Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO

Luca Crosato; Chongfeng Wei; Edmond S. L. Ho; Hubert P. H. Shum

doi:10.1109/ICHMS53169.2021.9582640

Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO

Luca Crosato^*, Chongfeng Wei^*, Edmond S. L. Ho, Hubert P. H. Shum

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

15 Citations (Scopus)

71 Downloads (Pure)

Abstract

As Autonomous Vehicles (AV) are becoming a reality, the design of efficient motion control algorithms will have to deal with the unpredictable and interactive nature of other road users. Current AV motion planning algorithms suffer from the freezing robot problem, as they often tend to overestimate collision risks. To tackle this problem and design AV that behave human-like, we integrate a concept from Psychology called Social Value Orientation into the Reinforcement Learning (RL) framework. The addition of a social term in the reward function design allows us to tune the AV behaviour towards the pedestrian from a more reckless to an extremely prudent one. We train the vehicle agent with a state of the art RL algorithm and show that Social Value Orientation is an effective tool to obtain pro-social AV behaviour.

Original language	English
Title of host publication	2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS 2021)
Editors	Andreas Nürnberger, Giancarlo Fortino, Antonio Guerrieri, David Kaber, David Mendonca, Malte Schilling, Zhiwen Yu
Place of Publication	Piscataway, US
Publisher	IEEE
Pages	85-90
Number of pages	6
ISBN (Electronic)	9781665401708
ISBN (Print)	9781665401715
DOIs	https://doi.org/10.1109/ICHMS53169.2021.9582640
Publication status	Published - 8 Sept 2021
Event	IEEE ICHMS 2021: 2nd IEEE International Conference on Human-Machine Systems: Human Centered Systems for our Digital World - Magdeburg, Germany Duration: 8 Sept 2021 → 10 Sept 2021 https://www.ichms2021.de/

Conference

Conference	IEEE ICHMS 2021: 2nd IEEE International Conference on Human-Machine Systems
Country/Territory	Germany
City	Magdeburg
Period	8/09/21 → 10/09/21
Internet address	https://www.ichms2021.de/

Keywords

Autonomous Vehicle-Pedestrian Interaction
Autonomous Vehicle
Reinforcement Learning
Human-Robot Interaction (HRI)
Social Value Orientation
Social Behaviour

Access to Document

10.1109/ICHMS53169.2021.9582640

IEEE_ICHMS_2021_paper_14Accepted author manuscript, 990 KB

Cite this

Crosato, L., Wei, C., Ho, E. S. L., & Shum, H. P. H. (2021). Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO. In A. Nürnberger, G. Fortino, A. Guerrieri, D. Kaber, D. Mendonca, M. Schilling, & Z. Yu (Eds.), 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS 2021) (pp. 85-90). IEEE. https://doi.org/10.1109/ICHMS53169.2021.9582640

Crosato, Luca ; Wei, Chongfeng ; Ho, Edmond S. L. et al. / Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO. 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS 2021). editor / Andreas Nürnberger ; Giancarlo Fortino ; Antonio Guerrieri ; David Kaber ; David Mendonca ; Malte Schilling ; Zhiwen Yu. Piscataway, US : IEEE, 2021. pp. 85-90

@inproceedings{afe74fea00e543368d6370c2abffe381,

title = "Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO",

abstract = "As Autonomous Vehicles (AV) are becoming a reality, the design of efficient motion control algorithms will have to deal with the unpredictable and interactive nature of other road users. Current AV motion planning algorithms suffer from the freezing robot problem, as they often tend to overestimate collision risks. To tackle this problem and design AV that behave human-like, we integrate a concept from Psychology called Social Value Orientation into the Reinforcement Learning (RL) framework. The addition of a social term in the reward function design allows us to tune the AV behaviour towards the pedestrian from a more reckless to an extremely prudent one. We train the vehicle agent with a state of the art RL algorithm and show that Social Value Orientation is an effective tool to obtain pro-social AV behaviour.",

keywords = "Autonomous Vehicle-Pedestrian Interaction, Autonomous Vehicle, Reinforcement Learning, Human-Robot Interaction (HRI), Social Value Orientation, Social Behaviour",

