精细神经网络仿真方法研究进展

树突对大脑神经元实现不同的信息处理功能有着重要作用。精细神经元模型是一种对神经元树突以及离子通道的信息处理过程进行精细建模的模型,可以帮助科学家在实验条件的限制之外探索树突信息处理的特性。由精细神经元组成的精细神经网络模型可通过仿真对大脑的信息处理过程进行模拟,对于理解树突的信息处理机制、大脑神经网络功能背后的计算机理具有重要作用。然而,精细神经网络仿真需要进行大量计算,如何对精细神经网络进行高效仿真是一个具有挑战的研究问题。本文对精细神经网络仿真方法进行梳理,介绍了现有主流仿真平台与核心仿真算法,以及可进一步提升仿真效率的高效仿真方法。将具有代表性的高效仿真方法按照发展历程以及核心思路分为网络尺度并行方法、神经元尺度并行方法以及基于GPU(graphics processing unit)的并行仿真方法3类。对各类方法的核心思路进行总结,并对各类方法中代表性工作的细节进行分析介绍。随后对各类方法所具有的优劣势进行分析对比,对一些经典方法进行总结。最后根据高效仿真方法的发展趋势,对未来研究工作进行展望。     

A study on human gait dynamics: modeling and simulations on OpenSim platform

Multimedia Tools and Applications - The analysis of human gait dynamics allows an individual to obtain interesting biometric features through which gait disturbances can be observed from normal and...     

Kinematic simulation of human gait with a multi-rigid-body foot model

The paper builds a multi-rigid-body model of human with a 4-rigid-body foot in the 3D CAD software Solidworks,based on human anatomy.By controlling the rotation of the ankle and major joints of human body while walking,the Kinematic simulation was performed in the dynamics simulation software ADAMS.The paper analyzes the simulate results and points out deficiencies in the current work and the direction of research efforts in future.     

Solution methods for the double-support indeterminacy in human gait

There is a growing interest in predicting the gait motion of real subjects under virtual conditions, e.g., to anticipate the result of surgery or to help in the design of prosthetic/orthotic devices. To this end, the motion parameters can be considered as the design parameters of an optimization problem. In this context, determination of the joint efforts for a given motion is a required step for the subsequent evaluation of cost function and constraints. In the double-support phase of gait, the ground reaction forces include twelve unknowns, rendering the inverse dynamics problem indeterminate if no force plate data are available. In this paper, several methods for solving the inverse dynamics problem of the human gait during the double-support phase, using force plates or not, are presented and compared.     

The effects of different carrying methods on human gait parameters

Lifting and carrying are essential tasks that affect human balance and gait. Choosing a certain gait pattern could help reduce the risk of falling. This study investigates the differences in human gait parameters and lateral bending of the trunk when carrying a 5-gallon water bottle, compared to that of normal walking. Several gait parameters were considered, including the cadence, stride width, step length, total double support duration, walking speed, toe angle, and single support duration. A laboratory experiment was conducted considering 23 healthy males, 18–30 years in age, performing several carrying scenarios, with and without the use of two assistive devices (a bottle lifting handgrip handle and back and lumbar support). The ProtoKinetics Zeno™ Walkway Gait Analysis System and the ProtoKinetics Movement Analysis Software were used to measure the spatiotemporal gait parameters. In addition, the lumbar spine's lateral bending was measured using Kinovea software. The results showed that the assistive carrying devices helped achieve less deviation in the walking pattern while carrying, compared to that of normal walking, and reduced the lateral bending of the trunk, resulting in greater balance while carrying. This, in turn, helps reduce the chance of falling and the stress in the joints and muscles, thereby increasing stability. In conclusion, carrying two 5-gallon water bottles using a handgrip handle assistive device was the most preferred carrying method.     

节律性步态运动中CPG对肌肉的控制模式的仿真研究

以神经振荡器理论和Dingguo Zhang等人研究的CPG模型为基础,结合了神经生物学和生物动力学的观点,并根据人体腿部的肌肉结构修正了Dingguo Zhang等提出的中枢模式发生器(CPG)的数学模型.修正后的CPG模型突破了原来仅反映单腿的节律性运动的局限性,能够更好地描述步态运动中双腿的节律性与协调性,使得修正后的CPG模型与实际情况更为一致.依据本文所提出的修正模型所做的数值模拟结果表明,我们所得到的CPG的输出模式能够很好的表现出人体节律性步态运动中神经系统的调节作用.     

