外骨骼助力负重中下肢肌肉收缩与协作机制

为了评估助力外骨骼人机交互性能，探索更优的外骨骼人机结构和人机协调控制策略，通过实验评估了9名健康男性在常规负重和外骨骼助力负重行走中，下肢肌肉平均活动、肌肉成对收缩指标和肌肉整体协调机制的特征与变化. 实验结果表明:外骨骼助力增加了肌肉平均活动，尤其是踝关节跖屈肌腓骨长肌，显示了当前外骨骼在踝关节/脚部位的人机界面设计存在缺陷；降低了成对肌肉的平均收缩指标，增大了肌肉收缩范围；肌肉模块化协作复杂度相似，解释方差水平值分别为95%和96%，但常规负重下的模块仅能描述外骨骼助力时83%～91%的肌电变量，这表明人体在外骨骼助力下不会单一地依赖中心控制和模块控制来协调肌肉活动，而是采取灵活且有规律的神经肌肉调节机制；肌间协作权重和肌肉激活尺度系数呈完全不同的形态，采取了与常规步态下完全不同的控制策略；进一步设计踝关节/脚人机交互界面，参考助力下神经肌肉调节机制规律来设计人机交互策略，将提高外骨骼人机可用性和人机交互性能. 

Lower Limb Muscle Co-Contraction and Coupling Synergy in Exoskeleton Assistance for Load Carriage Walking

Human-robot interaction performance was evaluated, and better exoskeleton structures and human-robot adaptability control strategies were explored. The muscle activity, paired-muscle contraction index, and synergy coupling of the muscles of nine healthy male subjects carrying loads and walking with and without the exoskeleton prototype were investigated through evaluation of the conducted experiment. The results show that muscle activity significantly increased compared with normal load carrying, especially for the planter flexor muscle peronaeus longus. Some design defects were observed in the foot/ankle area of the current exoskeleton prototype, and the paired-muscle mean co-contraction index decreased. However, the co-contraction index range was greater than in normal load carriage. The module complexities were similar for both load carriage conditions, with total variances of 95% and 96%. Three modules could describe nearly 95% of the variance in electromyography data during normal load carriage. However, the same modules could only describe 83%–91% of the variance in muscle activity while walking with the exoskeleton. These results show that the muscle activities in healthy subjects will not be modulated by the central control strategy in the process of adaption to the assistance of the exoskeleton. The plot shape of the synergy weights and the muscle activation are significantly different from the normal gait. Hence, better human-robot interface designs in the ankle/foot area and human- robot interaction strategies referenced from the kinds of flexible but principle neuromuscular recruitment mechanisms, would improve the usability and the human-robot interaction performance of the exoskeleton.