Role of Trunk Rehabilitation on Trunk Control, Balance and Gait in Patients with Chronic Stroke: A Pre-Post Design

Purpose: Although proximal stability of the trunk is a prerequisite for balance and gait, to determine the role of trunk rehabilitation on trunk control, balance and gait in patients with chronic stroke is yet unknown. Method: Fifteen sub-jects (post-stroke duration (3.53 ± 2.98) years) who had the ability to walk 10 meters independently with or without a walking aid;scoring ≤ 21 on Trunk Impairment Scale (TIS), participated in a selective trunk muscle exercise regime, consisting of 45 minutes training per day, four days a week, and for four weeks duration in an outpatient stroke reha-bilitation centre. Results: The overall effect size index for trunk rehabilitation was 1.07. This study showed large effect size index for Trunk Impairment Scale (1.75), Berg Balance Scale (1.65) than for gait variables (0.65). After trunk rehabilitation, there was a significant improvement for gait speed (p= 0.015), cadence (p= 0.001) and gait symmetry (p=0.019) in patients with chronic stroke. In addition, all the spatial gait parameters had a significant change post-intervention. There was no significant change in temporal gait parameters with the exception of affected single limb support time. The level of significance was set at p < 0.05. Conclusion: The exercises consisted of selective trunk movement of the upper and the lower part of trunk had shown larger effect size index for trunk control and balance than for gait in patients with chronic stroke. Future randomized controlled studies incorporating large sample size would provide insight into the effectiveness and clinical relevance of this intervention.

Gait planning and intelligent control for a quadruped robot

We present a method for designing free gaits for a structurally symmetrical quadruped robot capable of performing statically stable, omnidirectional walking on irregular terrain. The robot＇s virtual model is constructed and a control algorithm is proposed by applying virtual components at some strategic locations. The deliberative-based controller can generate flexible sequences of leg transferences while maintaining walking speed, and choose optimum foothold for moving leg based on integration data of exteroceptive terrain profile. Simulation results are presented to show the gait＇s efficiency and system＇s stability in adapting to an uncertain terrain.

Control strategy of central pattern generator gait movement under condition of attention selection

As a typical rhythmic movement, human being＇s rhythmic gait movement can be generated by a central pattern generator （CPG） located in a spinal cord by self- oscillation. Some kinds of gait movements are caused by gait frequency and amplitude variances. As an important property of human being＇s motion vision, the attention selection mechanism plays a vital part in the regulation of gait movement. In this paper, the CPG model is amended under the condition of attention selection on the theoretical basis of Matsuoka neural oscillators. Regulation of attention selection signal for the CPG model parameters and structure is studied, which consequentially causes the frequency and amplitude changes of gait movement output. Further, the control strategy of the CPG model gait movement under the condition of attention selection is discussed, showing that the attention selection model can regulate the output model of CPG gait movement in three different ways. The realization of regulation on the gait movement frequency and amplitude shows a variety of regulation on the CPG gait movement made by attention selection and enriches the controllability of CPG gait movement, which demonstrates potential influence in engineering applications.

助行康复机器人自适应步态控制方法研究

为了提高助行康复机器人在不同坡度上的稳定性,合理的步态控制方法显得尤为重要。这里以助行康复机器人为研究对象,提出了一种将长短期记忆网络和动态运动基元结合的步行康复机器人自适应步态控制方法。将传感器采集信息与动态运动基元中步态控制项相结合,对参考步态进行调整。同时通过更改动态运动基元的时间项和空间项对步态进行二次调整以适应不同的坡度。通过试验对该控制方法的可用性进行验证。结果表明,该方法可以在(0~20)°斜坡进行自适应步态变换,具有一定的实用价值。

