基于改进的CMAC神经网络与PID并行控制的研究

提出一种改进的CMAC神经网络控制算法，利用满打满葬单元的先前学习次数作为可信度；将改进的CMAC与PID实现复合控制，由CMAC控制器实现前馈控制，PID控制实现反馈控制；仿真表明，改进算法的响应速度和精度有一定的改善。     

基于平衡学习的CMAC神经网络非线性滑模容错控制

以一改进的信度分配CMAC(cerebellar model articulation controllers)神经网络为在线故障诊断的手段,将变结构滑模摔制技术引入容错控制器设计之中,提出一种动态非线性系统主动容错控制方法.在常规CMAC学习算法中,误差被平均地分配给所有被激活的存储单元,不管各存储单元存储数据(权值)的可信程度.改进的CMAC中,利用激活单元先前学习次数作为可信度,其误差校正值与激活单元先前学习次数的-p次方成比例,从而提高神经网络的在线学习速度和精度;在此基础上利用滑模控制算法进行容错控制律的在线重构,实现动态非线性系统在线故障诊断与容错控制的集成.分析了系统的稳定性,仿真结果表明改进故障学习算法及容错控制的有效性.     

基于平衡学习的CMAC 神经网络非线性辨识算法

为提高小脑模型关节控制器(CMAC)神经网络在线学习的快速性和准确性，提出一种平衡学习的概念，并设计一种改进的CMAC学习算法．在常规的CMAC中，误差的校正值被平均地分配给所有激活存储单元，而不管这些存储单元的可信度；在改进的CMAC中，利用激活单元先前学习次数作为可信度，其误差校正值与激活单元先前学习次数的负k次方成比例．仿真结果表明，当k为一适当数值时，改进CMAC具有较快的学习速度和较高的精度，特别是在神经网络的初始学习阶段．     

CMAC最优设计及其算法——GA技术优化CMAC偏移矢量分布

采用GA(Genetic Algorithm)技术实现CMAC(cerebellar Model Articulation Controller)最优设计及算法.该方法解决了CMAC与其学习对象的整体优化问题,具有理论 意义和实用价值.仿真结果证明该方法是成功的和有效的.对不同的客观对象(如空间曲面), 可以采用GA技术找到CMAc的最优内部表示(偏移矢量分布),实现一般CMAC难以达到 的精度.该方法比Albus的CMAC和Parks等的CMAC学习效果都有不同程度的提高,适 合于要求高精度学习的情况.同时给出了任意偏移矢量分布的CMAC算法.     

改进的小脑控制神经网络算法研究

研究小脑控制器神经网络(CMAC)在模式识别中的应用问题.在算法优化过程中,针对模式识别中训练样本中存在着高维、大量冗余信息而传统CMAC不能够对输入信息空间维数降低,从而常导致CMAC网络训练速度慢、识别率低等问题.为了加快CMAC学习速度,提高识别率,提出一种基于粗糙集(RS)的CMAC模式识别方法(RS_CMAC).利用粗糙集约简并删除训练样本中的冗余信息,降低样本的维数,从而优化网络的结构;同时网络中引入了自适应动态学习率,加快网络的收敛速度和学习速度,从而提高识别率.以数码管模式识别为例对算法进行了验证性实验,仿真结果表明,与传统CMAC相比.识别精度有了明显提高,学习速度加快.RS_CMAC方法克服了传统CMAC缺点和不足,是一种有效的模式识别方法,为实际应用提供依据.     

具有微分输出的神经网络New-CMAC及其学习收敛性

基于传统的CMAC神经网络和局部加权回归技术,提出了与传统CMAC(cerebellar model articulation computer)有着同样存储空间量的改进的新CMAC网络New-CMAC,它具有传统的输出和具有其微分信息的输出,因而更适合于自动控制.接着,又提出了其新的学习算法,并研究了其学习收敛性.     

神经计算中坐标变换的网络模型(CMAC)的泛化特性

在神经计算中神经网络的泛化特性是一个非常重要的内容.该文简述了小脑模型 (CMAC--Cerebellar Model Areiculation Controller)的原理和学习算法,并用仿真方法讨 论了在机器人使用的坐标变换关系(输入直角坐标值,输出机器手的关节角度)下CMAC的 泛化性能:当泛化率为1:100时CMAC仍能正常工作.系统的精度虽能满足需要,但是进一 步提高却受到限制.本文还讨论了影响精度的各种因素及可能的改进方法.     

