基于力阻抗模型的上肢康复机器人交互控制系统设计

传统上肢康复机器人交互控制系统受到奇异位形影响,导致系统控制精准度较低,为此提出基于力阻抗模型的上肢康复机器人交互控制系统;设计上肢康复机器人交互控制系统结构,选取双串口12CSA60S2系列单片机作为下位机控制核心模块,利用椎齿轮改变驱动力方向,设计机械臂肘部结构,通过同步带传动,将器件隐藏于空手柄中;设计机械臂腕部结构,满足临床康复时上肢患者站姿与坐姿训练需求;选择箔式应变片BF350力传感器,设计电阻应变片桥接电路,处理传输信号;构建机器人目标阻抗模型,设计基于力阻抗控制策略,调节位置、速度和关节;为改善奇异位形情况,在奇异位形附近关节角速度指令直接由各个关节力矩阻尼控制得到,实现角速度精准输出,完成系统控制;由实验结果可知,该系统直线运动位置、旋转关节位置和伸缩关节位置跟踪结果与标准值基本一致,满足系统设计需求。     

基于阻抗模型的下肢康复机器人交互控制系统设计

为帮助下肢功能障碍患者进行康复训练,设计了下肢康复机器人。对于该机器人的控制,采用传统系统无法柔顺控制,导致机器人运动轨迹偏离预设轨迹。针对该现象,提出了基于阻抗模型的下肢康复机器人交互控制系统设计。通过分析总体控制方案,设计系统硬件结构框图。采用L型二维力传感器,确定两个方向的人机交互力。使用绝对值编码器安装在各个关节处,其输出值作为髋关节、膝关节、踝关节电机的转动位置,增量编码器安装在电机轴上,测量值用来作为后期控制方法的输入参数。构建阻抗控制模型,能够调节机器人位置和速度,具有消除力误差功能。依据此力矩对参考运动轨迹进行设计,实时获取患者康复训练的跟踪、主动柔顺和接近状态信息。在柔顺训练实验测出人机交互力,通过实验结果知,在检测到人体主动力矩异常时,系统能够重新优化轨迹,具有良好柔顺控制效果。     

康复机器人的同步主动交互控制与实现

提出了一种适用于康复机器人的人机交互控制方法. 结合一款具有平面并联结构的上肢康复机器人, 实现了与用户(患者)运动意图同步的、柔顺的主动康复训练. 在训练中, 利用自适应频率振荡器, 从表面肌电信号(Surface electromyography, sEMG)中获取运动模式信息, 然后结合运动模式和期望的正常运动轨迹, 生成与主动运动意图同步的参考训练轨迹. 本文通过仿真和实际实验对所提出的方法进行了验证, 振荡器可以在2~5s内快速实现与用户主动运动意图的同步, 然后利用阻抗控制器给予柔顺的辅助. 通过调节阻抗参数, 可以为患者的运动训练提供不同程度的辅助.     

下肢康复机器人的自适应人机交互控制策略

针对康复机器人运动过程中的人机交互性问题,提出一种下肢康复机器人自适应人机交互控制策略.提取伸屈运动中下肢表面肌电信号（Surface electromyography,sEMG）和足底压力特征,分别用于表征下肢运动意图和人机交互力（Interaction force,IF）信息,建立基于sEMG-IF的人机交互信息融合模型,实现下肢康复机器人运动轨迹的在线规划;考虑主动康复运动过程中的人机交互作用,建立具有时变动态特性的人机系统动力学模型,设计间接模糊自适应控制器对期望轨迹进行跟踪控制,实现下肢康复机器人自适应人机交互控制.通过对5名被试者进行下肢康复机器人运动控制实验研究,验证所提方法的可行性和有效性.     

