Collaborative Research: FMitF: Track I: Designing Safe and Robust Human-machine Interactions with Fuzzy Mental Models

人为错误是跨安全关键领域的故障的主要因素。人为错误的一个常见原因是模式混乱。模式混淆源于人类操作员对系统的心理模型与实际系统设计之间的错位。模式混乱可能会导致自动化意外，这可能会使操作员迷失方向，增加操作员出错的可能性，并侵蚀操作员对系统的信任。该项目旨在开发一种设计人机界面(HMI)的新方法，以最大限度地减少此类错误。特别是，该项目将开发一个新颖的、正式的心理建模基础，以及严格评估HMI的安全性和消除可能的错误的技术。这项研究中产生的知识和工具将提供给研究人员和设计人员，并有可能应用于广泛的安全关键系统。反过来，这些知识将有助于避免系统灾难，防止伤害，拯救生命，并保护全社会的关键资源。这个项目的核心思想是模糊的心理模型。这些将明确建模人类操作员如何理解系统状态和功能所固有的模糊性。与现有的心理模型概念相比，这种新型的心理模型将能够捕捉到更广泛的人机交互问题。为了支持这种新的心理建模方法，该项目将开发以下方法：(1)系统地从人类中引出模糊心理模型，(2)自动分析HMI以识别心理模型和系统之间的潜在错位，以及(3)生成修复HMI中的错位缺陷和减少人为错误的可能性的建议。这些方法的有效性将在汽车、医疗设备、网络安全和航空系统等多个领域的HMI上得到演示。这项工作将改进评估界面可用性的方法，扩大正式方法在新环境中的应用，并为研究人员、设计师和工程师提供资源，以提高网络人类系统的可靠性。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Human error is a major factor in failures across safety-critical domains. One common cause of human error is mode confusion. Mode confusion arises from a misalignment between a human operator's mental model of the system and the actual system design. Mode confusion can result in automation surprise, which can disorient the operator, increase the likelihood of operator mistakes, and erode the operator's trust in the system. This project aims to develop a new method for designing human-machine interfaces (HMIs) that minimizes this type of error. In particular, the project will develop a novel, formal foundation of mental modeling, along with techniques for rigorously evaluating the safety of HMIs and eliminating possible errors. The knowledge and tools produced in this research will be made available to researchers and designers and have potential applications to a wide range of safety-critical systems. This knowledge, in turn, will help avoid system disasters, prevent injuries, save lives, and protect critical resources across society.The central idea in this project is fuzzy mental models. These will explicitly model the vagueness inherent to how human operators understand system states and functions. This new type of mental model will be capable of capturing a wider range of human-machine interaction problems than are possible with existing mental model concepts. In support of this new mental modeling approach, this project will develop methods for (1) systematically eliciting fuzzy mental models from humans, (2) automatically analyzing an HMI to identify potential misalignments between a mental model and the system, and (3) generating suggestions for repairing misalignment flaws in HMIs and reducing the likelihood of human errors. The efficacy of these methods will be demonstrated on HMIs across a number of domains, such as automobiles, medical devices, web security, and aeronautical systems. The work will lead to improved methods for evaluating the usability of interfaces, widen the application of formal methods to new contexts, and provide resources for researchers, designers, and engineers to improve the reliability of cyber-human systems.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.