基于曲面形状误差的多层前向神经网络快速训练

如何显著提高多层前向神经网络训练速度一直是国内外共同关注的一个问题，而解决这个问题的关键在于充分了解导致现有网络训练算法训练效率低的根本原因．文中首先提出了网络输出函数的曲面形状误差和偏移误差的概念，并将指导网络训练的平方和误差分解为这两种误差，进而分析了这两种误差的主要特性，给出了导致现有算法网络训练效率低的主要原因，最后提出了新的网络训练误差模型和具体的网络训练算法．典型实例计算结果表明，与目前常用的网络训练算法相比，该文所提出的算法可显著减少网络训练时间。     

神经网络磁链估计的感应电机反步法研究

为实现感应电机的位置渐近跟踪，基于反步法并取转矩和磁链控制信号作为虚拟控制，设计了感应电机位置控制系统．采用多层前馈神经网络估计转子磁链，以Levenberg-Marquardt算法训练网络并调整权值．最后基于Lyapunov稳定性理论证明了系统的稳定性．仿真结果表明，所设计的神经网络磁链观测器具有良好的估计效果，位置跟踪误差迅速收敛，具有较优的伺服跟踪特性．     

速度受限的全方向康复步行训练机器人Backstepping补偿跟踪控制

全方向康复步行训练机器人需要跟踪医生指定的训练轨迹、针对运动过程中速度过快而影响康复者安全性问题,提出一种基于Backstepping补偿方法的速度受限型控制器设计方案,目的是保证康复者的安全训练.通过限制实际运动速度的幅值,并根据所提出轨迹跟踪误差和速度跟踪误差的实时补偿方法,实现了全方向康复步行训练机器人在安全速度下的轨迹跟踪.基于李亚普诺夫稳定性理论和LaSalle不变性原理,证明了补偿误差系统和跟踪误差系统的渐近稳定性.通过仿真结果对比分析,表明了所提控制器设计方法的有效性和可行性.     

气动肌肉关节的无模型自适应CMAC迟滞补偿控制

为补偿迟滞性对气动肌肉关节轨迹跟踪控制精度的破坏,首先建立了关节模型;推导得到迟滞力方程,测试分析了关节迟滞性;而后设计了无模型自适应CMAC神经网络迟滞补偿算法,该算法采用了充、放气双重结构;采用梯度下降法实时反馈调整充、放气过程的网络权值;采用4阶傅里叶拟合函数对网络权值降噪;基于高斯函数和邻域误差,设计误差可信度评估函数来调节学习率,抑制干扰对神经网络的影响;而后用三角波轨迹跟踪控制对神经网络进行了学习训练;最后将训练好的神经网络用于突发干扰下的正弦波轨迹跟踪控制.实验结果表明,该算法能自适应非线性曲线跟踪控制中的迟滞变化,有效抑制突发干扰,提高控制精度.     

一类未知多变量非线性系统的动态神经网络自适应控制

对一类未知的非线性的多变量系统，提出了用动态神经网络实现直接自适应控制的策略，基于Ｌｙａｐｕｎｏｖ理论，获得一个稳定并且连续的学习律，避免了递归训练过程，闭环系统被证明是鲁棒稳定的，跟踪误差收敛到一个小的残集，这种方法的特点是即不需要离线学习阶段也不要求初始的参数误差足够小，仿真结果验证了提出的动态网络的自适应控制算法的有效性。     

ESN网络在机器人轨迹跟踪控制中的应用

为了研究具有模型不确定性的机器人操作手的轨迹跟踪控制,采用一种新的递归神经网络——回声状态网络(ESN)设计了动态控制器.采用PID控制器补偿ESN网络的逆建模误差,并在网络训练过程中加入白噪声项,以保证动态系统的稳定性.最后针对两关节机械手的轨迹跟踪控制问题进行了数值仿真,仿真结果表明了该方法的有效性.     

CG-DRR:数字重建放射影像循环一致性生成对抗算法

目的 当前数字重建放射影像（digitally reconstructed radiograph,DRR）生成算法难以同时保证影像生成效率和质量。为此,提出一种基于循环一致性生成对抗网络（generative adversarial network, GAN）的新型DRR生成算法（cycle-GAN-DRR, CG-DRR）,在保证生成效率的同时,兼顾生成影像与真实X射线影像的灰度和结构一致性。方法CG-DRR算法包含数据预处理、网络训练和网络应用3个阶段。数据预处理阶段准备后续网络训练需要的图像数据;网络训练阶段使用预处理后的数据训练循环一致性生成对抗网络;在应用阶段,训练后网络输入光线跟踪法生成的DRR图像实现DRR图像灰度校正。结果 使用4个常用图像相似性指标和2个梯度相似性指标分别评估原始DRR图像与灰度校正后的DRR图像的灰度和几何结构一致性。与传统光线跟踪法相比,盆腔和胸腔数据平均峰值信噪比分别提高18.22 dB和8.82 dB,平均绝对误差分别减少0.18和0.07,归一化均方根误差分别减少0.23和0.10,结构相似度指数分别提高23.5%和13.5%,与当前最新Re...     