基于广义基函数的CMAC学习算法的改进及收敛性分析

基于广义基函数的CMAC(Cerebeliar Model Articulation Controller)学习算法(称 C-L算法)收敛条件依赖于基函数和学习样本,很难同时满足学习快速性与收敛性.提出了一 种改进学习算法,并证明改进算法是收敛的,而且收敛条件不依赖于基函数和学习样本.仿真 结果表明改进算法优于C—L算法和标准的Albus算法.     

一类具有多维存储结构的CMAC神经网络

对于多输入 CMAC神经网络 ,概念映射决定了网络泛化特性 .从而直接影响到网络学习速度和网络的逼近精度 .本文在对多输入 CMAC网络进行深入分析的基础上 ,指出了概念映射的重要性及现有映射算法存在的不足 ;巧妙地将网络存储器由原来的一维改为多维从而得到了一类新型的多输入 CMAC神经网络 ,同时给出了新型网络的概念映射算法并分析论证了算法的合理性、严密性 .计算机仿真结果验证了所提出新型网络及其概念映射算法的有效性     

CMAC 算法收敛性分析及泛化能力研究

利用矩阵理论和线性方程组迭代收敛的一般性原理,在不附加特殊条件折情况下,证明了CMAC算法在批量和增量两种学习方式下的收敛定理,对在关联矩阵正定条件下得出的结论进行推广和改进。在此基础上提出了一种学习率自寻优的CMAC改进算法,并提出一种简单可行的评价CMAC网络整体泛化性能的指标,通过计算仿真验证了收敛定量的正确性和改进算法的优越性,并研究得出了CMAC网络各个参数对其泛化性能影响的相关结论。     

超闭球CMAC的性能分析及多CMAC结构

如何选择合适网络参数是传统CMAC(Cerebellar Model Articulation Controller)应用 中的一个难题.采用泛化均方差(GMSE)和学习均方差(LMSE)来分别评价超闭球CMAC的泛 化能力与记忆精度,并引入权调整率的概念,来研究CMAC结构参数与学习性能的关系.研究 结果表明,在样本分布和量化级数不变时,泛化均方差和学习均方差是权调整率的非增函数.因 此超闭球CMAC在满足存储空间和计算速度的要求下尽量使得权调整率较大.还提出了并行 CMAC结构以进一步提高单个超闭球CMAC的非线性逼近能力.仿真结果证明了该方法的有 效性.     

A Modified CMAC Algorithm Based on Credit Assignment

A Credit-Assignment CMAC (CA-CMAC) algorithm is proposed to reduce learning interference in conventional CMAC. In the proposed CA-CMAC, the error of the training sample distributed to the addressed memory cell is proportional to the cell's credibility, which is the inverse of the cell's activated times. The learning process of CA-CMAC is analyzed and conventional CMAC is proved to be a special case of CA-CMAC. Furthermore, the convergence properties of CA-CMAC both in batch learning and in incremental learning are investigated; meanwhile, the convergence theorems in the two learning schemes are obtained, respectively. Finally, simulations are carried out to testify the theorems and compare the performance of CA-CMAC with that of CMAC. Simulation results prove that CA-CMAC converges faster than conventional CMAC.     

The Kernel Recursive Least Squares CMAC with Vector Eligibility

The cerebellar model articulation controller (CMAC) neural network is an associative memory that is biologically inspired by the cerebellum, which is found in the brains of animals. In recent works, the kernel recursive least squares CMAC (KRLS–CMAC) was proposed as a superior alternative to the standard CMAC as it converges faster and does not require tuning of a learning rate parameter. One improvement to the standard CMAC that has been discussed in the literature is eligibility, and vector eligibility. With vector eligibility the CMAC is able to control online motion control problems that it could not previously, stabilize the system much faster, and converge to a more intelligent solution. This paper integrates vector eligibility with the KRLS–CMAC and shows how the combination is advantageous through two simulated control experiments.     

Robust and fast learning for fuzzy cerebellar model articulation controllers.

