一种改进的神经网络模型在故障诊断中的应用

利用粗糙集理论对知识的约简能力及神经网络的自学习、自适应能力,构建了粗糙集-神经网络故障诊断模型,并对BP和Elman两种神经网络比较分析。仿真结果表明,与BP结合的方法更能简化神经网络结构,减少网络的训练时间,提高故障诊断的准确率。     

基于CMAC的双足步行机器人逆运动学控制

针对双足步行机器人(Biped Walking Robot)腿部逆运动学模型求解问题,采用一种基于CMAC神经网络的机器人逆运动学控制方法,设计CMAC神经网络控制系统.控制系统采用2个CMAC神经网络控制器分别用来逼近步行机器人支撑腿与摆动腿的逆模型,跟踪通过三维线性倒立摆模型生成的给定腰部轨迹.建立步行机器人正运动学模型来调整CMAC神经网络权值,实现了步行器人腿部逆运动学映射.仿真结果表明,CMAC神经网络控制系统可以在保证机器人位姿良好的情况下跟踪给定的参考轨迹.三维运动学仿真结果进一步验证了控制算法的有效性.     

粗集-神经网络在雷达信号识别中的应用研究

信号识别是侦察系统信号处理的目的,是整个雷达对抗信号处理中关键性的一个环节。为解决雷达信号识别的问题,提出将粗集和神经网络紧密结合建立新的识别模型,该模型充分融合了粗集强大的规则提取能力和神经网络优良的分类能力。实验表明,该模型减少了识别的主观因素,简化了神经网络结构,能够对雷达信号有效地识别。     

基于粗糙集和遗传算法的神经网络模型研究

提出一种基于粗糙集理论和遗传算法的神经网络模型和它的构造方法.该模型先利用粗糙集理论进行属性约简;利用遗传算法优化BP网络参数;用约简结果和优化的BP网络参数进行网络训练.仿真实验结果表明,该模型能简化网络训练样本,优化神经网络结构,提高系统的学习效率和精度.此方法是有效可行的,具有理论意义和实用价值.     

基于数据融合的自适应神经网络模糊控制系统设计

基于数据融合的思想，提出一种非线性系统的自适应神经网络模糊控制器的设计方法。该方法利用数据融合技术降低了模糊控制器的输入维数，简化了模糊控制器的设计。用自适应神经模糊推理系统的神经网络自学习功能完成模糊控制器的设计。仿真结果表明，自适应神经网络模糊控制系统性能优于采用普通的模糊控制器的情况，为数据融合与智能系统技术在非线性系统中的应用作了有益的探索.     

整合-激发神经网络的同步脉冲发放

研究一种简化的整合-激发神经网络的集体行为.该模型显示出短时行为并快速收敛到周期激发模式,同时伴有局部的同步脉冲发放行为.用标准欧拉算法研究了该模型的动态行为,结果显示当用标准欧拉算法研究用以描述大型整合-激发神经网络的动态行为的微分方程时,必须用很小的时间步长才能准确产生局部的同步脉冲发放.为此,提出一种改进的欧拉算法,该方法用一个线性插值来确定激发时间,大大提高了网络的性能.     

一个八自由度步行车辆模型及腿机构的运动学分析

在越野行驶时,步行车辆的性能明显优于轮式车辆.本文提出了一个八自由度六足步行车辆模型.它在越野行走时,可以完成象轮式车辆在平路行驶时的功能.在不平路面上,通过简单的控制装置,可以保持车体的水平,其结构较以往的步行车辆的结构大为简化.同时,提出一种适合于该车辆的腿机构,并对其进行了运动学分析.分析结果表明,该腿机构有良好的性能,完全满足步行车辆的要求.     

基于粗糙集-BP神经网络的机车滚动轴承故障诊断

论文提出了一种基于粗糙集理论与BP神经网络相结合的机车滚动轴承故障诊断方法.首先对原始故障诊断样本的连续属性进行离散化处理,然后利用粗糙集理论,对条件属性进行约简,删除冗余信息,最后将约简的最小属性集作为BP神经网络的输入,并设计BP神经网络对滚动轴承进行诊断.仿真结果表明粗糙集-BP模型不仅简化神经网络结构,而且提高了收敛速度和故障诊断正确率.     

粗糙集-神经网络在作战效能评估中的应用

为了提高作战效能评估的准确度,将粗糙集理论和神经网络引入到作战效能评估研究中,提出了粗糙集与神经网络相结合的作战效能评估方法.应用粗糙集简化神经网络训练样本数据集,在保留重要信息的前提下消除冗余的数据,仿真实验表明评估精度提高了,并且能获得更好的效果.以坦克作战效能为例,构建了坦克的效能评估模型,给出了基于粗糙集和神经...     

