自动化水平及依赖对多域指挥认知增强的影响

doi:10.12305/j.issn.1001-506X.2026.10.19

系统工程与电子技术 ›› 2026, Vol. 48 ›› Issue (10): 3473-3484.doi: 10.12305/j.issn.1001-506X.2026.10.19

• 系统工程 • 上一篇

自动化水平及依赖对多域指挥认知增强的影响

曲昊¹^,²(), 吴晓莉¹^,²(), 晏彪¹^,²(), 张欣悦¹^,²(), 武愈涵¹^,²(), 吴传宇¹^,²(), 欧依琳¹^,²()

1. 南京理工大学人机融合与智能交互研究中心，江苏南京 210094
2. 语言信息智能处理及应用工信部重点实验室，江苏南京 210094

收稿日期:2025-09-24 接受日期:2026-02-11 出版日期:2026-10-25 发布日期:2026-09-30
通讯作者: 吴晓莉 E-mail:quhao_1218@163.com;wuxl@njust.edu.cn;yanbiao0109@163.com;969264885@qq.com;15951975551@163.com;chuanyuw213@163.com;oulihn@163.com
作者简介:曲　昊（2001—），女，硕士研究生，主要研究方向为认知增强
晏　彪（1996—），男，博士研究生，主要研究方向为人机交互
张欣悦（2001—），女，硕士研究生，主要研究方向为多模态交互
武愈涵（2001—），女，硕士研究生，主要研究方向为人机交互
吴传宇（2002—），男，硕士研究生，主要研究方向为多模态生理测评
欧依琳（2001—），女，硕士研究生，主要研究方向为人机交互
基金资助:
江苏省自然科学基金（BK20221490）;国家自然科学基金（52175469）资助课题

Impact of automation levels and dependency on cognitive enhancement for multi-domain command

Hao Qu¹^,²(), Xiaoli Wu¹^,²(), Biao Yan¹^,²(), Xinyue Zhang¹^,²(), Yuhan Wu¹^,²(), Chuanyu Wu¹^,²(), Yilin Ou¹^,²()

1. Research Centre of Human Cyber Physical Integration and Intelligent Interaction，Nanjing University of Science and Technology，Nanjing 210094，China
2. Key Laboratory of Ministry Industry and Information Technology for Language Information Processing and Application，Nanjing 210094，China

Received:2025-09-24 Accepted:2026-02-11 Online:2026-10-25 Published:2026-09-30
Contact: Xiaoli Wu E-mail:quhao_1218@163.com;wuxl@njust.edu.cn;yanbiao0109@163.com;969264885@qq.com;15951975551@163.com;chuanyuw213@163.com;oulihn@163.com

摘要/Abstract

摘要：

在舰上多域协同作战环境中，自动化水平对指挥员认知表现具有显著影响。针对自动化水平对指挥员认知表现影响机制尚不明确的问题，提出一种基于多源数据融合的认知评估方法。该方法首先构建典型舰上多域指挥任务并设置四级自动化水平，然后采集被试的行为数据、眼动数据与主观负荷评分，最后对不同自动化条件下的认知表现进行综合分析，识别最优配置区间。实验结果显示，低自动化条件下被试认知参与度较高但负荷较重，高自动化条件下任务效率提升但正确率有所下降，并出现一定程度的自动化依赖。结果表明二级自动化在效率与认知增强之间实现较优平衡，为舰上指挥系统人机协同设计提供参考。

关键词: 人机协同, 认知增强, 舰上指挥任务, 自动化水平, 主动思维, 眼动追踪

Abstract:

In shipboard multi-domain collaborative operational environment, the level of automation exerts a significant influence on commanders’ cognitive performance. To address the unclear mechanism by which automation levels influence cognitive performance, this study proposes a cognitive evaluation method based on multi-source data fusion. The method first constructs a typical shipboard multi-domain command task with four levels of automation. Then it collects behavioral data, eye-tracking data, and subjective workload ratings from participants. Finally, it conducts a comprehensive analysis of cognitive performance under different automation conditions to identify the optimal configuration range. Experimental results indicate that lower levels of automation are associated with higher cognitive engagement but heavier workload, whereas higher levels improve task efficiency while reducing accuracy and leading to a certain degree of automation dependency. Overall, Level 2 automation provides an optimal balance between efficiency and cognitive enhancement, offering practical guidance for the human-machine collaboration design of shipboard command systems.

