GA-BP神经网络在下肢运动步态识别中的应用研究

为了提高下肢表面肌电信号步态识别的准确性,提出了一种基于遗传算法（GA）优化的BP神经网络分类器设计方法。首先,对采集的下肢表面肌电信号进行小波滤波及特征值提取,其次,构造基于GA优化的BP神经网络分类器,然后,以提取的表面肌电信号特征作为输入对分类器进行训练,最后利用训练好的分类器进行测试。实验结果表明,基于GA优化的BP神经网络分类器能成功识别下肢正常行走的五个步态,平均识别率达到98%以上,效果明显优于BP神经网络分类器的识别效果。     

Evolutionary Robotics: From Algorithms to Implementations (Wang, L. et al.; 2006) [Book Review]

This book concentrates on the application of soft computing - evolutionary algorithms, neural networks, and fuzzy logic - for intelligent sensing, learning, and robot navigation. It consists of nine chapters which cover such topics as: evolvable hardware in robots; autonomous robot navigation systems; intelligent sensor fusion and learning; task-oriented developmental learning; bipedal walking through reinforcement learning; and swing time generation for bipedal walking. The book is well written, readable, and is concise yet comprehensive. It is more suited as a convenient reference book for researchers and practicing professionals, rather than a classroom textbook. If worked examples and end-of-chapter problems were added, the book would easily serve as a text for a graduate course on Intelligent Walking Robots.     

一种下肢康复机器人步态的生成与调节方法

针对下肢康复外骨骼机器人康复训练参考步态的标准化设计,对平地行走、上楼梯和下楼梯等不同情境下的步态轨迹的设计问题进行研究,提出了由无线惯性传感器采集人体步态,利用基于关键点的步态生成与调节方法,得到了融合过渡过程的参考步态曲线.利用得到的参考步态曲线指导下肢康复外骨骼式机器人辅助人体在平地行走、上楼梯和下楼梯等情景下进...     

Initial System Evaluation of an Overground Rehabilitation Gait Training Robot (NaTUre-gaits)

For decades, robotic devices have been suggested to enhance motor recovery by replicating clinical manual-assisted training. This paper presents an overground gait rehabilitation robot, which consists of a pair of robotic orthoses, the connected pelvic arm in parallel and a mounted mobile platform. The overground walking incorporates pelvic control together with active joints on the lower limb. As a preliminary evaluation, system trials have been conducted on healthy subjects and a spinal cord injury (SCI) subject, respectively. Electromyography signals were recorded from muscles of the lower limb for each subject. Three experiments were carried out: (i) health volunteers walking at self-preferred walking speed, (ii) a SCI subject walking with the help of three helpers and (iii) the same SCI subject walking with the assistance provided by the gait device. In the experiment, the muscle activation of overground walking was compared between the manual-assisted and robotic-assisted methods. The initial results show that the performance of the device can provide impact-less overground walking and it is comparable to the performance obtained by manual assistance in gait rehabilitation training.     

基于图像处理的人体步态信息采集与处理

为了获取健康人体的正常步态信息, 提出了一种快捷有效的获取方法. 通过在下肢关节点处粘贴标记点, 利用摄像机获取正常人行走的图像, 对图像进行二值化处理, 提取出标记点坐标. 经过最小二乘拟合分析可得到人体脚心在一个步态周期内的运动轨迹及运动速度. 最后对下肢康复机器人进行步态规划, 得到下肢康复机器人的步态轨迹及其速度,并对不同年龄人群的步态速度曲线进行了分析. 实验结果表明, 该系统可行性好, 工作稳定, 为下肢康复机器人的运动学分析与控制提供有力的理论依据和验证方法.     

利用多源运动信息的下肢假肢多模式多步态识别研究

运动状态识别对智能下肢假肢的控制非常关键,本文利用下肢表面肌电信号、腿部角度和足底压力信号在运动模式和步态分析中的优势和特点,对下肢假肢的多模式多步态识别进行研究.通过建立下肢运动信息系统,获取下肢多源运动信息.先提取下肢肌电信号的小波包能量作为特征,建立多个HMM对下肢假肢的运动模式进行识别;再根据大小腿和膝关节的角...     

