MULTIJOINT CONTROL FOR LOWER EXTREMITY NEUROPROSTHESES

描述(改编自申请人的摘要)：长期 本研究的目标是开发电刺激系统， 为截瘫和偏瘫患者提供居家和工作场所的移动性 因脊髓损伤、中风或其他原因而瘫痪的患者 头部受伤。这项工作的目的是开发 不需要大量矫形器的康复装置。那里 是需要改进对运动的控制，从而达到更好的效果 平衡、更平稳、更节能的运动，可靠 完成爬楼梯、下楼梯、拐杖行走。这 项目将解决关键控制问题，其解决方案是 神经假体达到这些目标所必需的。潜在的 拟议工作的科学假设是，特定的运动 膝盖、脚踝和臀部可以通过反馈控制器来实现， 在实时测量关节角度的基础上，足部终点 位置和脚的接触力，调制共同刺激 肌肉，从而产生净力矩和机械阻抗 每个关节，从而获得所需的运动。第二个假设是 有多组刺激波形可以达到同样的效果 运动，并且每组中涉及的协同刺激量是 关系到控制系统达到规定要求的能力 尽管有机械干扰，但仍有轨道。第三个假设是 协同刺激量与有效时间长短有关 控制(例如，疲劳抗力)。该项目将测试这些 通过开发和表征这种性能的假设 反馈控制器。具体目标是：1)发展和发展 电刺激肌肉预测模型的实验验证 回应。将为脚踝、膝盖和髋关节开发模型 (单独和组合)，并将考虑共同刺激 激动型肌肉电极、刺激脉冲频率的调制和 宽度、被动和外部机械载荷以及肌肉疲劳。 将电刺激和关节角度与生成的网络相关联的模型 关节力矩以及由此产生的主动刚度和阻尼力 将考虑生物力学系统；2)开发和评估 调节电刺激的反馈控制器的能力 达到关节角度和/或接触力的指定轨迹； 3)将这些控制器中最好的部分应用到实现中 截瘫患者的楼梯上升和下降改善，通过 电脑控制的电刺激。

DESCRIPTION (Adapted from the Applicant's Abstract): The long-term goal of this research is to develop electrical stimulation systems that provide home and workplace mobility to paraplegic and hemiplegic patients having paralysis arising from spinal cord injury, stroke or head injury. This work is directed toward the development of rehabilitative devices that do not require substantial orthoses. There is a need for improved control of motion, so as to achieve better balance, smoother and more energy efficient locomotion, reliable accomplishment of stair climbing and descent, and crutch walking. This project will addresses critical control problems whose solution is necessary for neuroprostheses to attain these goals. The underlying scientific hypothesis of the proposed work is that specified motions of the knee, ankle and hip can be achieved by feedback controllers which, on the basis of real time measurements of joint angles, foot endpoint position and foot contact forces, modulate the co-stimulation of muscles, and consequently the net moments and mechanical impedances at each joint, so as to obtain the desired motion. A second hypothesis is that there are multiple sets of stimulus waveforms that achieve the same motion, and that the amount of co-stimulation involved in each set is related to the ability of the control system to achieve the prescribed trajectory despite mechanical disturbances. A third hypothesis is that the amount of co-stimulation is related to the duration of effective control (e.g., the fatigue resistance). This project will test these hypotheses by developing and characterizing the performance of such feedback controllers. Specific Aims are: 1) to develop and experimentally verify predictive models of electrically-stimulated muscle response. Models will be developed for the ankle, knee and hip joints (separately and in combination), and will account for co-stimulation of agonist muscle electrodes, modulation of stimulus pulse frequency and width, passive and external mechanical loads, and muscle fatigue. Models relating electrical stimulation and joint angles to generated net joint moments and the active stiffness and damping of the resulting biomechanical systems will be considered; 2) to develop and evaluate the ability of feedback controllers that modulate electrical stimulation to achieve specified trajectories of joint angles and/or contact forces; and 3) the best of these controllers will be applied to the achievement of improved stair ascent and descent by paraplegic patients, through computer-controlled electrical stimulation.