Key words: human-machine collaboration, cognitive enhancement, shipboard command task, level of automation, active thinking, eye tracking

中图分类号:

曲昊, 吴晓莉, 晏彪, 张欣悦, 武愈涵, 吴传宇, 欧依琳. 自动化水平及依赖对多域指挥认知增强的影响[J]. 系统工程与电子技术, 2026, 48(10): 3473-3484.

Hao Qu, Xiaoli Wu, Biao Yan, Xinyue Zhang, Yuhan Wu, Chuanyu Wu, Yilin Ou. Impact of automation levels and dependency on cognitive enhancement for multi-domain command[J]. Systems Engineering and Electronics, 2026, 48(10): 3473-3484.

图/表 17

图1

表1

图2

图3

图4

表2

表3

表4

表5

表6

表7

图5

图6

图7

图8

图9

图10

参考文献 33

1	Vincenzi D A, Terwilliger B A, Ison D C. Unmanned aerial system （UAS） human-machine interfaces: new paradigms in command and control[J]. Procedia Manufacturing, 2015, 3, 920. doi: 10.1016/j.promfg.2015.07.139
2	Robinson R M, Mccourt M J, Marathe A R, et al. Degree of automation in command and control decision support systems[C]//IEEE International Conference on Systems, Man, and Cybernetics, 2016: 1184.
3	Pang E S, Dai L C. Research on task complexity measurements in human-computer interaction in nuclear power plant DCS systems based on emergency operating procedures[J]. Entropy, 2025, 27 (6): 600. doi: 10.3390/e27060600
4	Poornikoo M, Øvergård K I. Levels of automation in maritime autonomous surface ships （MASS）: a fuzzy logic approach[J]. Maritime Economics & Logistics, 2022, 24 (2): 278. doi: 10.1057/s41278-022-00215-z
5	Veitch E, Alsos O A. A systematic review of human-AI interaction in autonomous ship systems[J]. Safety Science, 2022, 152, 105778. doi: 10.1016/j.ssci.2022.105778
6	Parasuraman R, Riley V. Humans and automation: use, misuse, disuse, abuse[J]. Human Factors, 1997, 39 (2): 230.
7	Parasuraman R, Sheridan T B, Wickens C D. A model for types and levels of human interaction with automation[J]. IEEE Trans. on Systems, Man, and Cybernetics-Part A: Systems and Humans, 2000, 30 (3): 286. doi: 10.1109/3468.844354
8	He R Z, He H, Zhang Y X, et al. Automating dependency updates in practice: an exploratory study on github dependabot[J]. IEEE Trans. on Software Engineering, 2023, 49 (8): 4004. doi: 10.1109/TSE.2023.3278129
9	Roberts E, Abbott K. Investigating the under-reporting of automation dependency in air transport accidents worldwide[J]. Transportation Research Procedia, 2025, 88, 217. doi: 10.1016/j.trpro.2025.05.027
10	Figalová N, Bieg H J, Reiser J E, et al. From driver to supervisor: comparing cognitive load and EEG-based attentional resource allocation across automation levels[J]. International Journal of Human-Computer Studies, 2024, 182, 103169. doi: 10.1016/j.ijhcs.2023.103169
11	Passalacqua M, Pellerin R, Yahia E, et al. Practice with less AI makes perfect: partially automated AI during training leads to better worker motivation, engagement, and skill acquisition[J]. International Journal of Human-Computer Interaction, 2025, 41 (4): 2268. doi: 10.1080/10447318.2024.2319914
12	Alberti A L, Agarwal V, Gutowska I, et al. Automation levels for nuclear reactor operations: a revised perspective[J]. Progress in Nuclear Energy, 2023, 157, 104559. doi: 10.1016/j.pnucene.2022.104559
13	Markauskaite L, Marrone R, Poquet O, et al. Rethinking the entwinement between artificial intelligence and human learning: what capabilities do learners need for a world with AI[J]. Computers and Education: Artificial Intelligence, 2022, 3, 100056. doi: 10.1016/j.caeai.2022.100056
14	Bostrom N, Sandberg A. Cognitive enhancement: methods, ethics, regulatory challenges[J]. Science and Engineering Ethics, 2009, 15, 311.
15	Fu Q W, Zhang L J, Xu Y Q, et al. The review of human-machine collaborative intelligent interaction with driver cognition in the loop[J]. Systems Research and Behavioral Science, 2025, 42 (4): 954. doi: 10.1002/sres.3141
