An Adaptive Wearable Parallel Robot for the Treatment of Ankle Injuries
An Adaptive Wearable Parallel Robot for the Treatment of Ankle Injuries
复制标题
DOI：
10.1109/tmech.2012.2219065
复制
发表时间：
2014-02-01
影响因子：
6.4
通讯作者：
Parsons, John G.
中科院分区：
文献类型：
--
作者：
Jamwal, Prashant K.;Xie, Sheng Q.;Parsons, John G.
本文提出了一种新的自适应可穿戴踝关节机器人的发展，通过物理康复治疗踝关节扭伤。踝关节机器人具有生物灵感的设计，在仔细研究现有踝关节机器人的改进机会后设计。机器人设计适应不同生理能力和年龄组的受试者。踝关节机器人采用气动肌肉驱动器（PMA），模仿骨骼肌的驱动方式。为了解决PMA的非线性特性，基于模糊的干扰观测器（FBDO）已经开发。基于Mamdani推理的自适应模糊逻辑控制器的另一个实例已经开发并附加了FBDO以补偿PMA的瞬态性质。该控制方案可以同时控制踝关节机器人的四个并联驱动器，实现三个转动自由度。为了评估机器人的设计，干扰观测器，和自适应模糊逻辑控制器，实验进行。踝关节机器人是由一个神经系统完整的受试者使用的。机器人与人的互动保持主动-被动，而机器人则在治疗师通常采用的预定义轨迹上操作。轨迹跟踪结果的报告中存在的不可预测的人类用户的干预，使用顺应性和非线性致动器，和并联运动结构的踝关节机器人。
This paper presents the development of a novel adaptive wearable ankle robot for the treatments of ankle sprain through physical rehabilitation. The ankle robot has a bioinspired design, devised after a careful study of the improvement opportunities in the existing ankle robots. Robot design is adaptable to subjects of varying physiological abilities and age groups. Ankle robot employs lightweight but powerful pneumatic muscle actuators (PMA) which mimics skeletal muscles in actuation. To address nonlinear characteristics of PMA, a fuzzy-based disturbance observer (FBDO) has been developed. Another instance of an adaptive fuzzy logic controller based on Mamdani inference has been developed and appended with the FBDO to compensate for the transient nature of the PMA. With the proposed control scheme, it is possible to simultaneously control four parallel actuators of the ankle robot and achieve three rotational degrees of freedom. To evaluate the robot design, the disturbance observer, and the adaptive fuzzy logic controller, experiments were performed. The ankle robot was used by a neurologically intact subject. The robot-human interaction was kept as active-passive while the robot was operated on predefined trajectories commonly adopted by the therapists. Trajectory tracking results are reported in the presence of an unpredicted human user intervention, use of compliant and nonlinear actuators, and parallel kinematic structure of the ankle robot.