Task-Based AR-HUD in Autonomous Driving: Enhancing Driver Agency, Engagement, Attention, and Takeover Performance

Sun, Bingxin; Zhao, Danhua · 2026 · Crossref

DOI: 10.54941/ahfe1007865

archive: archived pipeline: cataloged

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the critical safety challenge in SAE Level 3 autonomous driving where drivers fall "out-of-the-loop" during prolonged passive monitoring, leading to degraded situational awareness and poor takeover performance. The authors propose a "task-based AR-HUD" (Augmented Reality Head-Up Display) that transforms vehicle motion planning into actionable task opportunities (e.g., lane changes, overtaking) requiring driver approval. This design aims to maintain the driver’s cognitive flow and sense of agency without imposing excessive cognitive load, thereby balancing psychological needs with functional safety. The study utilized a high-fidelity driving simulator to compare three progressive AR-HUD visual feedback modalities: linear (baseline line guidance), dynamic (incorporating motion effects), and task-based (requiring active driver decision-making). Twenty healthy adults aged 18–45 with prior autonomous driving experience participated in the experiment, which involved viewing first-person driving videos of urban and motorway segments under cloudy weather conditions. Participants completed a User Evaluation Scale after each trial, measuring perceived control, willingness to use, driver agency, safety, comfort, and emotional experience. Semi-structured interviews were also conducted to gather qualitative insights. Results indicated that the task-based AR-HUD significantly enhanced driver engagement, as evidenced by a higher engagement score compared to linear and dynamic modalities (F(2, 57) = 9.84, p = 0.0002). It also significantly increased the willingness to use the system (F(2, 57) = 11.27, p < 0.0001) and improved playfulness and emotional experience (p = 0.025 and p = 0.017, respectively). Notably, while the task-based modality improved engagement and emotional states, it resulted in lower visual comfort scores compared to the dynamic modality (F(2, 57) = 14.62, p < 0.00001), suggesting a trade-off between cognitive engagement and visual strain. The study concluded that task-based interactions effectively redirect attention to the road and sustain the driver’s "in-the-loop" state by providing a continuous sense of purpose and control feedback. The significance of this research lies in its shift from treating AR-HUDs as mere information display tools for takeover moments to interactive interfaces that sustain continuous driver participation. By validating that task-based interactions enhance driver agency and mitigate the boredom associated with passive monitoring, the study provides empirical evidence for designing human-machine collaborative systems that prioritize psychological engagement alongside safety. This approach offers a theoretical foundation for future high-level autonomous driving interfaces that prevent driver disengagement without disrupting the relaxation benefits of automation.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

StageOutcomeToolModelPromptAttemptsCompleted
discover success Crossref 1 2026-08-09
archive success canonical_url 1 2026-08-09
extract success cached 5 2026-08-23
clean success clean 2 2026-08-10
chunk success chunk 2 2026-08-10
embed success embed Qwen/Qwen3-Embedding-8B 2 2026-08-10
promote success 1 2026-08-09
summarize success llm qwen3.8-27b-gittensor summ-v5 3 2026-08-23
tag success vector_similarity 17 2026-08-11
verify success 2 2026-08-09

Summary generated by qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).