author = "Luca Crosato and Chongfeng Wei and Ho, {Edmond S. L.} and Shum, {Hubert P. H.}",

year = "2021",

month = sep,

day = "8",

doi = "10.1109/ICHMS53169.2021.9582640",

language = "English",

isbn = "9781665401715",

pages = "85--90",

editor = "Andreas N{\"u}rnberger and Giancarlo Fortino and Antonio Guerrieri and David Kaber and David Mendonca and Malte Schilling and Zhiwen Yu",

booktitle = "2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS 2021)",

publisher = "IEEE",

address = "United States",

note = "IEEE ICHMS 2021: 2nd IEEE International Conference on Human-Machine Systems : Human Centered Systems for our Digital World ; Conference date: 08-09-2021 Through 10-09-2021",

url = "https://www.ichms2021.de/",

}

Crosato, L, Wei, C, Ho, ESL & Shum, HPH 2021, Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO. in A Nürnberger, G Fortino, A Guerrieri, D Kaber, D Mendonca, M Schilling & Z Yu (eds), 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS 2021). IEEE, Piscataway, US, pp. 85-90, IEEE ICHMS 2021: 2nd IEEE International Conference on Human-Machine Systems, Magdeburg, Saxony-Anhalt, Germany, 8/09/21. https://doi.org/10.1109/ICHMS53169.2021.9582640

Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO. / Crosato, Luca; Wei, Chongfeng; Ho, Edmond S. L. et al.
2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS 2021). ed. / Andreas Nürnberger; Giancarlo Fortino; Antonio Guerrieri; David Kaber; David Mendonca; Malte Schilling; Zhiwen Yu. Piscataway, US: IEEE, 2021. p. 85-90.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO

AU - Crosato, Luca

AU - Wei, Chongfeng

AU - Ho, Edmond S. L.

AU - Shum, Hubert P. H.

PY - 2021/9/8

Y1 - 2021/9/8

N2 - As Autonomous Vehicles (AV) are becoming a reality, the design of efficient motion control algorithms will have to deal with the unpredictable and interactive nature of other road users. Current AV motion planning algorithms suffer from the freezing robot problem, as they often tend to overestimate collision risks. To tackle this problem and design AV that behave human-like, we integrate a concept from Psychology called Social Value Orientation into the Reinforcement Learning (RL) framework. The addition of a social term in the reward function design allows us to tune the AV behaviour towards the pedestrian from a more reckless to an extremely prudent one. We train the vehicle agent with a state of the art RL algorithm and show that Social Value Orientation is an effective tool to obtain pro-social AV behaviour.

AB - As Autonomous Vehicles (AV) are becoming a reality, the design of efficient motion control algorithms will have to deal with the unpredictable and interactive nature of other road users. Current AV motion planning algorithms suffer from the freezing robot problem, as they often tend to overestimate collision risks. To tackle this problem and design AV that behave human-like, we integrate a concept from Psychology called Social Value Orientation into the Reinforcement Learning (RL) framework. The addition of a social term in the reward function design allows us to tune the AV behaviour towards the pedestrian from a more reckless to an extremely prudent one. We train the vehicle agent with a state of the art RL algorithm and show that Social Value Orientation is an effective tool to obtain pro-social AV behaviour.

KW - Autonomous Vehicle-Pedestrian Interaction

KW - Autonomous Vehicle

KW - Reinforcement Learning

KW - Human-Robot Interaction (HRI)

KW - Social Value Orientation

KW - Social Behaviour

U2 - 10.1109/ICHMS53169.2021.9582640

DO - 10.1109/ICHMS53169.2021.9582640

M3 - Conference contribution

SN - 9781665401715

SP - 85

EP - 90

BT - 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS 2021)

A2 - Nürnberger, Andreas

A2 - Fortino, Giancarlo

A2 - Guerrieri, Antonio

A2 - Kaber, David

A2 - Mendonca, David

A2 - Schilling, Malte

A2 - Yu, Zhiwen

PB - IEEE

CY - Piscataway, US

T2 - IEEE ICHMS 2021: 2nd IEEE International Conference on Human-Machine Systems

Y2 - 8 September 2021 through 10 September 2021

ER -

Human-centric Autonomous Driving in an AV-Pedestrian Interactive Environment Using SVO

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this