绳驱动柔性上肢外骨骼机器人设计与控制

为了提高上肢外骨骼机器人的拟人化程度及关节柔性,设计了一种由串联弹性驱动器和鲍登线驱动的4自由度柔性上肢外骨骼机器人.首先,设计一种六连杆双平行四边形机构,建立肩关节虚拟转动中心,满足人体肩部3自由度运动需求.然后,设计基于串联弹性驱动器和鲍登线的驱动模块,将驱动器和机器人关节分离,降低结构的复杂度,减轻关节质量,实现力矩/位置信息的反馈.最后,构建机器人运动学及动力学模型,设计关节阻抗控制器并对样机肘关节进行阻抗控制实验.由实验结果可知,刚度系数在0.5 N·m/(°)～1.5 N·m/(°)时,力矩跟踪均方根为0.33 N·m;阻尼系数在0.001 N·m·s/(°)～0.01 N·m·s/(°)时,力矩跟踪均方根为0.57 N·m.实验结果表明,调节阻抗控制器中的阻抗系数能够改变关节的刚度和阻尼特性,从而提高人机连接的柔顺性.因此该机器人可以满足康复训练需求.     

一种基于肌电信号的自适应人机交互控制方法

针对踝关节康复机器人运动过程中的人机交互性问题, 本文提出一种基于肌电信号的鲁棒自适应人机交 互控制方法. 针对患者难以保持某一动作、肌电信号微弱等特点, 提出一种新的关节角度估计方法. 该方法充分利 用了踝关节运动时胫骨前肌与腓肠肌的拮抗关系, 将踝关节的动作类型与单个肌肉群的收缩进行关联, 利用归一化 的特征值完成运动意图的辨识和运动角度的估计. 为了保证人机交互的安全性, 提出一种刚度、阻尼参数在线自适 应调节的阻抗控制算法. 基于交互力矩对机器人末端的运动角度与运动速度实时进行调节, 使其对外表现出等效 柔性. 实验研究表明所提出的人机交互控制方法是有效的, 并具有一定应用前景.     

基于变阻抗补偿的上肢康复机器人人机交互方法研究

针对偏瘫患者在进行康复训练过程中不能很好地主动参入并且康复任务表现能力低的问题，提出了一种基于变阻抗补偿的上肢康复机器人人机交互方法。该方法可以提高康复过程中的柔顺性和任务表现。首先介绍了上肢康复机器人的基本结构，并对其进行运动学分析；再介绍了导纳控制和阻抗补偿控制系统框架；之后通过受试者的交互力和交互力变化速率建立出受试者的运动意图，在根据运动意图来实时调整阻抗补偿参数改变系统的柔顺性；最后通过三种不同的模式进行轨迹跟踪和到达目标点的实验比较，结果表明所提出的控制算法可以适应受试者的运动意图，提高任务表现。     

下肢康复外骨骼人机耦合动力学建模与控制

在行走过程中,患者与外骨骼之间的交互力,决定了患者使用外骨骼的安全性、舒适度以及外骨骼的性能,因而在康复外骨骼的相关研究中受到广泛的关注.本文以可穿戴式的下肢康复外骨骼为研究对象,考虑了患者和外骨骼的耦合作用,给出了人机交互力模型,同时考虑了足底-地面接触力的影响,从而基于第一类拉格朗日方程建立了人机耦合的动力学模型.进一步,考虑患者在不同康复训练阶段的特点和实际需求,给出了相应的控制策略.特别地,在主动康复阶段,从人机交互力模型的角度,给出了基于人机交互力的外骨骼位置阻抗控制中参考轨迹修正量的实际物理解释,并且设计了阻抗PID控制策略.仿真结果表明,使用人机耦合的动力学模型,能够对人机交互力进行有效的分析;进一步,本文在康复训练的主动阶段所设计的控制策略,能够有效地降低人机交互力的水平.     

基于模糊补偿的主从式上肢外骨骼康复机器人训练控制方法

针对主从式上肢外骨骼康复机器人主臂信息获取、从臂快速响应等问题,提出了基于关节位姿、速度和力/力矩等信息的运动意图建模方法及基于模糊补偿的康复训练控制策略.根据人体工程学原理,提出了一种同构同型的主从式双臂康复机器人新型结构;利用D-H算法给出了笛卡儿空间的主从臂运动学模型,建立了患者健肢运动意图信息和从臂各关节动作的人机协作映射关系;以患者运动意图力矩作为输入,基于模糊补偿算法提出了患者-主臂-从臂协作控制策略,并利用李亚普诺夫定理证明了该控制系统的稳定性.仿真结果表明,康复机器人从臂可以根据患者运动意图跟随主臂运动,能有效地防止抖动误动,可避免对患肢的二次伤害.实验结果表明从臂运动轨迹平滑,无剧烈波动,控制轨迹跟踪主臂效果好.     