仿生机器人运动步态控制:强化学习方法综述

仿生机器人是一类典型的多关节非线性欠驱动系统，其步态控制是一个非常具有挑战性的问题。对于该问题，传统的控制和规划方法需要针对具体的运动任务进行专门设计，需要耗费大量时间和精力，而且所设计出来的控制器往往没有通用性。基于数据驱动的强化学习方法能对不同的任务进行自主学习，且对不同的机器人和运动任务具有良好的通用性。因此，近年来这种基于强化学习的方法在仿生机器人运动步态控制方面获得了不少应用。针对这方面的研究，本文从问题形式化、策略表示方法和策略学习方法3个方面对现有的研究情况进行了分析和总结，总结了强化学习应用于仿生机器人步态控制中尚待解决的问题，并指出了后续的发展方向。     

A review of motion analysis methods for human Nonverbal Communication Computing

Human Nonverbal Communication Computing aims to investigate how people exploit nonverbal aspects of their communication to coordinate their activities and social relationships. Nonverbal behavior plays important roles in message production and processing, relational communication, social interaction and networks, deception and impression management, and emotional expression. This is a fundamental yet challenging research topic. To effectively analyze Nonverbal Communication Computing, motion analysis methods have been widely investigated and employed. In this paper, we introduce the concept and applications of Nonverbal Communication Computing and also review some of the motion analysis methods employed in this area. They include face tracking, expression recognition, body reconstruction, and group activity analysis. In addition, we also discuss some open problems and the future directions of this area.     

Approaches to human arm movement control—A review

Human arm movement control theories are reviewed in the current work. The paper addresses the main paradigms that have been used in modeling studies of human arm movement control and details the models that resulted from each approach. The main motivation of the paper is to provide the background knowledge produced by the Control and Computational Modeling communities that may contribute to the development of new rehabilitation methods and devices in a more principled way. With this goal in mind, the main insights coming from the analysis of these computational models are identified and their potential relevance for the rehabilitation practice is described.     

Identification of elastic properties utilizing non-destructive vibrational evaluation methods with emphasis on definition of objective functions: a review

Structural and Multidisciplinary Optimization - Defining a suitable objective function is important as it affects the solution search process and the solution quality. In vibrational material...     

A device for three-dimensional registration of human movement

A device for three-dimensional registration of human movement was tested. For this, a special garment with 43 attached light-emitting diodes was developed. The garment was specially designed for biomechanical purposes and made of a highly elastic material in the form of an overall. A recording system working with infra-red light was also tested considering both technical aspects and software. Test recordings of predetermined body movements within a defined work area of over 10m² showed that the device satisfies general practical requirements. The garment worked very well for different movements and for bodies of different shapes and sizes. Technically the system was tested with respect to noise and linearity, operating function, range and manageability. It worked well with some limitations such as reflections. The tested software functions, such as recording speed, processing time, interpolation and user compatibility, proved to fulfil their main needs.     

A study on gait entropy image analysis for clothing invariant human identification

A simple and common human gait may be viewed as a strong biometric cue to solve human identification problem through understanding the intrinsic patterns of gait biometrics. An individual’s gait pattern appears to be different in gallery and probe gait sequences due to wearing dissimilar clothing types. The gait dataset captures the possible changes found in silhouette shape image which provides the difficulty in distinguishing among individuals. In this paper, a robust feature selection technique has been addressed through Gait Entropy Image (GEnI) analysis. The GEnI has the capacity to accumulate most significant motion information. The width of GEnI, along the horizontal axis is taken as discriminative feature which produces a small intra-class variance. This information is studied as an evidence of feature invariance. The standard statistical tests such as pair-wise clothing correlation and intra-clothing variance are performed on gait dataset to evaluate the reliability of feature. Experimental results demonstrate the efficiency of proposed feature selection method using k-nearest neighbor (k-NN), minimum distance classifier (MDC), and support vector machine (SVM) algorithms. The performance analysis of recognition system has been evaluated on OU-ISIR Treadmill B gait database with different error metrics after performing N-fold cross validation method.     