基于社区性训练的助行康复系统研究与实现

为了循序渐进地恢复脊髓损伤患者受损部分神经及下肢运动功能,在已有早期减重步行训练的基础上,增加减重模式下空间运动功能。构建可根据患者下肢运动功能损伤程度实现不同模式康复训练、状态监测和助行功能的系统,进行包括减重、助行装置、移动平台和控制模块的结构设计。集成平地行走和上下楼梯训练步态轨迹模型,通过对比正常髋、膝关节的运动角度数据,验证应用该轨迹的可行性与合理性,并进行了设定行走路线的社区任务导向性训练轨迹规划试验。研究结果表明,助行系统支持患者自主设定模拟社区任意行走模式,有助于帮助患者恢复部分肢体运动功能,以尽快融入社会。

基于改进涡流搜索算法的外骨骼迭代学习控制

为提升康复外骨骼机器人的步态跟踪性能,提出一种基于改进涡流搜索算法的迭代学习控制方法。首先针对传统迭代学习控制抗扰性差和控制信息缺失问题,引入PD控制器、自适应遗忘因子、误差过渡曲线和控制信息搜索等策略,改进迭代学习控制律;其次,基于多种策略对涡流搜索算法进行改进,提出了一种改进涡流搜索算法,改进后的算法可优化迭代学习控制的PD参数;最后进行行走实验,将提出的迭代学习控制方法与现有的同类算法进行仿真和数值比较,并测试了扰动情况下的跟踪性能。实验结果表明,所提方法的误差更小,跟踪性能更强。该算法改进了迭代学习控制的不足,具有较强的抗扰性能,保证了使用时的稳定性。

基于最大熵深度强化学习的双足机器人步态控制方法

针对双足机器人连续直线行走的步态稳定控制问题,提出一种基于最大熵深度强化学习(DRL)的柔性演员-评论家(SAC)步态控制方法。首先,该方法无需事先建立准确的机器人动力学模型,所有参数均来自关节角而无需额外的传感器;其次,采用余弦相似度方法对经验样本分类,优化经验回放机制;最后,根据知识和经验设计奖励函数,使双足机器人在直线行走训练过程中不断进行姿态调整,确保直线行走的鲁棒性。在Roboschool仿真环境中与其他先进深度强化学习算法,如近端策略优化(PPO)方法和信赖域策略优化(TRPO)方法的实验对比结果表明,所提方法不仅实现了双足机器人快速稳定的直线行走,而且鲁棒性更好。

基于姿态信息的双足移动机器人越障步态控制

在机器人运动控制中,姿态信息获取是不可缺少的步骤,尤其是在机器人越障步态控制中,因此提出基于姿态信息的双足移动机器人越障步态控制方法。在双腿摆动和地面碰撞2种情境下,创建双足移动机器人动力学模型,获得机器人行走变化规律。利用AdaBoost算法分类机器人姿态属性,通过关联性指标与骨骼密度实现机器人姿态信息感知,明确操作者越障意图。使用摄像头得到障碍物图像数据,采用神经网络的前馈补偿能力进行机器人越障步态控制。实验结果表明,所提方法越障步态控制精准度高、稳定性强,可提高双足移动机器人的移动能力。

仿生机器鱼步态控制及闭环运动控制方法综述

鱼类所具有的推进效率高、机动性强、环境扰动小等优点引发了国内外学者对仿生机器鱼的研究。底层步态控制方法和闭环运动控制方法是当前机器鱼控制研究的两大热点。按照推进模式的分类方法概述各类机器鱼的样机研制情况以及其性能优劣,介绍机器鱼的推进机理及其水动力学研究进展,进而重点探讨轨迹逼近方法和中枢模式发生器这2种底层步态控制思路,综述机器鱼的典型闭环运动控制方法。中枢模式发生器具有更强的灵活性、稳定性和可操作性,易于引入反馈项而实现闭环控制,在机器鱼底层步态控制中占主导地位。针对机器鱼的显著特点改进后的基于学习的控制方法与多种方法相结合后的混合控制方法具有更为广阔的发展前景,符合仿生机器鱼智能化的发展方向。根据工作条件和运动要求建立合理的步态控制系统以及准确高效的闭环运动控制系统可提高机器鱼的整体性能。