基于多传感器数据融合的智能小车避障的研究

针对智能小车避障问题,提出了一种将模糊逻辑和神经网络相结合的融合方法—Takagi-Sugeno(T-S)模糊神经网络方法。基于此方法的数据融合算法应用在智能小车避障运动中,采用多只超声波传感器和红外线传感器探测障碍物的距离和方向,采集的各种数据利用T-S模糊神经网络进行融合。通过实验仿真表明:此方法能够使智能小车对障碍物的灵活避障和导航行进。     

基于步态检测算法的辅助行走系统设计

为了保证盲人和视障患者安全方便出行和减少误报警,设计一种基于步态检测算法的辅助行走系统。信号处理算法上利用小波包对信号进行降噪,提取三轴加速度的均值、标准差、方差与小波能量谱,并结合足底压力信息组成多维参数作为步态特征,选取基于遗传算法优化支持向量机进行步态识别。硬件部分主要包括三轴加速度信号和足底压力信号采集硬件设计。软件部分主要包括对足底压力信号和三轴加速度信号的采集和处理以及一款能够自动拨号安卓手机APP。实验结果表明,该系统基于多传感器融合的步态检测技术,步态检测识别率平均达90.48%。该系统便携性好、功耗低、测障效果好,在辅助行走领域具有一定的研究意义和实用价值。     

Electromyography-based control of active above-knee prostheses

This paper presents a new electromyography (EMG)-based control approach for above-knee (AK) prostheses, which enables the user to control the prosthesis motion directly with his or her muscle activating neural signals. Furthermore, the unique ‘active-reactive’ control structure mimics the actuation mechanism of a human biological joint, and thus provides the user an experience similar to that of a biological lower limb in the control process. In the proposed control approach, surface EMG is utilized to provide a non-intrusive interface to the user's central nervous system, through which the muscle-activating signals can be obtained. With the EMG signals as inputs, an ‘active-reactive’ control algorithm is developed based on the analysis on a simplified musculoskeletal structure of human biological joint. This control algorithm incorporates an ‘active’ component, which reflects the user's active effort to actuate the joint, and a ‘reactive’ component, which models the reaction of the joint to the motion as a result of the controllable impedance displayed on the joint. With this unique structure, the controller enables the active control of the joint motion, while at the same time achieves a natural interaction with the environment through the modulation of the joint impedance. The effectiveness of the proposed control approach was demonstrated through a set of free swing experiments, in which the user was able to control the prosthesis to follow arbitrary motion commands, and a set of level walking experiments, in which the user achieved natural walking gait similar to the typical walking gait of healthy subjects.     

Multi-channel Fusion Based Adaptive Gait Trajectory Generation Using Wearable Sensors

This paper presents a method to regenerate lower limb joint angle trajectories during gait cycle by judging human intention using wearable sensor system. Myoelectric signals from user are used to detect the intention of gait initiation and gait phases. Multi-channel redundant fusion technique is implemented to obtain a robust stride time and gait phase calculation algorithm. Joint trajectories corresponding to particular gait events and phases are regenerated using a Radial basis neural network. The network is trained with joint angle data measured by Inertial Measurement Unit (IMU) from users with varying anthropomorphic features. Generated trajectory is adaptive to anthropomorphic as well as gait velocity variation. Contribution of this paper is in development of a wearable sensor system, multi-channel redundant fusion to calculate stride time and an adaptive gait trajectory generation algorithm. The proposed method of trajectory generation is used to regenerate lower limb joint motion in sagittal plane for wearable robotic devices like prosthesis and active lower limb exoskeleton.     

车路协同路侧感知融合方法的研究

随着智能网联汽车技术和产业的不断发展,智能网联汽车逐渐成为人们交通出行的选择之一。但受智能网联汽车自身环境感知系统对特定道路交通场景信息处理的局限,无法实现在所有行驶工况下安全高效的运行,其需车路协同路侧感知技术的辅助方能更安全高效的运行。海量的车路协同感知数据是城市道路和高速公路车路协同、运行分析和科学管理的宝藏,理解和分析这些数据是车路协同路侧感知融合的关键。面对车路协同路侧多传感器的不同数据,如何高效准确地挖掘和提取雷达、视频在不同时间、不同空间维度的数据,实现对重点交通场景(如视野盲区、急转弯道、隧道、桥梁)和交通事件、环境、设施安全等的雷达、视频数据进行快速融合检测、识别与检索,通过蜂窝车联网C-V2X网络在一定时延范围内有效地将路侧感知融合结果数据发送给智能网联汽车,确保其安全高效的行驶,是面向智能网联汽车辅助驾驶的车路协同路侧感知融合的关键问题。基于智能网联汽车其自身环境感知能力,对道路智能基础设施感知网络中的多传感器融合方法进行研究分析,提出了基于误差方差的多传感器融合算法,与非智能道路相比,其效率更高,更加智能化,可有效解决道路交通运行环境中存在的常见问题,为人们提供更加安全、高效、优质的交通出行服务。     