A simulation model for studies of intestine cell dynamics

A dynamic simulation model for the epithelial cell structures of the intestine has been developed. The model is based on hypotheses found in the literature. It is shown that basic cell dynamics can be reproduced by simulation experiments. The simulation program is designed so that model output comparable with experimental findings can be obtained. The use of a dynamic simulation model can be interpreted as a simulated experiment in vivo. The model therefore allows analysis of dynamic behaviour that cannot be performed by normal experimental work. This means that the model can be used as a tool for verification/falsification of hypotheses about dynamics in both normal and abnormal cell structures.     

A comprehensive review of past and present vision-based techniques for gait recognition

Global security concerns have raised a proliferation of video surveillance devices. Intelligent surveillance systems seek to discover possible threats automatically and raise alerts. Being able to identify the surveyed object can help determine its threat level. The current generation of devices provide digital video data to be analysed for time varying features to assist in the identification process. Commonly, people queue up to access a facility and approach a video camera in full frontal view. In this environment, a variety of biometrics are available—for example, gait which includes temporal features like stride period. Gait can be measured unobtrusively at a distance. The video data will also include face features, which are short-range biometrics. In this way, one can combine biometrics naturally using one set of data. In this paper we survey current techniques of gait recognition and modelling with the environment in which the research was conducted. We also discuss in detail the issues arising from deriving gait data, such as perspective and occlusion effects, together with the associated computer vision challenges of reliable tracking of human movement. Then, after highlighting these issues and challenges related to gait processing, we proceed to discuss the frameworks combining gait with other biometrics. We then provide motivations for a novel paradigm in biometrics-based human recognition, i.e. the use of the fronto-normal view of gait as a far-range biometrics combined with biometrics operating at a near distance.     

Predictive dynamics: an optimization-based novel approach for human motion simulation

Predictive dynamics is a novel approach for simulating human motion. It avoids direct integration of differential-algebraic equations in order to create the resulting simulations for redundant digital human models. Instead, it formulates an optimization problem by defining appropriate performance measures and constraints to recover the real motion of the dynamic system. In the formulation, both kinematics and kinetics parameters serve as unknowns, and equations of motion are treated as equality constraints. Procedures to choose physical performance measures and appropriate constraints based on the available information about the bio-system are presented. The proposed methodology is illustrated and studied by first predicting the swinging motion of a single pendulum with externally applied torque. The pendulum can represent the motion of upper and lower extremities. This simple problem has analytical solutions and is used to gain insights for the predictive dynamics approach. In addition, a complex human walking task is simulated by using the approach, and realistic results are obtained. Such motion prediction capabilities have a wide variety of applications for industries ranging from automotive to military to clinical analysis and design.     

Evolutionary search for understanding movement dynamics on mixed networks

This paper describes an approach to using evolutionary algorithms for reasoning about paths through network data. The paths investigated in the context of this research are functional paths wherein the characteristics (e.g., path length, morphology, location) of the path are integral to the objective purpose of the path. Using two datasets of combined surface and road networks, the research demonstrates how an evolutionary algorithm can be used to reason about functional paths. We present the algorithm approach, the parameters and fitness function that drive the functional aspects of the path, and an approach for using the algorithm to respond to dynamic changes in the search space. The results of the search process are presented in terms of the overall success based on the response of the search to variations in the environment and through the use of an occupancy grid characterizing the overall search process. The approach offers a great deal of flexibility over more conventional heuristic path finding approaches and offers additional perspective on dynamic network analysis.     

Multi-body dynamics simulation and gait pattern analysis of a bio-inspired quadruped robot for unstructured terrains using adaptive stroke length

This paper presents the Finite element analysis (FEA), Multi-body dynamics (MBD) simulation and gait pattern analysis of a bio-inspired quadruped robot. A novel power transmission mechanism with varying gear train centre distance has been designed for generating different gait structures to suit unstructured terrain. The adjustable length of the lower link aids in improving the stability of the robot (or) make the robot to travel longer distances within a stipulated time. MBD simulation has been performed using MSC/ADAMS for torque, speed, energy and force analysis. The novel mechanism designed was able to successfully negotiate unstructured obstacles of different heights and these results have been presented in the paper. The results show a direct relationship between stroke length, lift-off distance and gear centre distance. A full scale experimental setup, along with the suitable supporting mechanisms is developed. The experimental analysis has been carried out for measuring the Ground Reaction Force (GRF). The GRF is found to be 45 N on an average, when the robot is walking.     

Brain-computer interfaces for human gait restoration

In this review article,we present more than a decade of our work on the development of brain-computer interface(BCI)systems for the restoration of walking follo...