团队合作
动作(物理)
模棱两可
任务(项目管理)
人机交互
感觉线索
人机交互
认知心理学
计算机科学
灵活性(工程)
心理学
机器人
人工智能
工程类
物理
程序设计语言
法学
系统工程
统计
量子力学
数学
政治学
作者
Julie Shah,Cynthia Breazeal
出处
期刊:Human Factors
[SAGE Publishing]
日期:2010-04-01
卷期号:52 (2): 234-245
被引量:116
标识
DOI:10.1177/0018720809350882
摘要
OBJECTIVE: We conducted an empirical analysis of human teamwork to investigate the ways teammates incorporate coordination behaviors, including verbal and nonverbal cues, into their action planning. BACKGROUND: In space, military, aviation, and medical industries, teams of people effectively coordinate to perform complex tasks under stress induced by uncertainty, ambiguity, and time pressure. As robots increasingly are introduced into these domains, we seek to understand effective human-team coordination to inform natural and effective human-robot coordination. METHOD: We conducted teamwork experiments in which teams of two people performed a complex task, involving ordering, timing, and resource constraints. Half the teams performed under time pressure, and half performed without time pressure. We cataloged the coordination behaviors used by each team and analyzed the speed of response and specificity of each coordination behavior. RESULTS: Analysis shows that teammates respond to explicit cues, including commands meant to control actions, more quickly than implicit cues, which include short verbal and gestural attention getters and status updates. Analysis also shows that nearly all explicit cues and implicit gestural cues were used to refer to one specific action, whereas approximately half of implicit cues did not often refer to one specific action. CONCLUSION: These results provide insight into how human teams use coordination behaviors in their action planning. For example, implicit cues seem to offer the teammate flexibility on when to perform the indicated action, whereas explicit cues seem to demand immediate response. APPLICATION: We discuss how these findings inform the design of more natural and fluid human-robot teaming.
科研通智能强力驱动
Strongly Powered by AbleSci AI