In this paper, the online learning capability and the robust property for the learning algorithms of cerebellar model articulation controllers (CMAC) are discussed. Both the traditional CMAC and fuzzy CMAC are considered. In the study, we find a way of embeding the idea of M-estimators into the CMAC learning algorithms to provide the robust property against outliers existing in training data. An annealing schedule is also adopted for the learning constant to fulfill robust learning. In the study, we also extend our previous work of adopting the credit assignment idea into CMAC learning to provide fast learning for fuzzy CMAC. From demonstrated examples, it is clearly evident that the proposed algorithm indeed has faster and more robust learning. In our study, we then employ the proposed CMAC for an online learning control scheme used in the literature. In the implementation, we also propose to use a tuning parameter instead of a fixed constant to achieve both online learning and fine-tuning effects. The simulation results indeed show the effectiveness of the proposed approaches.     

Single-input CMAC control system

This paper attempts to propose a single-input cerebellar model articulation controller (CMAC) control system, which contains only one controller implemented by the CMAC. The single-input CMAC control system adopts two learning stages. An off-line learning stage is to enable the output behavior of the CMAC to approximate the control surface of a fuzzy PD-type controller. An on-line learning stage follows to improve the system stability by the modified learning rule. The linear interpolation scheme is also applied to the recall process at the on-line learning stage to ensure better accuracy of the CMAC output. Simulation results show that the single-input CMAC controller is superior to the fuzzy PD-type controller.     

用多个改进低维CMAC神经网络学习多维函数的研究

提出了一种多维非线性函数的多神经网络学习方法。即用变量代换的方法把一个多维非线性函数分解为若干低维函数。用多个改进的低维小脑模型神经网络分别映射这些低维函数。提高了收敛性。减少了存储空间。大大提高了学习精度。且易于实现。给出了大量学习非线性函数的仿真实验，其结果表明。采用这种方法的学习精度比用一个CMAC的学习精度提高l0倍以上。     

Learning convergence in the cerebellar model articulationcontroller

A new way to look at the learning algorithm in the cerebellar model articulation controller (CMAC) proposed by J.S. Albus (1975) is presented. A proof that the CMAC learning always converges with arbitrary accuracy on any set of training data is obtained. An alternative way to implement CMAC based on the insights obtained in the process is proposed. The scheme is tested with a computer simulation for learning the inverse dynamics of a two-link robot arm.     

A genetic algorithm based robust learning credit assignment cerebellar model articulation controller

In this paper, a novel approach of genetic algorithm based robust learning credit assignment cerebellar model articulation controller (GCA-CMAC) is proposed. The cerebellar model articulation controller (CMAC) is a neurological model, which has an attractive property of learning speed. However, the distributions of errors into the addressed hypercubes of CMAC are not proportional to their credibility and may cause unacceptable learning performance. The credit assignment CMAC (CA-CMAC) can solve this problem by using the creditability of hypercubes that the calculated errors are assigned proportional to the inverse of learning times. Afterward, the obtained learning times can be optimized by genetic algorithm (GA) to increase its accuracy. In this paper, the proposed algorithm is to combine credit assignment ideas and GA to provide accurate learning for CMAC. Moreover, we embed the robust learning approach into the GCA-CMAC and dynamically adjust the learning constant for training data with noise or outliers. From simulation results, it shows that the proposed algorithm outperforms other CMACs.     

Contouring accuracy improvement of a piezo-actuated micro motion stage based on fuzzy cerebellar model articulation controller

The motion accuracy and performance of piezoelectric actuators (PEA) are hampered by their inherent hysteresis nonlinearity and time-varying parameters. In order to overcome these drawbacks, an integrated motion control scheme is developed in this paper. In the proposed controller scheme, for each axis of the micro motion stage, a fuzzy CMAC feedforward controller combined with a conventional proportional-integral (PI) feedback controller and a critic-based learning mechanism (FCMAC-CLM) is used to improve the tracking performance and deal with the adverse effects due to hysteresis nonlinearity and external disturbance. In addition, one of the most important issues in the contour following tasks performed by a dual-axis micro motion stage is the contour error reduction. In order to improve the contouring accuracy, a fuzzy CMAC that is capable of real-time learning and self-adjusting is exploited to develop a fuzzy CMAC cross-coupled controller (FCMAC-CCC). Several experiments have been conducted to evaluate the tracking performance and contour accuracy of the proposed approach.