基于人体运动意图卡尔曼预测的外骨骼机器人控制及实验

为在外骨骼控制中准确获取人体运动意图,本文使用力矩传感器测量人机交互信息.基于人体下肢摆动腿的单摆模型获得摆动腿关节的运动轨迹,并使用卡尔曼滤波进行预测,从而弥补意图延时.使用PD(比例-微分)控制律控制外骨骼跟踪人体摆动腿的关节轨迹,编码器反馈外骨骼关节的实时位置,形成位置闭环控制.进行外骨骼摆动腿实验,结果表明,测得的人机交互信息经过卡尔曼滤波后,可以预测人体摆动腿的运动意图,外骨骼机器人能够实现对人体摆动腿关节轨迹的跟随,所提方法可行.     

上肢康复外骨骼机器人的模糊滑模导纳控制

为了辅助上肢运动功能障碍患者进行不同模式的康复训练,基于上肢康复外骨骼机器人系统,提出了一种模糊滑模导纳控制策略,实现训练过程的人机协调控制.首先,介绍了康复外骨骼的整体结构和实时控制平台.然后,分析了模糊滑模导纳控制算法的推导过程,并根据李亚普诺夫稳定性判据证明系统的稳定性.最后,在不同系统导纳参数条件下,分别进行被动训练模式和主动训练模式实验,并对比分析了实验过程中运动偏差、人机交互力以及肱二头肌表面肌电信号的变化特点.实验结果表明,选择合适的目标导纳模型可以优化康复训练强度与难度,提高人机交互柔顺性与患者参与度,满足不同瘫痪程度和康复进度患者的训练需求.     

Voltage-based adaptive impedance force control for a lower-limb rehabilitation robot

This paper presents a novel voltage-based adaptive impedance force control for a lower limb rehabilitation robot. The impedance parameters are adaptively regulated by a gradient descent algorithm for adjusting the human force in performing therapeutic exercises. Although the proposed control is based on voltage control strategy, it differs from the common torque control strategy. One of the advantages is that it is free from the dynamical models of the robot and patient. Compared with a torque control scheme, it is simpler, less computational, and more efficient while it considers the actuators. The control approach is verified by stability analysis. Simulation results show the efficiency of the control approach applied on a lower limb rehabilitation robot driven by an electric motor. A comparison on performing isometric exercise shows that the voltage-based adaptive impedance force control is superior to both voltage-based impedance control and torque-based impedance control.     

Muscle circumference sensor and model reference-based adaptive impedance control for upper limb assist exoskeleton robot

In this paper, we have addressed two issues for upper limb assist exoskeleton: (1) estimation of human desired motion intention (DMI) using non-biological-based sensors; and (2) compliant control using model reference-based adaptive approach. For non-biological-based DMI estimation, we have employed Muscle Circumference Sensor (MCS) and load cells. MCS measures human elbow joint torque using human arm kinematics, biceps/triceps muscle model, and physiological cross-sectional area of these muscles. So, using MCS, we have measured Biceps/Triceps internal muscle activity and we have tried to reduce it by providing robotic assistance. To extract DMI, we have employed radial basis function neural network (RBFNN). RBFNN uses position, velocity, and human force to estimate DMI which is further tracked by the impedance control law. This algorithm is based on model reference-based adaptive impedance control law which drives the overall assist exoskeleton to the desired reference impedance model, giving required compliance. To highlight the effectiveness, we have compared proposed control algorithm with simple impedance and adaptive impedance control algorithms. Experimental results demonstrate the reduced muscle activity and active compliance for subject wearing the robot.     

Control System Design for an Upper-Limb Rehabilitation Robot

Control system implementation is one of the major difficulties in rehabilitation robot design. The purpose of our study is to present newly developed control strategies for an upper-limb rehabilitation robot. The Barrett WAM Arm manipulator is used as the main hardware platform for the functional recovery training of the past-stroke patient. Passive and active recovery training have been implemented on the WAM Arm. A fuzzy-based PD position control strategy is proposed for the passive recovery exercise to control the WAM Arm stably and smoothly to stretch the impaired limb to move along predefined trajectories. An adaptive impedance force controller is employed in the active motion mode in which a fuzzy logic regulator is used to adjust the desired impedance between the robot and impaired limb to generate adaptive force in agreement with the change of the impaired limb's muscle strength. In order to evaluate the change of the impaired limb's muscle power, the impaired limb's mechanical impedance parameters as an objective evaluation index is estimated online by using a recursive least-squares algorithm with an adaptive forgetting factor. Experimental results demonstrate the effectiveness and potential of the proposed control strategies.     

