Measurement and simulation of human-device interaction: application to lower-limb orthoses

在接下来的十年里，预计将有一个快速扩散的设备，如支架和动力外骨骼，帮助人类运动。 虽然被动支架已经使用多年，并且一些外骨骼设备现在已被批准上市，但设备设计师仍然不完全了解当前设备是否安全有效，或者应该如何调整设备以实现不同个体的最佳性能。 我的研究计划将解决这一领域的两个主要限制，并广泛应用于辅助人类运动的设备。 首先，我们没有足够的工具来测量和预测外部辅助设备如何与人体四肢的软组织相互作用。 当外部设备向身体施加力时，下面的皮肤、肌肉和脂肪会发生显著变形。 这引起装置和人体关节轴线的未对准，这不仅影响力传递的效率，而且还引起皮肤和其他组织的不期望的和不舒适的剪切载荷。 在拟议的研究中，我们将开发简单，低成本的设备来测量个人肢体的组织硬度，然后创建计算模型来预测如何为每个人优化辅助设备设计。 我们将使用一系列实验测试案例来开发和验证这种方法，从简单的1-DOF压痕开始，到人类下肢和仪器矫形器之间的6-DOF交互（例如，膝盖和脚踝支架），其具有被动弹簧和主动致动器控制。我们缺乏足够的工具来预测当外部设备施加力以辅助运动时人类神经肌肉系统将如何响应。 由于测量人体肌肉产生的力通常是不可行的，因此通常使用计算模型来估计肌肉力。 然而，传统的模型假设肌肉的努力是协调的，只是为了最大限度地提高耐力，这些模型没有考虑到改变肌肉协调模式的残疾人，或使用外部设备的人。 在拟议的研究中，我们将探索新的模拟方法来估计肌肉力量，同时考虑外部设备的稳定作用。 我们的方法将使用竞争目标的加权组合，以最大限度地减少代谢成本，同时实现髋关节，膝关节和踝关节的理想关节刚度。 最后，预计本研究计划中开发的实验和计算工具将为学术界和工业界的辅助设备设计者带来直接利益。 利用我们的实验和计算专业知识，可以测试当前设备如何影响人体软组织肌肉，韧带和软骨的内部负荷。 新的合作将导致设计新颖，更有效的辅助设备。

Over the next decade, there is expected to be a rapid proliferation of devices, such as braces and powered exoskeletons, that assist human locomotion. While passive braces have been used for years, and some exoskeleton devices are now approved for market, device designers still don't fully understand whether current devices are safe and effective, or how devices should be tuned to achieve optimal performance for different individuals. My research program will address two primary limitations in this field, with broad applications for devices that assist human motion. First, we don't have adequate tools to measure and predict how external assistive devices interact with soft tissues of human limbs. When external devices apply forces to the body, there is significant deformation of underlying skin, muscle, and fat. This induces misalignment of device and human joint axes, which not only affects the efficiency of force transmission but also causes undesirable and uncomfortable shear loading of skin and other tissues. In the proposed research, we will develop simple, low-cost devices to measure tissue stiffness across an individual's limb, then create computational models to predict how assistive device design might be optimized for each individual. We will develop and validate this approach using a progression of experimental test cases, beginning with simple 1-DOF indentation and culminating in full 6-DOF interaction between the human lower-limb and instrumented orthoses (e.g. knee and ankle braces) with both passive springs and active actuator controls.Second, we lack adequate tools to predict how the human neuromuscular system will respond when external devices apply forces to assist locomotion. Since it is not generally feasible to measure the forces generated by human muscles, computational models are often used to estimate muscle forces. Yet, traditional models assume that muscular efforts are coordinated only to maximize endurance, and these models do not account for altered muscle coordination patterns seen in people with disabilities, or people using external devices. In the proposed research, we will explore novel simulation approaches to estimate muscle forces while accounting for the stabilizing effect of external devices. Our approach will use a weighted combination of competing objectives to minimize metabolic cost while achieving the desired joint stiffness at hip, knee, and ankle joints. Finally, it is anticipated that the experimental and computational tools developed in this research program will yield direct benefits for assistive device designers in both academia and industry. Leveraging our experimental and computational expertise, it will be feasible to test how current devices affect internal loading of soft tissues muscles, ligaments, and cartilage - in the human body. New collaborations will lead to the design of novel, more effective assistive devices.