Skinner 操作条件反射的一种仿生学习算法与机器人控制

针对两轮自平衡机器人的运动平衡控制问题,提出了基于Skinner 操作条件反射理论的BP 神经网络 与资格迹相结合的仿生自主学习算法作为两轮机器人的学习机制．该算法利用资格迹能解决延迟影响、加快学习速 度和提高可靠性的特点,将其与BP 神经网络相结合构成复合学习算法,能够预测机器人将要获得的行为评价函数, 并依据概率取向机制以一定的概率选择最大评价值对应的最优行为,从而使机器人能够在未知环境下通过与环境的 交互、学习和训练,获得像人或动物一样的自主学习技能,实现对两轮机器人的运动平衡控制．最后,分别用基于 Skinner 操作条件反射理论的BP 算法和BP 资格迹复合算法对两轮机器人做了仿真实验并进行了比较．结果表明, 基于Skinner 操作条件反射理论的BP 资格迹复合仿生自主学习算法的学习机制能够使机器人获得良好的动态性能 和较快的学习速度,体现了机器人较强的自主学习技能和平衡控制能力．     

Statistic learning-based defect detection for twill fabrics

Template matching methods have been widely utilized to detect fabric defects in textile quality control. In this paper, a novel approach is proposed to design a flexible classifier for distinguishing flaws from twill fabrics by statistically learning from the normal fabric texture. Statistical information of natural and normal texture of the fabric can be extracted via collecting and analyzing the gray image. On the basis of this, both judging threshold and template are acquired and updated adaptively in real-time according to the real textures of fabric, which promises more flexibility and universality. The algorithms are experimented with images of fault free and faulty textile samples.     

模糊控制理论及其应用的发展

对模糊控制进行了综述,介绍了模糊控制理论、模糊控制理论的发展、研究以及应用情况。指出模糊控制是一种形式简单、内涵丰富、易于实现的智能控制方法。     

多变量强耦合时变系统的PID神经网络控制

介绍了一种新的神经网络———PID神经网络及其多变量强耦合时变控制系统。文中给出了网络的结构和算法,分析了时变对象的特点,对一组二变量强耦合时变系统进行了实时仿真。仿真结果显示:PID神经网络对多变量强耦合时变对象具有良好的解耦性能和自学习控制特性。     

一种在线自学习的移动机器人模糊导航方法

针对现有移动机器人模糊导航对未知不确定环境缺乏自适应性的缺点,提出了一种具备在线自学习能力的模糊导航方法.通过设计模糊规则并确定动作先验值,完成初始模糊导航系统的构建.利用Q强化学习算法对模糊规则中各行为的值函数进行在线增量学习,实现模糊决策的逐步求精.仿真实验表明,移动机器人导航系统能够在运行过程中不断调整导航策略,实现对未知不确定环境的自适应.同时由于导航先验知识的引入,有效地克服了强化学习初始阶段进行盲目搜索导致的学习速率低、收敛速度慢的缺点,实现了移动机器人可靠导航.     

专家控制方法在浮选过程中的应用

针对浮选过程被控对象复杂、数学模型不确定以及控制要求高等特点，提出一种用于阳离子浮选过程的专家控制方法，使系统具备一定的自学习和自组织的能力．所采用的规则表示法能自然、完整地表示浮选过程的知识，又便于知识库的维护．实际应用结果表明，所提出的专家控制方法能够满足浮选过程控制的要求．     

Multivariable temperature measurement and control system of large-scaled vertical quench furnace based on temperature field

A temperature control system of 31m vertical forced air-circulation quench furnace is proposed,which is a kind of equipment critical for thermal treamlent of aluminum alloy components that are widely used in aerospace industry. For the effective operation of the furnace, it is essential to analyze the radial temperature distribution of the furnace. A set of thermodynamic balance equations modeling is established firstly. By ufilizmg the numerical analysis result to modify the temperature measurements, the control accuracy and precision of the temperature are truly guaranteed. Furthermore, the multivariable decouplmg self-learning PID control algorithm based on the characteristics of strong coupling between the multi-zones in the large-scaled furnace is implemented to ensure the true homogeneity of the axial temperature distribution. Finally, the redundant structure composed of industrial control computers and touch panels leads to great improvement of system reliability.     

HIGH-ORDER CONTRASTS FOR SELF-ADAPTIVE SOURCE SEPARATION

This paper is concerned with the problem of separating independent non-Gaussian sources. This is done by adaptively maximizing a contrast function based on fourth-order cumulants of the (mixed) obser$softhyphen;vations. The first cla ss of solutions involves a first stage where the signal vector is adaptively whitened. In order to implement in the second stage the proper separating task, new contrast functions are proposed, especially when all the source kurtosises have the same sign. These contrasts involve only self-cumulants of the outputs. The second class of solutions requires a single separating stage. However, the associated contrasts involve cross-cumulants in addition to self-cumulants. They essentially apply to correlated ve ctors with normalized powers (rather than to white vectors). The resulting adaptive one-stage and two-stage systems achieve satisfactory separation performance independently of the statistics of sources and of the kind of linear mixture.     

基于远程通信和学习功能的电厂火检监控系统

介绍了一个已经成功运行的基于远程通信和学习功能的火检监控系统.该系统结合通信技术、计算机技术、自动控制技术以及信息处理技术,将原来现场检测环节中较为庞大的信号线路简化到只有2根通信线缆,并增加了自学习和远程编程监控功能,不仅优化了集散控制系统(DCS),提高了数据的准确度和整个系统的可靠度,也使现场检测设备的维护更加容易和简便.     

基于前馈模糊自学习控制的强迫周期振动主动控制

运用前馈模糊自学习方法对强迫周期振 主动控制进行研究。指出模糊自学习算法在应用上存在的问题,并进行改进;针对不同规则数、不同学习次数进行控制仿真研究,揭示规则数和学习次数对控制性能的影响;对模糊自学习系统的鲁棒性进行探讨;结果表明前馈模糊自学习系统对强迫周期振动具有良好的抑制效果。