基于X86平台和RSI的工业机器人步态自动控制系统设计

工业环境中的任务和工作场景是动态变化的,机器人需要能够根据环境的变化调整步态,以满足新的任务需求;为此,设计基于X86平台和机器人传感器接口(RSI,robot sensor interface)的工业机器人步态自动控制系统;以复杂指令集计算机为基础的X86架构设计机器人控制器主板,使系统具有高集成度和扩展性;利用超声波传感器和红外线传感器获取步态自动控制传感信号;使用基于AS5040型高精密非接触磁性转动编码器的步态关节控制器,通过总线扩展,定位关节运动方向;分析机械臂前后摆动步态规划轨迹,控制髋关节;使用RSI应用程序包控制点位运动,实现步态自动控制;实验结果表明,设计系统的膝关节x方向与实际轨迹只存在最大为20 mm的误差,y方向与实际轨迹一致;髋关节x方向与实际轨迹只存在最大为20 mm的误差,y方向与实际轨迹只存在最大为15 mm的误差,能够提高控制精度,控制效果较好。

基于改进SSA优化PID的下肢康复机器人控制

针对下肢康复机器人训练过程中步态跟踪误差大的问题,提出一种改进麻雀搜索算法(ISSA)优化PID的控制策略。首先,利用拉格朗日方程对下肢机器人进行动力学分析,得到动力学模型;然后,通过引入Kent映射混沌初始化种群、Tent混沌扰动和柯西变异扰动,以及改变探索者-跟随者比例系数和探索者位置更新公式等方面对麻雀搜索算法进行改进,通过8种测试函数证实了算法改进的有效性;最后,在Matlab/Simulink中建立下肢机器人PID控制器和ISSA-PID控制器系统的仿真模型,以正常人体步态数据进行拟合得到的函数曲线作为系统实际输入进行跟踪实验。结果表明:相比于传统PID控制,ISSA-PID控制髋关节和膝关节的误差分别降低了63.3%和72.5%,实现了精准的步态跟踪,验证了该控制策略的有效性。

单侧前交叉韧带重建患者步行双侧下肢肌肉协同模式差异

目的探究单侧前交叉韧带重建患者步态活动时双侧下肢肌肉协同模式差异。方法2023年4月至6月,于武汉体育学院附属医院招募单侧前交叉韧带重建患者男性12例、女性8例,采集步行时双侧下肢肌肉的肌电,采用非负矩阵分解算法提取健、患侧腿肌肉协同元参与数量、肌肉协同元峰值激活用时、肌肉相对权重。结果男性健侧腿解析出6类肌肉协同元,患侧腿解析出5类肌肉协同元,缺乏以胫骨前肌为主要参与肌肉的协同元2;女性健、患侧腿则均解析出6类肌肉协同元。相比于健侧腿,男性参与患侧腿活动的各类肌肉协同元的峰值激活用时均无显著变化(P>0.05);女性的协同元3、协同元5的峰值激活用时增加(P<0.05)。男性患侧腿协同元1的股直肌相对权重降低(P<0.05);女性患侧腿协同元2的股外侧肌相对权重升高(P<0.05),股二头肌相对权重降低(P<0.05),协同元3的股直肌的相对权重降低(P<0.05),协同元6的股二头肌相对权重降低(P<0.05)。结论步行期间,男性患侧腿冻结主导踝关节背屈功能的肌肉协同元参与以增强踝关节稳定性,降低承重反应期股直肌的相对权重以弱化膝关节落地缓冲功能;女性主导患侧腿承重反应期、支撑相中期动作的协同元激活延迟,增强承重反应期股外侧肌的相对权重,降低承重反应期股直肌、股二头肌与支撑相中期股二头肌的相对权重,以限制膝关节屈曲活动。

Exercise-induced muscle fatigue in the unaffected knee joint and its influence on postural control and lower limb kinematics in stroke patients

This study aimed to investigate the effects of exercise-induced muscle fatigue in the unaffected knee joint on postural control and kinematic changes in stroke patients. Forty participants（20 stroke patients, 20 age-matched healthy participants） were recruited. To induce fatigue, maximum voluntary isometric contractions were performed in the unaffected knee joint in a Leg Extension Rehab exercise machine using the pneumatic resistance. We measured static and dynamic balance and lower-limb kinematics during gait. Changes in postural control parameters anteroposterior sway speed and total center of pressure distance differed significantly between the stroke and control groups. In addition, changes in gait kinematic parameters knee and ankle angles of initial contact differed significantly between stroke（paretic and non-paretic） and control groups. Muscle fatigue in the unaffected knee and ankle impaired postural control and debilitates kinematic movement of ipsilateral and contralateral lower limbs, and may place the fatigued stroke patients at greater risk for falls.