16	Broadbent D P, D‘innocenzo G, Ellmers T J, et al. Cognitive load, working memory capacity and driving performance: a preliminary fNIRS and eye tracking study[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2023, 92, 121. doi: 10.1016/j.trf.2022.11.013
17	Daronnat S, Azzopardi L, Halvey M, et al. Inferring trust from users’ behaviours; agents’ predictability positively affects trust, task performance and cognitive load in human-agent real-time collaboration[J]. Frontiers in Robotics and AI, 2021, 8, 642201. doi: 10.3389/frobt.2021.642201
18	Zeitlhofer I, Zumbach J, Schweppe J. Complexity affects performance, cognitive load, and awareness[J]. Learning and Instruction, 2024, 94, 102001.
19	Boyer M, Cummings M L, Spence L B, et al. Investigating mental workload changes in a long duration supervisory control task[J]. Interacting with Computers, 2015, 27 (5): 512. doi: 10.1093/iwc/iwv012
20	Mao R Z, Li G Y, Hildre H P, et al. A survey of eye tracking in automobile and aviation studies: implications for eye-tracking studies in marine operations[J]. IEEE Trans. on Human-Machine Systems, 2021, 51 (2): 87. doi: 10.1109/THMS.2021.3053196
21	Xu K M, Koorn P, De Koning B, et al. A growth mindset lowers perceived cognitive load and improves learning: integrating motivation to cognitive load[J]. Journal of Educational Psychology, 2021, 113 (6): 1177. doi: 10.1037/edu0000631.supp
22	Yan K T, Shao J, Zhu Z D, et al. Display interface design for rollers based on cognitive load of operator[J]. Journal of the Society for Information Display, 2021, 29 (8): 659. doi: 10.1002/jsid.1009
23	Wenk N, Penalver-andres J, Buetler K A, et al. Effect of immersive visualization technologies on cognitive load, motivation, usability, and embodiment[J]. Virtual Reality, 2023, 27 (1): 307. doi: 10.1007/s10055-021-00565-8
24	Sethupathy U K A. Empowering intelligent decision-making: architecting resilient real-time data platforms with actionable visual dashboards[J]. International Journal of Advanced Engineering Technologies and Innovations, 2021, 1(2): 211.
25	Huang Q, Xu X, Wei Y, et al. The impacts of level of automation and cognitive secondary task on the cognitive load of armored vehicle crews[J]. Cognition, Technology & Work, 2025, 27 (3): 541.
26	Gkintoni E, Antonopoulou H, Sortwell A, et al. Challenging cognitive load theory: the role of educational neuroscience and artificial intelligence in redefining learning efficacy[J]. Brain Sciences, 2025, 15 (2): 203. doi: 10.3390/brainsci15020203
27	韩玉龙, 严建钢, 林云, 等. 舰载无人机编队协同对海突击作战关键技术综述[J]. 飞航导弹, 2015, (5): 47.
28	Greenacre M, Groenen P J F, Hastie T, et al. Principal component analysis[J]. Nature Reviews Methods Primers, 2022, 2 (1): 100. doi: 10.1038/s43586-022-00184-w
29	Gouraud J, Delorme A, Berberian B. Influence of automation on mind wandering frequency in sustained attention[J]. Consciousness and Cognition, 2018, 66, 54. doi: 10.1016/j.concog.2018.09.012
30	Wohleber R W, Matthews G, Lin J, et al. Vigilance and automation dependence in operation of multiple unmanned aerial systems （UAS）: a simulation study[J]. Human Factors, 2019, 61 (3): 488. doi: 10.1177/0018720818799468
31	Causse M, Mercier M, Lefrançois O, et al. Impact of automation level on airline pilots’ flying performance and visual scanning strategies: a full flight simulator study[J]. Applied Ergonomics, 2025, 125, 104456. doi: 10.1016/j.apergo.2024.104456
32	Yang B, Inoue K, Yan Z H, et al. Influences of level 2 automated driving on driver behaviors: a comparison with manual driving[J]. IEEE Trans. on Intelligent Transportation Systems, 2023, 25 (1): 144. doi: 10.1109/tits.2023.3308569
33	Cui X, Zhang Y J, Zhou Y W, et al. Measurements of team workload: a time pressure and scenario complexity study for maritime operation tasks[J]. International Journal of Industrial Ergonomics, 2021, 83, 103110. doi: 10.1016/j.ergon.2021.103110

