CRII: CPS: Intuitive Human-in-the-Loop Control for Medical Cyber-Physical Systems

人在环控制策略，其中用户执行任务更好，感觉更有信心这样做，是一个重要的研究领域的网络物理系统。人类非常擅长学习控制复杂的系统，特别是那些具有非直觉运动约束的系统（例如，汽车、自行车、轮椅、可操纵的针）。随着网络物理系统（物理系统与网络控制层集成）的出现，人类控制不再局限于系统输入。用户还可以通过许多不同的远程操作映射来控制系统输出。考虑到所有这些灵活性，人类用户控制任意系统最直观的方式是什么？如何量化直觉？该项目的重点是医疗针头的人在环控制，它以类似自行车的运动学来操纵。这些针可用于多种医疗干预，包括组织活检、肿瘤消融、脓肿引流和局部药物输送。我们已经探索了人类控制这些可操纵针的各种远程操作映射；然而，我们发现了客观性能指标（例如，任务时间和错误）与实验后舒适性或易用性调查之间的不一致。用户偶尔会报告对控制映射的偏好，这在客观上降低了性能，反之亦然。为了捕捉不同的控制映射如何影响体力、精神负荷、分心、困倦和情绪反应的细微差别，测量用户与物理系统的实时互动非常重要。生理传感器，如脑电图（EEG）、皮肤电反应（GSR）和肌电图（EMG），可用于提供这些实时测量，并定量分类新的远程操作算法的直观性。更广泛的影响：直观和自然的人在环控制界面将通过在手术和康复中的应用改善人类的健康和福祉。这项研究的结果将在研究者的实验室网站、会议研讨会以及德克萨斯大学达拉斯分校和德克萨斯大学西南医学中心联合举办的新的医疗机器人研讨会上公开发布。外展活动、实验室参观以及对各级代表性不足的学生的指导，将扩大STEM的参与。另外，调查员的距离？面向医疗专业人员的美国医院实验室将让非工程师参与设计和创新

Human-in-the-loop control strategies in which the user performs a task better, and feels more confident to do so, is an important area of research for cyber-physical systems. Humans are very adept at learning to control complex systems, particularly those with non-intuitive kinematic constraints (e.g., cars, bicycles, wheelchairs, steerable needles). With the advent of cyber-physical systems, (physical systems integrated with cyber control layer), human control is no longer constrained to system inputs. Users can also control system outputs through a number of different teleoperation mappings. Given all this flexibility, what is the most intuitive way for a human user to control an arbitrary system and how is intuitiveness quantified?The project focuses on human-in-the-loop control for medical needles, which steer with bicycle-like kinematics. These needles could be used in a variety of medical interventions including tissue biopsy, tumor ablation, abscess drainage, and local drug delivery. We have explored a variety of teleoperation mappings for human control of these steerable needles; yet, we have found inconsistencies between objective performance metrics (e.g., task time and error), and post-experimental surveys on comfort or ease-of use. Users occasionally report a preference for control mappings, which objectively degrade performance, and vice versa. It is important to measure the real-time engagement of the user with the physical system in order to capture the nuances of how different control mappings affect physical effort, mental workload, distraction, drowsiness, and emotional response. Physiological sensors such as electroencephalography (EEG), galvanic skin response (GSR), and electromyography (EMG), can be used to provide these real-time measurements and to quantitatively classify the intuitiveness of new teleoperation algorithms.Broader Impacts: Intuitive and natural human-in-the-loop control interfaces will improve human health and well being, through applications in surgery and rehabilitation. The results of this study will be disseminated publicly on the investigator's laboratory website, a conference workshop, and a new medical robotics seminar to be held jointly between UT Dallas and UT Southwestern Medical Center. Outreach activities, lab tours, and mentoring of underrepresented students at all levels, will broaden participation in STEM. Additionally, the proximity of the investigator?s hospital-based lab to medical professionals will engage non-engineers in design and innovation