Learning to Control Flexible Objects Using Error-Tolerant Movement Strategies

学习使用容错运动策略控制灵活物体

• 批准号: 8212927
• 负责人: Christopher James Hasson
• 金额: $ 3.07万
• 依托单位: NORTHEASTERN UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-01-17 至 2012-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8212927
• 关键词: Activities of Daily Living; Address; Coffee; Development; Elderly; Elements; Future; Goals; Human; Individual; Lead; Learning; Life; Light; Methods; Movement; Nature; Nervous system structure; Noise; Outcome; Pattern; Performance; Physiological; Process; Recording of previous events; Research; Safety; Specific qualifier value; System; Task Performances; Time; Translations; Transportation; Work; base; flexibility; mathematical model; population movement; public health relevance; research study; response; robotic device

DESCRIPTION (provided by applicant): Humans must manipulate and transport a variety of objects when performing daily life activities. Simple rigid objects readily respond to human control actions and usually have no movement constraints. In contrast, flexible objects contain elements that cannot be directly controlled and pose unique constraints on manipulative actions. For example, coffee in a cup can only be controlled by manipulating the cup, and the coffee may spill if the movements are inappropriate. Most studies to date have focused on tasks with rigid objects that have a well-defined optimal movement pattern that leads to a single goal; however, many everyday tasks do not have precisely defined optimal strategies - many strategies lead to the desired goal. Importantly, some strategies may have less sensitivity to control errors compared to others, and these may be preferred. Extending previous research, we define an error-tolerant movement strategy as one that minimizes the effects of movement variability on task performance. Tolerance is particularly important in the control of flexible objects, where violations of energetic constraints can lead to a loss of stability with undesired consequences, such as spilling coffee. Critical unanswered questions for flexible object control include whether individuals are sensitive to error tolerance and how tolerance changes with learning. Accordingly, the goal of this research is to develop an analysis approach to quantify error tolerance in the continuous control of flexible objects, and examine whether humans learn error-tolerant strategies when transporting flexible objects. Three specific aims will be pursued: 1) develop an experimental paradigm based on the task of transporting a flexible object with energetic constraints, create a mathematical model of the task, and implement the task using a robotic device; 2) extend an existing analysis approach called TNC (Tolerance, Noise, Covariation) to the continuous transportation task; 3) show that humans learn error tolerant movement strategies for the transportation task during normal and perturbed task performances. This research combines theoretical and experimental work to address important gaps in our basic understanding of how humans manipulate flexible objects in the presence of movement variability and perturbations. Additionally, given that older adults have increased variability in their movements, the development of tolerant movement strategies may be a key feature of movement control in older adults. Hence, this study will lay the groundwork for future studies in older adults, which will explore whether older adults move slower as a strategic response to increase tolerance rather than simply due to physiological limitations. PUBLIC HEALTH RELEVANCE: When performing everyday activities such as carrying a cup of coffee, humans should move in a way that minimizes the chance that the coffee will be spilled if the cup is unexpectedly disturbed. This research proposes that humans naturally learn such error-tolerant movement strategies. It may explain why older adults move slower, which may be a strategy to increase error tolerance due to their increased movement variability.

描述（由申请人提供）：人类在进行日常生活活动时必须操纵和运输各种物体。简单的刚性物体很容易响应人类的控制动作，通常没有运动约束。相比之下，柔性对象包含无法直接控制的元素，并对操纵动作构成独特的约束。例如，杯子中的咖啡只能通过操纵杯子来控制，如果动作不适当，咖啡可能会溢出。迄今为止，大多数研究都集中在具有刚性物体的任务上，这些任务具有明确定义的最佳运动模式，从而导致单一目标;然而，许多日常任务没有精确定义的最佳策略-许多策略导致期望的目标。重要的是，与其他策略相比，一些策略可能对控制错误的敏感性较低，并且这些策略可能是优选的。扩展以前的研究，我们定义了一个容错的运动策略，最大限度地减少运动变异性对任务性能的影响。宽容是特别重要的控制灵活的对象，违反能源约束可能会导致失去稳定性与不希望的后果，如溢出咖啡。灵活的对象控制的关键未回答的问题包括个人是否对错误容忍度敏感，以及容忍度如何随着学习而变化。因此，本研究的目标是开发一种分析方法来量化柔性物体的连续控制中的容错能力，并研究人类在运输柔性物体时是否学习容错策略。本论文的主要目标是：1）建立一个基于柔性物体在能量约束下的搬运实验模型，建立数学模型，并利用机器人装置实现该任务; 2）扩展现有的分析方法TNC（公差，噪声，协变）的连续运输任务; 3）表明人类在正常和扰动任务执行过程中学习运输任务的容错运动策略。这项研究结合了理论和实验工作，以解决我们对人类如何在运动变异性和扰动存在的情况下操纵柔性物体的基本理解中的重要空白。此外，鉴于老年人动作的变异性增加，制定宽容的动作策略可能是老年人动作控制的一个关键特征。因此，这项研究将为未来在老年人中的研究奠定基础，这些研究将探讨老年人移动较慢是否是增加耐受性的战略反应，而不仅仅是由于生理限制。 公共卫生关系：当进行日常活动时，例如端着一杯咖啡，人们应该以一种方式移动，以最大限度地减少咖啡在杯子意外被打扰时溢出的机会。这项研究表明，人类自然学习这种容错的运动策略。这可以解释为什么老年人移动较慢，这可能是一种策略，以增加错误的容忍度，由于他们增加的运动变异性。