任务流程	任务目标
发现目标	监控态势地图区域，及时发现潜在威胁
目标处理	对可疑目标进行目标类型的识别与威胁等级的分类，做出应对策略
攻击决策	根据目标信息决策是否攻击目标，选择攻击平台与攻击方式
行动执行	最终确认攻击执行

指标类别	眼动指标
注视类	TDF
	ADF
	NF
	TFF
	DFF
访问类	TDV
	ADV
	NV

被试编号	TDF_Mean	ADF_Mean	NF_Mean	TFF_Mean	DFF_Mean	TDV_Mean	ADV_Mean	NV_Mean
1	1.10	0.16	3.00	3.12	0.13	1.20	0.54	1.06
2	1.18	0.17	3.67	4.16	0.15	1.27	0.49	1.11
3	2.48	0.21	7.24	7.72	0.17	2.76	0.72	2.17
4	2.20	0.21	5.08	8.22	0.18	2.33	0.86	1.64
5	2.64	0.27	6.15	6.79	0.25	2.81	0.90	1.88
6	1.00	0.11	3.88	4.16	0.12	1.15	0.35	1.40
7	1.98	0.21	5.03	4.24	0.17	2.12	0.60	1.68
8	0.92	0.12	4.00	4.04	0.12	1.20	0.40	1.44
9	1.14	0.18	2.93	3.70	0.15	1.22	0.38	1.15
10	1.01	0.17	2.67	4.69	0.16	1.08	0.42	1.24
11	1.47	0.16	3.94	4.78	0.14	1.60	0.42	1.40
12	1.26	0.19	3.00	2.86	0.16	1.34	0.51	0.96
13	1.07	0.16	2.88	3.59	0.15	1.13	0.41	1.11
14	0.23	0.07	1.38	2.75	0.08	0.43	0.30	0.67
15	1.82	0.20	6.26	5.72	0.20	2.20	0.60	2.07
16	0.87	0.13	2.85	2.16	0.10	0.96	0.34	1.17
17	0.84	0.16	3.19	4.07	0.16	0.96	0.31	1.58
18	1.10	0.11	3.78	4.08	0.10	1.18	0.28	1.57

原始变量	主成分1
TDF_Mean	0.980
ADF_Mean	0.891
NF_Mean	0.926
TFF_Mean	0.880
DFF_Mean	0.855
TDV_Mean	0.985
ADV_Mean	0.904
NV_Mean	0.837

被试编号	FAC1
1	−0.49507
2	−0.23180
3	1.77318
4	1.40454
5	2.11401
6	−0.56717
7	0.75416
8	−0.51334
9	−0.37824
10	−0.33821
11	−0.03621
12	−0.31333
13	−0.50979
14	−1.81515
15	1.17291
16	−1.00724
17	−0.40400
18	−0.60926

自动化水平及依赖对多域指挥认知增强的影响

Impact of automation levels and dependency on cognitive enhancement for multi-domain command