Dynamically Adapt to Uneven Terrain Walking Control for Humanoid Robot

Dynamically adapt to uneven ground locomotion is a crucial ability for humanoid robots utilized in human environments.However,because of the effect of current pattern generation method,adapting to unknown rough ground is limited.Moreover,to maintain large support region by four-point contact during the landing phase is usually a key problem.In order to solve these problems,a landing phase control and online pattern generation in three dimensional environments is proposed.On the basis of robot-environment non-planar interactive modes,a method of landing control based on optimal support region is put forward to realize stable four-point contact by flexible foot,and a controller is employed to adapt to the changes of ground without using prior knowledge.Furthermore,an adaptable foothold planning is put forward to the online pattern generation considering walking speed,uneven terrain,and the effect of lateral movement to the locomotion stability.Finally,the effectiveness of landing control and online pattern generation is demonstrated by dynamic simulations and real robot walking experiments on outdoor uneven ground.The results indicate that the robot kept its balance even though the ground is unknown and irregular.The proposed methods lay a foundation for studies of humanoid robots performing tasks in complex environments.

下肢机器人联合膝关节控制训练在脑卒中后下肢功能障碍患者中的应用效果

目的探讨下肢机器人联合膝关节控制训练在脑卒中后下肢功能障碍患者中的应用效果。方法选取2021年3月至2023年3月南京脑科医院收治的106例脑卒中后下肢功能障碍患者为研究对象,根据随机数字表法将其分为对照组和观察组,每组53例。对照组实施膝关节控制训练,观察组给予下肢机器人联合膝关节控制训练治疗。治疗3个月后,评估比较两组临床疗效、症状评分、步态参数、运动耐力。结果治疗后,观察组临床疗效优于对照组(P<0.05)。治疗后,两组下肢Fugl-Meyer运动功能量表、Berg平衡量表、Holden步行功能分级量表评分均高于治疗前,且观察组高于对照组(P<0.05)。治疗后,两组屈髋峰值、屈膝峰值高于治疗前,步幅长于治疗前,步频、步速快于治疗前,且观察组屈髋峰值、屈膝峰值高于对照组,步幅长于对照组,步频、步速快于对照组(P<0.05)。治疗后,两组峰值摄氧量(VO_(2)peak)、无氧阈值(AT)高于治疗前,6分钟步行试验(6 MWT)距离长于治疗前,且观察组VO_(2)peak、AT高于对照组,6 MWT距离长于对照组(P<0.05)。结论下肢机器人联合膝关节控制训练对脑卒中下肢功能障碍患者效果显著,能提升下肢运动功能,增强运动耐力、提高生活质量。

Walking Stability Control Method for Biped Robot on Uneven Ground Based on Deep Q-Network

A gait control method for a biped robot based on the deep Q-network (DQN) algorithm is proposed to enhance the stability of walking on uneven ground. This control strategy is an intelligent learning method of posture adjustment. A robot is taken as an agent and trained to walk steadily on an uneven surface with obstacles, using a simple reward function based on forward progress. The reward-punishment (RP) mechanism of the DQN algorithm is established after obtaining the offline gait which was generated in advance foot trajectory planning. Instead of implementing a complex dynamic model, the proposed method enables the biped robot to learn to adjust its posture on the uneven ground and ensures walking stability. The performance and effectiveness of the proposed algorithm was validated in the V-REP simulation environment. The results demonstrate that the biped robot's lateral tile angle is less than 3° after implementing the proposed method and the walking stability is obviously improved.

