Experience, Context, and the Visual Perception of Human Movement.

Why are human observers particularly sensitive to human movement? Seven experiments examined the roles of visual experience and motor processes in human movement perception by comparing visual sensitivities to point-light displays of familiar, unusual, and impossible gaits across gait-speed and identity discrimination tasks. In both tasks, visual sensitivity to physically possible gaits was superior to visual sensitivity to physically impossible gaits, supporting perception-action coupling theories of human movement perception. Visual experience influenced walker-identity perception but not gait-speed discrimination. Thus, both motor experience and visual experience define visual sensitivity to human movement. An ecological perspective can be used to define the conditions necessary for experience-dependent sensitivity to human movement. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Generation of adaptive gait patterns for quadruped robot with CPG network including motor dynamic model

This paper describes the generation of adaptive gait patterns using new Central Pattern Generators (CPGs) including motor dynamic models for a quadruped robot under various environments. The CPGs act as the flexible oscillators of the joints and adjust joint angles to required values. The CPGs are interconnected with each other and sets of their coupling parameters are adjusted by a genetic algorithm so that the quadruped robot can realize stable and adequate gait patterns. Generation of gait patterns results in the formation of the CPG networks suitable for the formation of not only a straight walking pattern but also of rotating gait patterns. Experimental results demonstrate that the proposed CPG networks are effective for the automatic adjustment of the adaptive gait patterns for the tested quadruped robot under various environments. Furthermore, the target tracking control based on image processing is achieved by combining the general gait patterns. © 2006 Wiley Periodicals, Inc. Electr Eng Jpn, 155(1): 35–43, 2006; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20225     

基于特征融合的步态识别算法研究

根据人体步态变化特点,提出一种基于特征融合和神经网络的步态识别算法。首先采用时域差分法对运动人体轮廓进行分割,然后分别提取空间特征和频率特征,将两步态特征融合在一起,从而实现步态的分类和识别。在CASIA步态数据库上进行仿真实验,仿真结果表明,该方法不仅克服了单一特征提取方法存在的缺陷,同时提高了步态识别正确率。     

管道内壁四足爬壁机器人的运动学与步态规划

研究用于检测气体绝缘金属封闭开关（GIS）内部的负压吸附管道内壁四足爬壁机器人. 分别对机器人的腿部和机身进行运动学分析,采用改进的牛顿迭代法解决机身正运动学求解困难的问题. 对机器人沿管道轴向和圆周方向的爬壁运动进行步态规划,提出运动过程零冲击的轨迹规划方法. 使用Adams进行运动仿真,并在四足爬壁机器人样机上进行水平和垂直管道的全方位爬壁实验. 结果表明：机器人的运动轨迹与所规划的步态一致,运动过程中速度与加速度无突变,运动平稳,无明显冲击,运动学模型的正确性和所规划步态的合理性得到验证. 在GIS管道的实际检测应用中,实现机器人在不同工况下的平稳爬壁运动与检测.     

基于约束满足的多目标双足机器人动态步态参数优化*

目前的步态优化算法仅仅实现了对单一目标的优化,把双足机器人步态优化看做是多目标优化问题,构建了衡量稳定性、能量消耗、步行速度三个目标评价函数。考虑到直接对多个目标加权求和的方法不能很好地处理多目标问题,提出一种新的基于约束满足的多目标步态参数优化算法,其思想是把基于惩罚函数的SPEA2(strength Pareto evolutionary algorithm2 )应用到多目标双足机器人动态步态参数优化问题上,规划出了同时满足这三个目标的动态优化步态。通过仿真实验表明了算法的有效性。     

Online gait programming of humanoid walking via NMPC

In order to satisfy the requirement of realtime gait programming of humanoid walking with foot rotation,a kind of modified Nonlinear Model Predictive Control (NMPC) scheme was proposed. Based on setting suitable kinetic and kinematic virtual constraints of Single Support Phase (SSP) and three subphases of Double Support Phase (DSP) ,complex realtime gait programming problem was simplified to four online NMPC dynamic optimization problems. A numerical approach was proposed to transform the dynamical optimization problem to the finite dimensional static optimization problem which can be solved by Sequential Quadratic Programming (SQP) . It can be concluded from simulation that using this method on BIP model can realize online gait programming of dynamic walking with foot rotation and the biped stability can be satisfied such that there is no sliding during walking.     

基于灰度标准差的光流模型的步态识别算法

步态识别主要是通过人体走路的姿势来识别人的身份.提出了一种改进的光流法和隐马可夫模型相结合的技术,使用背景减除和阈值化为目标和背景图像建立模型,由使用灰度标准差筛选的光流法模型提取每一个运动目标的特征,取得时间序列的特征向量,把特征量化和状态分类的结果交给隐马可夫模型来判断运动目标的姿态.仿真实验表明,本算法识别率高,速度快.     

一种基于步幅长度及频域特征的步态识别方法

步态识别作为一种较新的生物认证技术有其独特优势,它旨在根据人行进过程中的行走步态模式来识别其身份.提出一种基于步幅长度及频域特征的步态识别方法:首先采用背景减除算法实现运动人体检测,并根据步幅长度变化特征进行周期分割.然后提取步幅长度极值结合关键帧的频幅均值构成五维特征向量,最后用标准模式分类器在中科院自动化所提供的CASIA步态数据库中进行实验,在降低了算法复杂度的同时,获得了令人鼓舞的实验结果.     

煤矿入井人员唯一性检测技术研究

通过分析掌纹、指纹、虹膜、人脸、步态、声纹等生物特征识别技术的特点以及煤矿现场对入井人员生物特征的影响,指出虹膜识别、人脸识别、步态识别、声纹识别适用于煤矿入井人员唯一性检测;提出了一种基于人员定位和生物特征识别的煤矿入井人员唯一性检测技术方案,将生物特征识别技术嵌入人员定位系统,利用人员定位识别卡实现识别卡数量及人员身份的唯一性检测;指出煤矿入井人员唯一性检测技术的研究关键点是严重污染人脸的识别算法、对设备遮挡情况下人员步态图像的采集及对混入人员语音信号的煤矿现场噪声消除算法。     

生物特征识别学科发展报告

从手机解锁、小区门禁到餐厅吃饭、超市收银,再到高铁进站、机场安检以及医院看病,人脸、虹膜和指纹等生物特征已成为人们进入万物互联世界的数字身份证。生物特征识别赋予机器自动探测、捕获、处理、分析和识别数字化生理或行为信号的高级智能,是一个典型而又复杂的模式识别问题,一直处于人工智能技术发展前沿,在新一代人工智能规划、“互联网+”行动计划等国家战略中具有重要地位。由于生物特征识别涉及公众利益攸关的隐私、道德和法律等问题,近期也引起了广泛的社会关注。本文系统综述了生物特征识别学科发展现状、新兴方向、存在问题和可行思路,深入梳理了人脸、虹膜、指纹、掌纹、静脉、声纹、步态、行人重识别以及多模态融合识别的研究进展,以人脸为例重点介绍了生物特征识别领域近些年受到关注的新方向——对抗攻击和防御、深度伪造和反伪造,最后剖析总结了生物特征识别领域存在的3大挑战问题——“感知盲区”、“决策误区”和“安全红区”。本文认为必须变革和创新生物特征的传感、认知和安全机制,才有可能取得复杂场景生物识别学术研究和技术应用的根本性突破,破除现有生物识别技术的弊端,朝着“可感”、“可知”和“可信”的新一代生物特征识别总体目标发展。