基于线性参数神经网络的非线性系统稳定自适应控制

提出适用于多种网络类型的神经网络稳定自适应控制设计思想,在神经网络逼近误差界未知的条件下,对该误差界进行在线自适应估计,研究基于线性参数神经网络的仿射非线性系统稳定自适应控制。采和Ｌａｐｕｎｏｖ函数方法证明系统状态变量、网络权值矢量、网络逼近误差界的在线估计及输出跟踪误差的收敛性。仿真结果表明,该方案跟踪性能良好,稳态误差较小,系统输出能快速跟踪目标信号。     

基于蚁群算法的BP网络优化算法

将BP网络的训练误差和检验误差用于引导蚂蚁行经路径上的信息更新机制和选择机制,并据此计算蚂蚁行径中的转移概率;又将蚂蚁行经路径上的存储单元存放的参数值赋予BP网络训练,而存储单元存放的参数和训练误差值亦随BP网络训练误差的调整而改变。通过交互迭代优化,最终得到调整后的BP网络的最佳参数值。数值模拟计算结果表明:与传统的BP算法相比,在达到同一数量级的训练误差情况下,基于蚁群算法优化参数的BP算法训练次数少,而模型的精度高,在一定程度上提高了BP网络的学习能力和泛化能力。     

基于LM算法的在线自适应RBF网结构优化算法

针对LM算法不能在线训练RBF网络以及RBF网络结构设计算法中存在的问题,提出一种基于LM算法的在线自适应RBF网络结构优化算法.该算法引入滑动窗口和在线优化网络结构的思想,滑动窗口的引入既使得LM算法能够在线训练RBF网络,又使得网络对学习参数的变化具有更好的鲁棒性,并且易于收敛.在线优化网络结构使得网络在学习过程中能够根据训练样本的训练误差和隐节点的相关信息,在线自适应调整网络结构,跟踪非线性时变系统的变化,使网络维持最为紧凑的结构,以保证网络的泛化性能.最后通过仿真实验验证了所提出算法的性能.     

Adaptive neural output feedback tracking control for a class of nonlinear systems

In this paper, an adaptive neural output feedback control scheme based on backstepping technique and dynamic surface control (DSC) approach is developed to solve the tracking control problem for a class of nonlinear systems with unmeasurable states. Firstly, a nonlinear state observer is designed to estimate the unmeasurable states. Secondly, in the controller design process, radial basis function neural networks (RBFNNs) are utilised to approximate the unknown nonlinear functions, and then a novel adaptive neural output feedback tracking control scheme is developed via backstepping technique and DSC approach. It is shown that the proposed controller ensures that all signals of the closed-loop system remain bounded and the tracking error converges to a small neighbourhood around the origin. Finally, two numerical examples and one realistic example are given to illustrate the effectiveness of the proposed design approach.     

基于神经网络观测器的船舶轨迹跟踪递归滑模 动态面输出反馈控制

针对三自由度全驱动船舶速度向量不可测问题,考虑船舶模型参数和外部环境扰动均未知的情况,提出一种基于神经网络观测器的船舶轨迹跟踪递归滑模动态面输出反馈控制方法.该方法设计神经网络自适应观测器估计船舶速度向量,且利用神经网络逼近模型参数不确定项,综合考虑船舶位置和速度误差之间关系构造递归滑模面,再采用动态面控制技术设计轨迹跟踪控制律和参数自适应律,并引入低频增益学习方法消除外界扰动导致的高频振荡控制信号.选取李雅普诺夫函数证明了该控制律能够保证轨迹跟踪闭环系统内所有信号的一致最终有界性.最后,基于一艘供给船进行仿真验证,结果表明,船舶轨迹跟踪响应速度快,所设计控制器对系统模型参数摄动及外界扰动具有较强的鲁棒性.     

基于递归神经网络的再入飞行器最优姿态控制

针对再入飞行器的姿态跟踪问题,基于递归神经网络提出最优跟踪控制.采用反步法和递归神经网络,设计自适应前馈控制,将再入飞行器的最优姿态跟踪问题转化为等价的姿态角误差/角速率误差最优调节问题.采用自适应动态规划技术,解决最优调节问题.引入神经网络估计最优控制中的代价函数,推导最优反馈控制律,同时保证Hamilton–Jacobi–Isaacs(HJI)方程估计误差最小化.采用Lyapunov理论,保证闭环系统中所有信号,包括姿态角跟踪误差是一致最终有界的.在MATLAB/Simulink中仿真验证了所提出控制策略的有效性.     

