跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.19474 · v1

Towards Informed Autonomy Design: Understanding the Interplay of Function Characteristics and Autonomy Preferences in Automotive Agents

首次提交 2026-11-14 23:52
版本更新 2026-11-14 23:52
雷达观测 2026-11-17 12:06
arXiv ↗PDF ↗

摘要

Proactive, autonomous interactions receive high interest in the field of intelligent assistants, including the automotive domain. A central challenge, however, is identifying appropriate levels of autonomy for different functions while maintaining user acceptance and trust. This work examines how function characteristics relate to suitable autonomy levels and whether these relationships generalize across functions. We conducted an expert workshop to define autonomy levels and identify relevant function characteristics. We then collected data from 126 participants, who rated 14 in-vehicle functions regarding autonomy preferences and function characteristics, from which we trained machine learning models to predict appropriate autonomy levels. Building on our findings, we developed our prototypical Autonomy Design Assistant that supports designers in making autonomy-related decisions, which we evaluated qualitatively with nine experts. Our results suggest that function characteristics can rank functions by their suitable level of autonomy and thereby guide more informed autonomy design decisions, whereas predictions for individual users remain uncertain.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-14 23:52 · 最新观测版本

Towards Informed Autonomy Design: Understanding the Interplay of Function Characteristics and Autonomy Preferences in Automotive Agents

Proactive, autonomous interactions receive high interest in the field of intelligent assistants, including the automotive domain. A central challenge, however, is identifying appropriate levels of autonomy for different functions while maintaining user acceptance and trust. This work examines how function characteristics relate to suitable autonomy levels and whether these relationships generalize across functions. We conducted an expert workshop to define autonomy levels and identify relevant function characteristics. We then collected data from 126 participants, who rated 14 in-vehicle functions regarding autonomy preferences and function characteristics, from which we trained machine learning models to predict appropriate autonomy levels. Building on our findings, we developed our prototypical Autonomy Design Assistant that supports designers in making autonomy-related decisions, which we evaluated qualitatively with nine experts. Our results suggest that function characteristics can rank functions by their suitable level of autonomy and thereby guide more informed autonomy design decisions, whereas predictions for individual users remain uncertain.

查看此版本解读 →查看此版本原文 ↗