基于粗糙集与神经网络的故障诊断研究

通过引入粗糙集理论,利用可辨识矩阵约简算法对故障诊断决策表进行属性约简,剔除其中不必要的属性,然后构造改进的BP神经网络作为粗糙集的后端处理机,构造了基于粗糙集与神经网络的故障诊断模型。仿真结果表明,该方法可以有效地减少输入层个数,简化神经网络结构,减少网络的训练时间,在故障诊断中有良好的应用前景。     

仿人机器人相似性运动研究进展

针对仿人机器人模仿人体运动问题, 从运动轨迹角度比较了基于运动解析方程方法与基于人体运动相似性方法的特点, 阐述了相似性运动系统基本结构, 分析了图像捕捉与处理、相似性特征处理、相似性运动约束与优化等模块功能, 阐述了相似性运动中的人体运动捕获与处理、运动关节解算、运动模型简化与重定向、关键姿势处理与相似度评价、关节空间位姿计算、动力学匹配约束等方面的研究现状, 最后提出了研究展望。     

Neural inverse modeling and control of a base-excited inverted pendulum

In this paper multilayer neural networks are used to control the balancing of a base-excited inverted pendulum. The pendulum has 2 degrees of rotational freedom and the base-point moves freely in three-dimensional space. The goal is to apply control torques to keep the pendulum in a prescribed orientation in spite of disturbing base-point movement. A control algorithm is proposed that utilizes a set of neural networks to compensate for the effect of system's nonlinearities. These networks are updated on-line, according to a learning algorithm, which guarantees the stability of the closed-loop system. Furthermore, since the pendulum's base-point movement is considered unmeasurable, a novel neural inverse model is employed to estimate it from measurable variables. The proposed neural controller has been tested through simulations. Its performance has also been compared with the performance of the most recently developed control technique on the same problem. It is shown that the proposed neural controller produces fast, yet well maintained damped responses with reasonable control torques and without a knowledge of the model or model parameters. Additionally, the developed controller does not require measurement of the base-point accelerations, which are difficult to obtain. The work presented here benefits practical problems such as the study of stable locomotion of human upper body and bipedal locomotion.     

Model and control of the locomotion of a biomimic musculoskeletal biped

This article presents a biomimic musculoskeletal biped which contains 7 segments and 18 muscles. The muscle model and body dynamics are constructed based on physiological theories. A motor control system is designed to mimic natural human locomotion, which contains a central pattern generator, a regulator, a compensator, and an impedance controller. The recurrent neural oscillator models the central pattern generator, and an artificial neural network is used to design the regulator. From the simulation study, we found that this biped can produce a rhythmic and stable walking movement similar to actual human walking.     

Adaptive controller design for underwater snake robot with unmatched uncertainties

Because of hydrodynamic model error of the present dynamic model, there is a challenge in controller design for the underwater snake-like robot. To tackle this challenge, this paper proposes an adaptive control schemes based on dynamic model for a planar, underwater snake-like robot with model error and time-varying noise. The adaptive control schemes aim to achieve the adaptive control of joint angles tracking and the direction of locomotion control. First, through approximation and reducibility using Taylor expansion method, a simplified dynamics model of a planar amphibious snake-like robot is derived. Then, the L1 adaptive controller based on piecewise constant adaptive law is applied on the simplified planar, underwater snake-like robot, which can deal with both matched and unmatched nonlinear uncertainties. Finally, to control the direction of locomotion, an auxiliary bias signal is used as the control input to regulate the locomotion direction. Simulation results show that this L1 adaptive controller is valid to deal with different uncertainties and achieve the joint angles tracking and fast adaptive at the same time. The modified L1 adaptive controller, in which the auxiliary bias item is added, has the ability to change the direction of locomotion, that is, the orientation angle is periodic with arbitrarily given constant on average.     

Biped Locomotion Control through a Biomimetic CPG-based Controller

Modern concepts of motor learning favour intensive training directed to the neural networks stimulation and reorganization within the spinal cord, the central pattern generator, by taking advantage of the neural plasticity. In the present work, a biomimetic controller using a system of adaptive oscillators is proposed to understand the neuronal principles underlying the human locomotion. A framework for neural control is presented, enabling the following contributions: a) robustness to external perturbations; b) flexibility to variations in the environmental constraints; and c) incorporation of volitional mechanisms for self-adjustment of gait dynamics. Phase modulation of adaptive oscillators and postural balance control are proposed as main strategies for stable locomotion. Simulations of the locomotion model with a biped robot in closed-loop control are presented to validate the implemented neuronal principles. Specifically, the proposed system for online modulation of previous learnt gait patterns was verified in terrains with different slopes. The proposed phase modulation method and postural balanced control enabled robustness enhancement considering a broader range of slope angles than recent studies. Furthermore, the system was also verified for tilted ground including different slopes in the same experiment and uneven terrain with obstacles. Adaptive Frequency Oscillators, under Dynamic Hebbian Learning Adaptation mechanism, are proposed to build a hierarchical control architecture with spinal and supra spinal centers with multiple rhythm-generating neural networks that drive the legs of a biped model. The proposed neural oscillators are based on frequency adaptation and can be entrained by sensory feedback to learn specific patterns. The proposed biomimetic controller intrinsically generates patterns of rhythmic activity that can be induced to sustain CPG function by specific training. This method provides versatile control, paving the way for the design of experimental motor control studies, optimal rehabilitation procedures and robot-assisted therapeutic outcomes.     