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献 33

相关文章 13

编辑推荐

Metrics

本文评价

被试编号	ATI	分组
2	−0.23180	主动组
3	1.77318
4	1.40454
5	2.11401
7	0.75416
10	−0.33821
11	−0.03621
12	−0.31333
15	1.17291
1	−0.49507	依赖组
6	−0.56717
8	−0.51334
9	−0.37824
13	−0.50979
14	−1.81515
16	−1.00724
17	−0.40400
18	−0.60926

被试组别	评估维度	一级	二级	三级	四级
主动组	脑力需求	76.89	67.44	48.72	26.44
	体力需求	55.06	48.78	36.28	19.89
	时间需求	64.39	57.00	38.50	31.06
	绩效表现	60.50	65.83	69.94	70.50
	努力程度	71.56	58.61	44.33	29.44
	受挫程度	58.83	53.72	44.50	28.50
	总计	64.537	58.565	47.046	34.306
依赖组	脑力需求	77.61	62.39	39.94	21.28
	体力需求	35.50	27.33	18.28	10.06
	时间需求	69.00	49.67	30.78	20.78
	绩效表现	54.83	59.22	58.33	54.22
	努力程度	70.94	55.17	28.28	17.56
	受挫程度	58.06	47.22	40.06	23.89
	总计	60.991	50.167	35.944	24.630

[1]	张少辉, 李璐璐, 李亚飞, 吴庆顺, 李冠峰, 徐明亮. 舰载机弹药保障作业规划方法研究综述[J]. 系统工程与电子技术, 2026, 48(4): 1303-1321.
[2]	武愈涵, 吴晓莉, 张欣悦, 晏彪, 王名珺. 面向认知增强的MUM-T态势图视觉调控方法[J]. 系统工程与电子技术, 2026, 48(4): 1292-1302.
[3]	晏彪, 吴晓莉, 张蓝, 刘潇, 方泽茜, 韩炜毅, 李琦桉. 有人/无人机协同指挥员的事件相关电位特征[J]. 系统工程与电子技术, 2026, 48(2): 578-587.
[4]	魏建林, 林彦超, 唐慧龙, 张旺, 王伟. 基于改进MCTS的多无人机多任务联合决策[J]. 系统工程与电子技术, 2026, 48(2): 556-568.
[5]	李正杰, 刘光远, 张浩为, 刘斌, 齐铖. 面向射频隐身的多无人机协同区域覆盖航迹优化方法[J]. 系统工程与电子技术, 2026, 48(1): 301-311.
[6]	洪芳宇, 叶青, 张利宁, 伍国华. 面向区域搜索的车载多无人机协同任务规划方法[J]. 系统工程与电子技术, 2026, 48(1): 144-156.
[7]	胡崇爽, 王纪凯, 李明浩, 豆亚杰, 姜江. 基于IVSFS的作战威胁人机协同评估框架[J]. 系统工程与电子技术, 2025, 47(6): 1855-1866.
[8]	熊威, 张栋, 任智, 杨书恒. 面向有人/无人机协同打击的智能决策方法研究[J]. 系统工程与电子技术, 2025, 47(4): 1285-1299.
[9]	尹中杰, 侯博, 靳啸龙, 范志良, 王海洋. 面向阵列天线抗干扰无人机的隐蔽诱骗方法[J]. 系统工程与电子技术, 2025, 47(2): 633-640.
[10]	张进富, 曹云峰, 郭邦君, 王浩宇. 面向有人/无人机协同作战的体系架构正向建模方法[J]. 系统工程与电子技术, 2025, 47(12): 3935-3951.
[11]	王名珺, 吴晓莉, 晏彪, 张欣悦, 武愈涵. 基于人机协同作战多通道交互系统的隐性知识挖掘[J]. 系统工程与电子技术, 2025, 47(10): 3313-3324.
[12]	刘瑶, 夏阳升, 石建迈, 陈超, 黄金才. 车载多无人机协同多区域覆盖路径规划方法[J]. 系统工程与电子技术, 2023, 45(5): 1380-1390.
[13]	李洪瑶, 李小强, 韩心中, 谢学立, 席建祥. 基于决策融合的多无人机协同目标检测识别算法[J]. 系统工程与电子技术, 2022, 44(3): 746-754.