A4 Article in conference proceedings
Modeling human road crossing decisions as reward maximization with visual perception limitations (2023)

Wang, Y., Srinivasan, A. R., Jokinen, J., Oulasvirta, A., & Markkula, G. (2023). Modeling human road crossing decisions as reward maximization with visual perception limitations. In 2023 IEEE Intelligent Vehicles Symposium (IV). IEEE. IEEE Intelligent Vehicles Symposium. https://doi.org/10.1109/IV55152.2023.10186617

JYU authors or editors

Publication details

All authors or editors: Wang, Yueyang; Srinivasan, Aravinda Ramakrishnan; Jokinen, Jussi, P. P.; Oulasvirta, Antti; Markkula, Gustav

Parent publication: 2023 IEEE Intelligent Vehicles Symposium (IV)

Place and date of conference: Anchorage, USA, 4.-7.6.2023

ISBN: 979-8-3503-4692-3

eISBN: 979-8-3503-4691-6

Journal or series: IEEE Intelligent Vehicles Symposium

ISSN: 1931-0587

eISSN: 2642-7214

Publication year: 2023

Publisher: IEEE

Publication country: United States

Publication language: English

DOI: https://doi.org/10.1109/IV55152.2023.10186617

Publication open access: Not open

Publication channel open access: 

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/92697

Web address of parallel published publication (pre-print): https://arxiv.org/abs/2301.11737

Abstract

Understanding the interaction between different road users is critical for road safety and automated vehicles (AVs). Existing mathematical models on this topic have been proposed based mostly on either cognitive or machine learning (ML) approaches. However, current cognitive models are incapable of simulating road user trajectories in general scenarios, and ML models lack a focus on the mechanisms generating the behavior and take a high-level perspective which can cause failures to capture important human-like behaviors. Here, we develop a model of human pedestrian crossing decisions based on computational rationality, an approach using deep reinforcement learning (RL) to learn boundedly optimal behavior policies given human constraints, in our case a model of the limited human visual system. We show that the proposed combined cognitive-RL model captures human-like patterns of gap acceptance and crossing initiation time. Interestingly, our model’s decisions are sensitive to not only the time gap, but also the speed of the approaching vehicle, something which has been described as a “bias” in human gap acceptance behavior. However, our results suggest that this is instead a rational adaption to human perceptual limitations. Moreover, we demonstrate an approach to accounting for individual differences in computational rationality models, by conditioning the RL policy on the parameters of the human constraints. Our results demonstrate the feasibility of generating more human-like road user behavior by combining RL with cognitive models.

Keywords: behaviour; pedestrians; traffic; road traffic; traffic safety; cognitive processes; perception (activity); mathematical models; machine learning; modelling (representation)

Free keywords: human behavior; computational rationality; noisy perception; reinforcement learning; adaptation models; pedestrians; intelligent vehicles; computational modeling; mathematical models; road safety

Contributing organizations

Ministry reporting: Yes

VIRTA submission year: 2023

JUFO rating: 1