Active torque-based gait adjustment multi-level control strategy for lower limb patient–exoskeleton coupling system in rehabilitation training

This paper proposes an active torque-based gait adjustment multi-level control strategy for lower limb patient–exoskeleton coupling system (LLPECS) in rehabilitation training. The proposed controller has three levels of high, middle, and low sub-controllers: gait adjustment layer (high-level), interaction torque design layer (middle-level), and trajectory tracking layer (low-level). The high-level sub-controller uses an adaptive central pattern generator (ACPG) to adjust the desired gait for rehabilitation training according to the patient’s active torque. In the middle-level sub-controller, the desired interaction torque is designed with neural networks and the estimated muscle torque by utilizing nonlinear disturbance observer (NDO). In the low-level sub-controller, a time delay estimation-based prescribed performance model free control is designed for the accurate tracking performance of the exoskeleton, so as to make the actual interaction torque track the desired value. An exoskeleton virtual prototype, which is developed in SolidWorks, has been imported to MATLAB/Simulink to conduct co-simulations in the SimMechanics environment. The results of co-simulations demonstrate the effectiveness of the proposed control strategy when the patient’s muscle torque is at different recovery degrees.     

下肢软质康复外骨骼机器人的模糊神经网络阻抗控制

针对脑卒中或交通意外等因素导致的运动功能障碍问题,设计了一种可用于康复训练的可穿戴式的软质膝关节外骨骼机器人．在重点介绍基于Hill肌肉模型的套索人工肌肉驱动系统设计和实时控制平台的基础上,分析了模糊神经网络阻抗控制算法的推导过程．最后,分别在定阻抗与变阻抗参数控制策略条件下,进行人机协同训练模式下的康复训练实验,并对比分析了康复外骨骼系统对受试者肌肉活性的影响．实验结果表明,定频率定幅值训练时的屈/伸扭矩分别增加了9.70%和9.06%,而变频率变幅值训练时的屈/伸扭矩提升了88.34%和57.68%．由此可知,选择符合人体生理肌肉刚度特性的阻抗模型可以改善下肢康复机器人系统的稳定性和安全性,提高人机交互的柔顺性和协调性．     

Optimal impedance via model predictive control for robot-aided rehabilitation

This paper proposes an optimal impedance controller for robot-aided rehabilitation of walking, aiming to increase the patient’s activity during the therapy. In an online procedure, the joint torques produced by the patient during the gait is estimated using the generalized momenta-based disturbance observer and the Extended Kalman filter algorithm. At the same time, a model predictive control is performed to obtain the instantaneous optimal stiffness parameters of the robot’s impedance controller, trying to maximize the patient’s active participation by increasing his/her joint torques. In this feasibility study, experiments with a healthy subject, considering a modular lower limb exoskeleton and a set of user’s behaviors, are performed to evaluate the proposed controller. The results show the robot stiffness converges to a value which increases the user’s active participation.     

Model reference adaptive impedance control for physical human-robot interaction

This paper presents a novel enhanced human-robot interaction system based on model reference adaptive control. The presented method delivers guaranteed stability and task performance and has two control loops. A robot-specific inner loop, which is a neuroadaptive controller, learns the robot dynamics online and makes the robot respond like a prescribed impedance model. This loop uses no task information, including no prescribed trajectory. A task-specific outer loop takes into account the human operator dynamics and adapts the prescribed robot impedance model so that the combined human-robot system has desirable characteristics for task performance. This design is based on model reference adaptive control, but of a nonstandard form. The net result is a controller with both adaptive impedance characteristics and assistive inputs that augment the human operator to provide improved task performance of the human-robot team. Simulations verify the performance of the proposed controller in a repetitive point-to-point motion task. Actual experimental implementations on a PR2 robot further corroborate the effectiveness of the approach.