Using Inverse Reinforcement Learning to Determine the Factors Contributing to the Visual Demand of Driving

Visual demand of driving, a critical component of mental workload, is intrinsically linked to driving safety and plays a pivotal role in estimating takeover time during automated driving transitions. While the visual occlusion method is a long-standing paradigm for quantifying this demand, prior studies often isolate a single factor or rely on linear regression models that fail to capture the complex, nonlinear nature of driving behavior. To address these gaps, this study employs a naturalistic, unconstrained visual occlusion experiment. We utilize Maximum Entropy Inverse Reinforcement Learning to infer reward policies from multi-dimensional driving trajectories, using occlusion duration as a behavioral outcome. This approach allows us to quantify the relative importance of multiple interacting factors in shaping visual demand of driving.

Authors

Institutions

Publication Details

Journal
Proceedings of the Human Factors and Ergonomics Society Annual Meeting
Published
2026-11-18
DOI
https://doi.org/10.1177/10711813261498775
Primary Topic
Human-Automation Interaction and Safety
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Using Inverse Reinforcement Learning to Determine the Factors Contributing to the Visual Demand of Driving

Paul Green, Liangxiu Chen, Chengjun Ma
Proceedings of the Human Factors and Ergonomics Society Annual Meeting
Human-Automation Interaction and Safety
article

Using Inverse Reinforcement Learning to Determine the Factors Contributing to the Visual Demand of Driving

Paul Green, Liangxiu Chen, Chengjun Ma
article en

Abstract

Visual demand of driving, a critical component of mental workload, is intrinsically linked to driving safety and plays a pivotal role in estimating takeover time during automated driving transitions. While the visual occlusion method is a long-standing paradigm for quantifying this demand, prior studies often isolate a single factor or rely on linear regression models that fail to capture the complex, nonlinear nature of driving behavior. To address these gaps, this study employs a naturalistic, unconstrained visual occlusion experiment. We utilize Maximum Entropy Inverse Reinforcement Learning to infer reward policies from multi-dimensional driving trajectories, using occlusion duration as a behavioral outcome. This approach allows us to quantify the relative importance of multiple interacting factors in shaping visual demand of driving.

Proceedings of the Human Factors and Ergonomics Society Annual Meeting
University of Michigan (US)
Human-Automation Interaction and Safety
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon