2021

ER -

Miekkala T , Pyykönen P , Kutila M, Kyytinen A. LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots. In Nedevschi S, Potolea R, Slavescu RR, editors, 2021 IEEE 17th International Conference on Intelligent Computer Communication and Processing (ICCP). IEEE Institute of Electrical and Electronic Engineers. 2022. p. 299-303 doi: 10.1109/ICCP53602.2021.9733448

LiDAR system benchmarking for VRU detection in heavy goods vehicle blind spots

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this