Learning an intermittent control strategy for postural balancing using an EMG-based human-computer interface.

Learning an intermittent control strategy for postural balancing using an EMG-based human-computer interface.
复制标题

DOI:
10.1371/journal.pone.0062956
复制
发表时间:
2013
期刊:
影响因子:
3.7
通讯作者:
Nomura T
Nomura T
中科院分区:
综合性期刊3区
文献类型:
--
作者:
Asai Y;Tateyama S;Nomura T

文献摘要

参考文献

被引文献

相似文献

人们一直认为,大脑通过调节拮抗剂肌肉的共同激活水平来稳定不稳定的身体动力学。在这里,我们使用一种新的基于肌电图的人机界面,在姿势平衡任务中批判性地重新检验了阻抗控制的既定理论,在该任务中,受试者被要求使用关于摆位置的视觉反馈信息来平衡一个虚拟倒立摆。该摆由一对拮抗关节力矩驱动,该拮抗关节力矩由受试者站立时对应的一对拮抗踝关节肌的激活实时确定。这种运动任务带来了令人沮丧的环境;感觉运动回路的反馈时滞大,作为不稳定性的来源,可能有利于采用非反应性的预编程阻抗控制,但踝关节肌肉相对难以协同激活,这阻碍了受试者采用阻抗控制。本研究旨在探索实验对象如何通过运动学习来解决这种令人沮丧的环境。三分之一的被试以阻抗式控制的方式适应平衡任务。然而,值得注意的是,大多数受试者没有采用阻抗控制。相反,他们获得了一种聪明而高效的策略,在一系列最佳时间内,两块肌肉同时失活,导致钟摆不主动驱动的间歇出现。肌肉失活和钟摆Çs摆动的特征表明,这些受试者采用的策略是一种间歇性控制,利用了在主动驱动关闭时钟摆状态空间中出现的鞍型不稳定直立平衡的稳定流形。
It has been considered that the brain stabilizes unstable body dynamics by regulating co-activation levels of antagonist muscles. Here we critically reexamined this established theory of impedance control in a postural balancing task using a novel EMG-based human-computer interface, in which subjects were asked to balance a virtual inverted pendulum using visual feedback information on the pendulum's position. The pendulum was actuated by a pair of antagonist joint torques determined in real-time by activations of the corresponding pair of antagonist ankle muscles of subjects standing upright. This motor-task raises a frustrated environment; a large feedback time delay in the sensorimotor loop, as a source of instability, might favor adopting the non-reactive, preprogrammed impedance control, but the ankle muscles are relatively hard to co-activate, which hinders subjects from adopting the impedance control. This study aimed at discovering how experimental subjects resolved this frustrated environment through motor learning. One third of subjects adapted to the balancing task in a way of the impedance-like control. It was remarkable, however, that the majority of subjects did not adopt the impedance control. Instead, they acquired a smart and energetically efficient strategy, in which two muscles were inactivated simultaneously at a sequence of optimal timings, leading to intermittent appearance of periods of time during which the pendulum was not actively actuated. Characterizations of muscle inactivations and the pendulum¡Çs sway showed that the strategy adopted by those subjects was a type of intermittent control that utilizes a stable manifold of saddle-type unstable upright equilibrium that appeared in the state space of the pendulum when the active actuation was turned off.
DOI: 10.1152/jn.1998.80.4.1787
发表时间: 1998-10-01
影响因子: 2.5
作者:
Doeringer, JA;Hogan, N
通讯作者: Hogan, N
DOI: 10.1152/jn.01123.2011
发表时间: 2012-07
影响因子: 2.5
作者:
Franklin S;Wolpert DM;Franklin DW
通讯作者: Franklin DW
DOI: 10.1007/s002219900286
发表时间: 2000-03-01
影响因子: 2
作者:
Foulkes, AJM;Miall, RC
通讯作者: Miall, RC
DOI: 10.1016/s0893-6080(09)80004-x
发表时间: 1993-01-01
期刊: NEURAL NETWORKS
影响因子: 7.8
作者:
GOMI, H;KAWATO, M
通讯作者: KAWATO, M
DOI: 10.1038/29528
发表时间: 1998-08-20
期刊: NATURE
影响因子: 64.8
作者:
Harris, CM;Wolpert, DM
通讯作者: Wolpert, DM