Autonomous mobile robots navigation using RBF neural compensator

This paper presents an approach to adaptive trajectory tracking of mobile robots which combines a feedback linearization based on a nominal model and a RBF-NN adaptive dynamic compensation. For a robot with uncertain dynamic parameters, two controllers are implemented separately: a kinematics controller and an inverse dynamics controller. The uncertainty in the nominal dynamics model is compensated by a neural adaptive feedback controller. The resulting adaptive controller is efficient and robust in the sense that it succeeds to achieve a good tracking performance with a small computational effort. The analysis of the RBF-NN approximation error on the control errors is included. Finally, the performance of the control system is verified through experiments.     

Indirect adaptive self-organizing RBF neural controller design with a dynamical training approach

This study proposes an indirect adaptive self-organizing RBF neural control (IASRNC) system which is composed of a feedback controller, a neural identifier and a smooth compensator. The neural identifier which contains a self-organizing RBF (SORBF) network with structure and parameter learning is designed to online estimate a system dynamics using the gradient descent method. The SORBF network can add new hidden neurons and prune insignificant hidden neurons online. The smooth compensator is designed to dispel the effect of minimum approximation error introduced by the neural identifier in the Lyapunov stability theorem. In general, how to determine the learning rate of parameter adaptation laws usually requires some trial-and-error tuning procedures. This paper proposes a dynamical learning rate approach based on a discrete-type Lyapunov function to speed up the convergence of tracking error. Finally, the proposed IASRNC system is applied to control two chaotic systems. Simulation results verify that the proposed IASRNC scheme can achieve a favorable tracking performance.     

An intelligent robust tracking control for electrically-driven robot systems

This article addresses the problem of designing intelligent robust tracking controls of robot systems actuated by brushed direct current motors. The structures of both mechanical and electrical dynamics are allowed to be completely unknown and adaptive fuzzy (or neural network) systems are employed to approximate these two uncertainties. Consequently, an adaptive fuzzy-based (or neural network-based) state feedback tracking controller is developed such that the resulting closed-loop system guarantees that all the states and signals are bounded and the tracking error can be made as small as possible. Finally, simulation examples are made to demonstrate the effectiveness and tracking performance.     

ANALYSIS AND DESIGN OF FEEDBACK CONTROL SYSTEMS WITH TRACKING ERRORS SQUARE

In this study, a command tracking error square control scheme is first proposed for analysis and design of feedback control systems. One of the tracking errors is low‐pass filtered and used in the feedback loop for gain adaptation; the other is used in the forward loop for command tracking control. The overall systems are nonlinear feedback systems, and can be reconfigured to an automatic gain control (AGC) loop with command tracking error input. The stability and robustness of the controlled systems are verified by time response, frequency response, and large parameter variation testing with a simple illustrating example and are finally applied to a complicated electro‐hydraulic velocity servo system with large load disturbance.     

一类非线性不确定系统的神经网络控制

针对一类非线性不确定系统,提出了一种自适 应神经网络控制方案．被控系统是部分已知的,其中系统已知的动态特性被用来设计保证标 称模型稳定的反馈控制器,而基于神经网络的动态补偿器则用于补偿系统的非线性不确定性 ,从而可以保证系统输出跟踪误差渐近收敛于0．     

带执行器饱和的N 连杆机械臂输出反馈动态面控制

针对带执行器饱和的多关节刚性机械臂系统,提出一种基于RBF神经网络补偿的输出反馈动态面控制.通过观测器实现角速度的观测,采用RBF网络实现执行器饱和的补偿;通过Lyapunov方法证明闭环系统的稳定性,实现高精度的角度和角速度跟踪.仿真结果表明,所提出的方法能够有效补偿系统存在的执行器饱和,显著减小跟踪误差,并且对于外界干扰具有一定的鲁棒性.     

Output feedback adaptive DSC for nonlinear systems with guaranteed L-infinity tracking performance

In this paper, the problem of output tracking for a class of uncertain nonlinear systems is considered. First, neural networks are employed to cope with uncertain nonlinear functions, based on which state estimation is constructed. Then, an output feedback control system is designed by using dynamic surface control (DSC). To guarantee the L-infinity tracking performance, an initialization technique is presented. The main feature of the scheme is that explosion of complex- ity problem in backstepping control is avoided, and there is no need to update the unknown parameters including control gains as well as neural networks weights, the adaptive law with one update parameter is necessary only at the first design step. It is proved that all signals of the closed-loop system are semiglobally uniformly ultimately bounded and the L-infinity performance of system tracking error can be guaranteed. Simulation results demonstrate the effectiveness of the proposed scheme.