Durfee W K, Palmer K I
Whitaker College, Massachusetts Institute of Technology, Cambridge 02139.
IEEE Trans Biomed Eng. 1994 Mar;41(3):205-16. doi: 10.1109/10.284939.
Designing advanced controllers for motor neural prosthesis applications requires appropriate models for electrically stimulated muscle. A nonlinear nonisometric muscle model based on a Hill-type structure is presented. Estimation algorithms were derived to parameterize the passive force-length, the passive force-velocity, the active force-length, and the active force-velocity properties, the isometric recruitment curve, and the linear contraction dynamics of the model. All parameters were based on experimental measurements rather than on values taken from the literature. The estimation methods were validated experimentally using isolated hind-limb muscles in two acute animal model preparations. The results demonstrated that the parameterized model is capable of predicting force output with reasonable accuracy for a wide range of simultaneously varying kinematic and stimulation inputs.
为运动神经假体应用设计先进的控制器需要适用于电刺激肌肉的模型。本文提出了一种基于希尔型结构的非线性非等长肌肉模型。推导了估计算法，用于对模型的被动力-长度、被动力-速度、主动力-长度、主动力-速度特性、等长募集曲线和线性收缩动力学进行参数化。所有参数均基于实验测量，而非取自文献中的值。在两种急性动物模型制备中，使用离体后肢肌肉对估计方法进行了实验验证。结果表明，对于广泛的同时变化的运动学和刺激输入，参数化模型能够以合理的精度预测力输出。