Application of the utility function method for behavioral organization in a locomotion task

The generation of a complete robotic brain for locomotion based on the utility function (UF) method for behavioral organization is demonstrated. A simulated, single-legged hopping robot is considered, and a two-stage process is used for generating the robotic brain. First, individual behaviors are constructed through artificial evolution of recurrent neural networks (RNNs). Thereafter, a behavioral organizer is generated through evolutionary optimization of utility functions. Two systems are considered: a simplified model with trivial dynamics, as well as a model using full newtonian dynamics. In both cases, the UF method was able to generate an adequate behavioral organizer, which allowed the robot to perform its primary task of moving through an arena, while avoiding collisions with obstacles and keeping the batteries sufficiently charged. The results for the simplified model were better than those for the dynamical model, a fact that could be attributed to the poor performance of the individual behaviors (implemented as RNNs) during extended operation.     

Computational evolution of human bipedal walking by a neuro-musculo-skeletal model

The acquisition process of bipedal walking in humans was simulated using a neuro-musculo-skeletal model and genetic algorithms, based on the assumption that the shape of the body has been adapted for locomotion. The model was constructed as 10 two-dimensional rigid links with 26 muscles and 18 neural oscillators. Bipedal walking was generated as a mutual entrainment between neural oscillations and the pendulous movement of body dynamics. Evolutionary strategies incorporated, for example, as fitness in the genetic algorithms were assumed to decrease energy consumption, muscular fatigue, and load on the skeletal system. An initial population of 50 individuals was created, and an evolutionary simulation of 5000 steps was conducted. As a result, the shape of the body changed from that of a chimpanzee to that of a modern human, and the body size nearly reached the size of a modern human. These simulation results show that improving locomotive efficiency and reducing the load on the musculo-skeletal system are important factors affecting the evolution of the human body shape and bipedal walking. Such computer simulations help us to understand the process of evolution and adaptation for locomotion in humans. This work was presented, in part, at the Third International Symposium on Artificial Life and Robotics, Oita, Japan, January 19–21, 1998     

Development of a pulse control-type MEMS microrobot with a hardware neural network

This article presents the micro-electro-mechanical systems (MEMS) microrobot which demonstrates locomotion controlled by hardware neural networks (HNN). The size of the microrobot fabricated by the MEMS technology is 4 × 4 × 3.5 mm. The frame of the robot is made of silicon wafer, and it is equipped with a rotary-type actuator, a link mechanism, and six legs. The rotary-type actuator generates rotational movement by applying an electrical current to artificial muscle wires. The locomotion of the microrobot is obtained by the rotation of the rotary-type actuator. As in a living organism, the HNN realized robot control without using any software programs, A/D converters, or additional driving circuits. A central pattern generator (CPG) model was implemented as an HNN system to emulate the locomotion pattern. The MEMS microrobot emulated the locomotion method and the neural networks of an insect with the rotary-type actuator, the link mechanism, and the HNN. The microrobot performed forward and backward locomotion, and also changed direction by inputting an external trigger pulse. The locomotion speed was 0.325 mm/s and the step width was 1.3 mm.     

基于改进SPSO-BP神经网络的温度传感器湿度补偿

针对在实际使用中湿度影响温度传感器准确性的问题,通过对基本粒子群算法的分析,得出不受速度向量影响的简化粒子群算法,同时采用线性递减惯性权重,提出了一种改进SPSO-BP神经网络温度传感器的湿度补偿方法.通过改进的简化粒子群算法的不断迭代,优化BP神经网络的权阈值,直到得到最优权阈值,并赋给BP神经网络.根据湿度影响实验中测得的数据,运用此方法建立湿度补偿模型,与BP神经网络方法对比分析.结果表明,改进SPSO-BP神经网络的模型结构简单、补偿精度高,收敛速度快,有效地对温度传感器进行了湿度补偿.     

胸鳍推进型机器鱼的CPG 控制及实现

结合仿生游动机理,针对胸鳍推进型机器鱼提出了一种基于中枢模式发生器（CPG）的运动控制方法． 该模型采用一类振荡频率和幅值可以独立控制的非线性微分方程作为其神经元振荡器模型,通过最近相邻耦合的方 式,对n 个这样的神经元振荡器进行耦合,构建了仿生机器鱼的CPG 网络模型．证明了此模型单个神经元振荡器的 极限环的存在性、唯一性及稳定性．在此基础上,通过对胸鳍推进的运动学分析,导出机器人直游、倒游、胸鳍—尾 鳍协调运动等多种模式的运动控制方法．仿真及实验结果验证了此中枢模式发生器模型的可行性与所提控制方法的 有效性．