多模态交互技术在自动驾驶舱环境中的应用

中图分类号：U463.67 文献标识码：A 文章编号：1003-8639（2025）09-0013-03

ApplicationofMultimodal Interaction TechnologyinAutomatic CockpitEnvironmen

Lü Jiawei，Lu Danxuan （Henan Vocational College of Surveying and Mapping， Zhengzhou 45oooo， China）

【Abstract】With the continuous development of intelligent connected vehicle technology，multi-modal interaction technology is increasingly widelyused in thefieldof autonomous driving.Therefore，this paperdeeply studies the applicationof multimodalinteraction technology inthe automaticcockpit environment，focusingon speech， gesture，physiological perception and soon.At thesame time，onthebasisof in-depth analysisof thecorechallnges such as modal conflictand system robustnessdecline inthecabin，a system-level solution including modaldecoupling structure，adaptive fault-tolerant mechanism and semantic discrimination priority strategy is proposed， hoping to promote the healthydevelopment of inteligentconnectedvehicle industry，sothat autonomous driving technologycan better serve human society.

【Key words】automatic cockpit; multimodal interaction; semantic fusion

0 引言

在智能化、信息化、互联化数字技术迭代加速的时代背景下，智能网联汽车正从“交通工具”向“移动智能空间”转型，这一转型推动汽车人机交互（HMI）关系发生根本性变革——已从传统单一的物理仪表按键交互，逐步迈向融合语音、手势、生理感知等多通道的场景化多模态交互。（剩余4550字）