Exploration of optimal prosthesis feedback information using computational motor control

更好的机器人假肢可以极大地提高上肢截肢患者的生活质量，他们中的许多人拒绝使用现有的设备，因为他们很难用控制完整手臂的同样直观的、潜意识的方式来控制它们。假肢控制是困难的，因为截肢者在他们的设备是否会对他们的控制信号做出适当反应以及感觉反馈信号是否准确地反映实际运动方面都经历了巨大的不确定性。研究人员一直专注于改善控制的孤立方面，例如通过改进过滤器或通过触觉设备模仿健全的感觉提示，但这些方法最大限度地减少了假肢控制的不确定性。人与假肢的相互作用是一个多方面的、时变的问题，很难解决。机器人假肢研究中缺少的是优化控制策略和感觉线索的原则性方法，这些方法考虑了人们在面对高度不确定性时做出的行为选择。 我们独特的方法是使用一种优化策略，该策略结合了人类为了处理不确定性而直觉做出的行为决定。对于健康的受试者，基于人类行为数据的计算运动控制模型很好地描述了受试者如何学习、估计和控制。这甚至适用于类似于假体使用的情况，例如将非直观关节映射到抽象自由度或依赖于信号的噪声。这表明，这些模型还可以预测截肢者如何使用他们更嘈杂的控制信号和有限的感觉反馈来学习控制假肢。建立模型并通过实验对其进行校准，使其具有预测性，将使我们能够研究不同的解码器和反馈设计如何影响行为。这种模型驱动的方法保证了更快、更高效的假体设计。我们计划量化截肢者归因于各种来源(他们的控制信号、假肢和世界)的不确定性；开发新的控制器来减少控制的不确定性；并提供触觉感觉线索，与现有的传感器和控制策略协同工作，以减少不确定性。 这项拟议的研究具有创新性，因为它在更广泛的背景下提出了控制问题，其中包括人类在优化其控制策略和感官线索时做出的高度复杂的行为决定。这种方法能够以以前的方法无法实现的方式集成多种效果。例如，我们的方法自然结合了这样一个事实，即人们更喜欢使用较少的努力来完成一项任务，但在需要更高精度的运动部分(如轨迹的最后部分)可以容忍更多的努力。另一方面，我们的方法不支持高确定性的触觉线索，如果这些线索为现有的感觉线索如视觉提供了多余的信息，或者如果触觉信息不能减少可控系统动力学的不确定性。由于截肢者存在较大的控制信号噪声源，我们的工作将导致计算电机控制和最优控制领域的技术改进。这项研究建立在我们团队在上肢假体设计和控制方面的丰富经验以及我们的合作者在计算运动控制领域的经验的基础上。拟议目标的实现将促进机器人控制领域以及人-机器人交互、感知、操纵和外骨骼等不同领域的发展，并将为各级教育提供丰富的平台。

Better robotic prostheses can dramatically improve the quality of life for persons with an upper limb amputation, many of whom reject existing devices because they have trouble controlling them in the same intuitive, subconscious way that they controlled their intact arms. Prosthesis control is difficult because amputees experience great uncertainty both with respect to whether their device will respond appropriately to their control signals and whether sensory feedback cues accurately reflect the actual movement. Researchers have focused on improving isolated aspects of control, for example by improving filters or mimicking able-bodied sensory cues through haptic devices, but these approaches have minimally reduced the uncertainty of prosthesis control. Human interaction with a prosthesis is a multifaceted, time-varying problem that is difficult to solve. What is missing from robotic prosthesis research are principled methods for optimizing control strategies and sensory cues that take into account behavioral choices people are known to make in the face of high uncertainty. Our unique approach is to use an optimization strategy that incorporates the behavioral decisions that humans intuitively make in order to deal with uncertainty. For healthy subjects, computational motor control models based on human behavioral data describe very well how subjects learn, estimate, and control. This is even true for cases that are analogous to prosthesis use, such as mapping non-intuitive joints to abstract degrees of freedom or signal-dependent noise. This suggests that those models could also predict how an amputee learns to control a prosthesis using their noisier control signals and limited sensory feedback. Building models and calibrating them with experiments to make them predictive will allow us to study how different decoder and feedback designs would affect behavior. This model-driven approach promises faster and more efficient prosthesis design. We plan to quantify the uncertainty that amputees attribute to various sources (their control signals, the prosthesis, and the world); to develop novel controllers that reduce the uncertainty of control; and to provide haptic sensory cues that work synergistically with available sensors and control strategies to reduce uncertainty. The proposed research is innovative because it poses the control problem in a broader context that incorporates the highly sophisticated behavioral decisions that humans make in optimizing their control strategy and sensory cues. This approach is able to integrate multiple effects in ways that were not possible using previous approaches. For example, our approach naturally incorporates the fact that people prefer to use less exerted effort to accomplish a task, but tolerate more effort during portions of movement that require greater precision (e.g. final portion of a trajectory). On the other hand, our approach does not favor high-certainty haptic cues if those cues provide redundant information to existing sensory cues such as vision, or if the haptic information does not reduce the uncertainty of controllable system dynamics. Due to the large sources of control-signal noise present in amputees, our work will lead to improved techniques within the fields of computational motor control and optimal control. This research builds on our team’s extensive experience in the design and control of upper-limb prostheses and our collaborator’s experience in the field of computational motor control. Achievement of the proposed aims will contribute to the field of robotic control and to such diverse fields as human-robot interaction, perception, manipulation, and exoskeletons, and will provide a rich platform for education at all levels.