Toward Human-Centered Design of Automated Vehicles: A Naturalistic Brake Policy

Frontiers in Future Transportation

Yalda Rahmati¹  Mahmood Tabesh²  Alireza Talebpour³  Arezoo Samimi Abianeh³ 

[1] Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States;M University, College Station, TX, United States;;Zachry Department of Civil Engineering, Texas A&

关键词: connected automated vehicles;    human drivers;    intersection;    braking profile;    neural networks;    Markov decision process;   

DOI  :  10.3389/ffutr.2021.683223

来源: DOAJ

【 摘 要 】

While safety is the ultimate goal in designing Connected and Automated Vehicles (CAVs), current automotive safety standards fail to explicitly define rules and regulations that ensure the safety of CAVs or those interacting with such vehicles. This study investigates CAV safety in mixed traffic environments with both human-driven and automated vehicles, focusing particularly on rear-end collisions at intersections. The central hypothesis is that the primary reason behind these crashes is the potential mismatch between CAVs’ braking decisions and human drivers’ expectations. To test this hypothesis, various Artificial Intelligence (AI) techniques, along with specialized statistical methods are adopted to learn and model the braking behavior of human drivers at intersections and compare the results to that of CAVs. Findings suggest systematical differences in CAVs’ and humans’ braking trajectories, revealing a mismatch between their braking patterns. Accordingly, a Markovian decision modeling framework is adopted to design a novel CAV braking profile that ensures 1) compatibility with human expectation, and 2) safe and comfortable maneuvers by CAVs in mixed driving environments. The findings of this study are expected to facilitate the development of higher levels of vehicle automation by providing guidelines to prevent rear-end collisions caused by existing differences in CAVs’ and humans’ braking strategies.

【 授权许可】

Unknown