| 摘要: |
| 目的 针对人机混合智能在航空领域中的应用进行探索，以提升飞行绩效和安全为核心目标。方法 构建了包含语言动作感知、生理模态感知和认知行为感知的综合智能感知系统，以实现对飞行员状态的全面理解。通过引入特征可视化、局部可解释和交互透明化技术，提高人工智能决策的透明度与可信度。详细分析了人在环内、人在环上及人在环外的三种人机协同模式的特点及适用场景。结果 提出了一种动态人机协同决策范式。该范式将机器的计算优势与人类的直觉感知、创造性决策优势相结合。结论 为航空智能化发展提供了理论支持和实践路径。 |
| 关键词: 混合智能 人机协同 语言动作感知 生理模态感知 透明决策 |
| DOI：10.19554/j.cnki.1001-3563.2025.08.004 |
| 分类号: |
| 基金项目: |
|
| Application Paradigm of Human-machine Hybrid Intelligence in Aerospace Field |
|
WANG Yuanhang, LI Zhen, XU Gang, GAO Zhongqi, WANG Yihang
|
|
(Shenyang Aircraft Design & Research Institute of AVIC, Shenyang 110035, China)
|
| Abstract: |
| The work aims to explore the application of human-machine hybrid intelligence in the aerospace field, with a core focus on enhancing the flight performance and safety. A comprehensive intelligent perception system integrating language and motion perception, physiological modality perception, and cognitive behavior perception was constructed to achieve a full understanding of the pilot's state. By introducing techniques such as feature visualization, local interpretability, and interactive transparency, the transparency and credibility of artificial intelligence decision-making were improved. It provided an in-depth analysis on the characteristics and applicable scenarios of three human-machine collaboration modes:human-in-the-loop, human-on-the-loop, and human-out-of-the-loop. A dynamic human-machine collaborative decision-making paradigm was proposed. This paradigm combined the computational advantages of the machine with the intuitive and creative decision-making capabilities of humans, providing theoretical support and practical pathways for the development of aviation intelligence. |
| Key words: ive Summarization and Rating-based Sentiment Classification using Deep Transfer Learning[J]. International Journal of Information Management Data Insights, 2024, 4(2):1-9. |