Gait simulation of new robot for human walking on sand

In order to simulate the gait of human walking on different terrains a new robot with six degrees of freedom was proposed. Based on sand bearing characteristic compliance control was introduced to control system in horizontal and vertical movement directions at the end of the robot,and position control in attitude. With Matlab/Simulink toolbox,the system control models were established,and the bearing characteristics of rigid ground,hard sand,soft sand and softer sand were simulated. The results show that 0,0.62,0.89 and 1.12 mm are the maximal subsidences of the four kinds of ground along the positive direction of x-axis,respectively,and 0,-0.96,-1.99 and -3.00 mm are the maximal subsidences along the negative direction of x-axis,respectively. Every subsidence along y-axis is negative,and 0,-4.12,-8.23 and -12.01 mm are the maximal subsidences of the four kinds of ground,respectively. Simulation results show that the subsidence of footboard points to inferior anterior in early stage of stand phase,while points to posterior aspect in late stage. The subsidence tends to point to posterior aspect in the whole. These results are basically consistent with the gait characteristics of human walking on sand. Gait simulation of the robot for human walking on sand is achieved.

新型四足并联军用机器人步态控制算法及仿真

在多关节步行机器人控制策略中,全部采用中央模式(CPG)网络进行控制的方法具有参数繁多、网络结构复杂的特点;机器人的工作环境通常多变且复杂,对机器人的灵活性和抗干扰能力提出了更高的要求,故仅采用单一的控制模式很难满足上述需求。针对上述问题,在机器人控制上把基于CPG的控制方法和基于虚拟模型的控制方法进行综合,对单腿为3-UPS机构的四足并联军用机器人设计了一种新型步态控制算法,用CPG完成机器人的基础步态,完成输入输出之间的非线性振荡器网络模型的搭建,并将模型的输出与关节电机的驱动力矩构成映射关系;再用虚拟模型生成行走时保持机器人平稳姿态所需要的足端虚拟力,并将足端虚拟力映射为关节驱动力矩。通过V-REP与MATLAB软件对该步态控制算法进行联合仿真实验。仿真结果表明:所提出的步态控制算法有效;新算法的优势在于简化控制网络的同时还能保证机器人在行走过程中拥有较强的灵活性和抗干扰能力,这种新型控制模式为四足并联军用机器人的步态控制提供了新的思路与方法。

基于改进强化学习的机器人双足步态控制方法

为提高双足机器人应对不同场地步态控制的自主性与稳定性,提出了一种基于改进强化学习的机器人双足步态控制方法。在构建机器人双足步态控制强化学习框架的基础上,对确定性策略梯度进行修正,基于Actor-Critic结构将网络训练样本一分为二,分别对行为价值进行独立估计,通过使用多层神经网络参数化策略并应用确定性策略梯度学习,从而学习到具有较强鲁棒性的步态控制策略。仿真实验表明:通过本文方法能够使机器人臀部、大腿、关节、小腿以及指向获得更为稳定的控制策略,步态收敛更迅速、步态收敛域更大,实现了不同速度下不同场地的稳健步行运动。

一种智能膝关节假肢及其步态对称性评价

目的提出一种评价智能膝关节假肢步态对称性的方法。方法设计智能膝关节假肢样机AiKneeOne以及可穿戴式步态数据采集系统。以步态周期中第一次双脚支撑时间、单脚支撑时间、第二次双脚支撑时间和摆动相时间作为参数,计算对称指数(SI)和对称指标比率RⅠ、RⅡ评估步态对称性指标。5例健康人模拟穿戴AiKneeOne,以0.5、0.7、1.1 m/s速度在平板上行走,采用可穿戴式下肢步态数据采集系统采集数据,进行实测。结果不同速度下,第一次双脚支撑和第二次双脚支撑时,SI和RⅡ绝对值较低、RⅠ接近1,对称性较好;单脚支撑和摆动相时,对称性有待提高。结论智能膝关节假肢AiKneeOne具备重建截肢者步态的潜力,步态对称性指标可用于智能膝关节假肢穿戴性能评估测试。