A differential evolution based neural network approach to nonlinear system identification

This paper addresses the effectiveness of soft computing approaches such as evolutionary computation (EC) and neural network (NN) to system identification of nonlinear systems. In this work, two evolutionary computing approaches namely differential evolution (DE) and opposition based differential evolution (ODE) combined with Levenberg Marquardt algorithm have been considered for training the feed-forward neural network applied for nonlinear system identification. Results obtained envisage that the proposed combined opposition based differential evolution neural network (ODE-NN) approach to identification of nonlinear system exhibits better model identification accuracy compared to differential evolution neural network (DE-NN) approach. The above method is finally tested on a one degree of freedom (1DOF) highly nonlinear twin rotor multi-input–multi-output system (TRMS) to verify the identification performance.     

Classification of EMG signals using combined features and soft computing techniques

The motor unit action potentials (MUPs) in an electromyographic (EMG) signal provide a significant source of information for the assessment of neuromuscular disorders. Since recently there were different types of developments in computer-aided EMG equipment, different methodologies in the time domain and frequency domain has been followed for quantitative analysis of EMG signals. In this study, the usefulness of the different feature extraction methods for describing MUP morphology is investigated. Besides, soft computing techniques were presented for the classification of intramuscular EMG signals. The proposed method automatically classifies the EMG signals into normal, neurogenic or myopathic. Also, multilayer perceptron neural networks (MLPNN), dynamic fuzzy neural network (DFNN) and adaptive neuro-fuzzy inference system (ANFIS) based classifiers were compared in relation to their accuracy in the classification of EMG signals. Concerning the impacts of features on the EMG signal classification, different results were obtained through analysis of the soft computing techniques. The comparative analysis suggests that the ANFIS modelling is superior to the DFNN and MLPNN in at least three points: slightly higher recognition rate; insensitivity to overtraining; and consistent outputs demonstrating higher reliability.     

Fusion of soft and hard computing: multi-dimensional categorization of computationally intelligent hybrid systems

The concept of fusion of soft computing and hard computing has rapidly gained importance over the last few years. Soft computing is known as a complementary set of techniques such as neural networks, fuzzy systems, or evolutionary computation which are able to deal with uncertainty, partial truth, and imprecision. Hard computing, i.e., the huge set of traditional techniques, is usually seen as the antipode of soft computing. Fusion of soft and hard computing techniques aims at exploiting the particular advantages of both realms. This article introduces a multi-dimensional categorization scheme for fusion techniques and applies it by analyzing several fusion techniques where the soft computing part is realized by a neural network. The categorization scheme facilitates the discussion of advantages or drawbacks of certain fusion approaches, thus supporting the development of novel fusion techniques and applications.     

Design of robust knowledge bases of fuzzy controllers for intelligent control of substantially nonlinear dynamic systems: II. A soft computing optimizer and robustness of intelligent control systems

The structure of intelligent control system (ICS) is analyzed, and the interrelations with conventional problems of the theory and practice of application of control systems are described. The analysis of the results of simulation of typical structures of intelligent control systems has allowed us to establish the following fact. The application of the technique of designing (presented in Part I), which is based on a fuzzy neural network (FNN), does not guarantee in general that the required accuracy of approximation of the training signal (TS) will be reached. As a result, under an essential change of external conditions, the sensitivity level of the controlled plant (CP) increases, which, on the whole, leads to a decrease in the robustness of the intelligent control system, and, as a consequence, to a loss of reliability (accuracy) of achieving the control goal. To eliminate the specified drawback of the neural network, a soft computing optimizer (SCO), which uses the technique of soft computing and allows one to eliminate the drawback, is applied, which results in an increase in the robustness level of the structure of the intelligent control system. The structure of the soft computing optimizer, which contains as a particular case the required configuration of an optimal fuzzy neural network, is considered. The main specific features of the functional operation of the soft computing optimizer and the stages of the process of designing robust knowledge bases (KB) of fuzzy controllers (FC) are described. The methodology of joint stochastic and fuzzy simulation of automatic control system based on the developed tool of the soft computing optimizer is discussed in order to test the robustness and to estimate the limiting structural capabilities of intelligent control systems. The efficiency of the control processes with application of the soft computing optimizer is demonstrated by particular typical examples (benchmarks) of models of dynamic controlled plants under the conditions of incomplete information about the parameters of the structure of the controlled plant and under the presence of unpredicted (abnormal) control situations. Examples of industrial application of robust intelligent control systems in actual control systems designed based on the soft computing optimizer are presented. Practical recommendations for improving the robustness level of intelligent control systems by using new types of computations and simulation are given     

A comparative investigation of non-linear activation functions in neural controllers for search-based game AI engineering

The creation of intelligent video game controllers has recently become one of the greatest challenges in game artificial intelligence research, and it is arguably one of the fastest-growing areas in game design and development. The learning process, a very important feature of intelligent methods, is the result of an intelligent game controller to determine and control the game objects behaviors’ or actions autonomously. Our approach is to use a more efficient learning model in the form of artificial neural networks for training the controllers. We propose a Hill-Climbing Neural Network (HillClimbNet) that controls the movement of the Ms. Pac-man agent to travel around the maze, gobble all of the pills and escape from the ghosts in the maze. HillClimbNet combines the hill-climbing strategy with a simple, feed-forward artificial neural network architecture. The aim of this study is to analyze the performance of various activation functions for the purpose of generating neural-based controllers to play a video game. Each non-linear activation function is applied identically for all the nodes in the network, namely log-sigmoid, logarithmic, hyperbolic tangent-sigmoid and Gaussian. In general, the results shows an optimum configuration is achieved by using log-sigmoid, while Gaussian is the worst activation function.     

The Development of a Virtual Myoelectric Prosthesis Controlled by an EMG Pattern Recognition System Based on Neural Networks

One of the major difficulties faced by those who are fitted with prosthetic devices is the great mental effort needed during the first stages of training. When working with myoelectric prosthesis, that effort increases dramatically. In this sense, the authors decided to devise a mechanism to help patients during the learning stages, without actually having to wear the prosthesis. The system is based on a real hardware and software for detecting and processing electromyografic (EMG) signals. The association of autoregressive (AR) models and a neural network is used for EMG pattern discrimination. The outputs of the neural network are then used to control the movements of a virtual prosthesis that mimics what the real limb should be doing. This strategy resulted in rates of success of 100% when discriminating EMG signals collected from the upper arm muscle groups. The results show a very easy-to-use system that can greatly reduce the duration of the training stages.     

Quasi-Newton BP神经网络算法在电梯急停故障诊断中的应用

为了判断电梯运行是否故障急停,结合电梯动态智能检测系统的结构和特点,采用基于多传感器数据融合技术的3层BP神经网络方法,并将其应用到电梯动态智能检测系统中;因对不同传感器采集的信号采用不同的处理方法,训练样本包括基于小波包分析的能量特征向量,峭度系数、峰峰值时域特征值;Quasi-Newton BP算法经104步完成对样本训练,精度是2.6x10-4,实现检测系统的智能化急停诊断;结果表明该算法优于弹性BP算法.     

基于足底压力和关节运动信息的步态检测系统设计

作为老年人高发疾病,脑卒中有着较高的几率造成患者行动障碍。及时检测、获取偏瘫患者行走时步态信息,是治疗师为患者制定康复计划的重要环节。目前,较新的步态检测系统大多采用三维分析,三维步态分析系统体积较大、成本高、操作不便,在患者主动康复训练中存在较大的局限。针对现今步态检测系统的局限,设计了一种结合无线传输技术,检测患者关节运动角度及足底压力的运动信息检测系统。系统采用薄膜压力传感器、惯性传感器分别采集足底压力及关节运动信息,实现了对患者的8路足底压力信息及下肢关节运动角度信息的快速、有效采集。实验表明,系统可以有效采集关节运动角度及足底压力信息,在未来的康复训练中有着很好的应用前景。