图像雅可比矩阵伪逆估计在视觉伺服中的应用

在一般无标定视觉伺服算法的基础上，提出了一种基于递推最小二乘法估计图像雅可比矩阵伪逆的无标定视觉伺服算法．这种方法无需计算雅可比矩阵伪逆．针对固定目标，文中给出了比例控制算法的稳定性条件．针对运动目标，推导了基于图像雅可比矩阵伪逆估计的伪高斯—牛顿视觉伺服算法和信任区视觉伺服算法．最后,仿真验证了三种视觉伺服算法的有效性．     

机器人运动学反解中的奇异点处理

本文研究机器人运动学反解中的奇异点处理问题，给出了机器人微分运动Ｊａｃｏｂｉａｎ矩阵Ｊ（ｑ）条件数的一个上界，并在此基础上提出机器人关节速度阻尼伪逆解方法中阻尼系数的一种自适应调整方法，该方法可以保证在奇异点附近伪逆解的稳定性。     

仿人机器人冗余度双臂运动学建模与解析

针对人形机器人,提出一种融合左右臂广义逆的组合伪逆方法。该方法能将左右臂的雅克比矩阵广义逆统一为一个伪逆矩阵,从而方便使用传统伪逆方法或优化的伪逆方法处理双臂。首先给出人形机器人双臂的前向运动学模型、左右手臂的基于伪逆模型;然后将这2个单独模型的雅克比矩阵的广义逆统一到一个雅克比矩阵中,得到一个组合和简化的伪逆模型。仿真结果表明,该方案具有可行性和准确性。     

不动点法在飞控系统控制分配中的应用

针对多操纵面战斗机飞行控制系统中的控制分配问题,提出了一种不动点法控制分配策略。方法在伪逆法的分配基础上,采用不动点迭代法进行优化求解,既利用了伪逆法实时性好,易于实现的优点,又解决了伪逆法无饱和约束限制的缺点,并且使可行解的范围扩大到了整个可达集。在Matlab/Simulink中建立控制分配系统的数字仿真模型,并进行计算机仿真,结果表明,策略不仅能得到期望的三轴力矩,而且能很好地分配各个操纵面,使其均工作在约束范围内。结果验证了不动点法的可行性和有效性。     

伪逆BP神经网络在汽车尾气检测中的应用

为了准确、快速、高效地检测汽车尾气中各气体的质量分数,结合传感器阵列和神经网络技术,设计了一种汽车尾气检测系统.为了提高神经网络模型的拟合和预测能力,提出伪逆BP集成神经网络模型:利用伪逆法求得不同BP神经网络集成的最优权重系数,再通过加权平均法集成BP神经网络.利用伪逆BP集成神经网络模型对传感器阵列信号进行回归分析,结果表明:该模型的预测相对误差均小于5％,对比传统Adaboost-BP集成神经网络模型,该模型实现更简单,收敛更快,收敛精度和预测精度更高.     

用阻尼伪逆法控制冗余度机器人的一种新方案

提出一种带次级性能指标优化的阻尼伪逆控制方案,它是通常的阻尼伪逆解的扩展形 式.使用文中给出的方案和阻尼因子、优化因子选择法,冗余度机器人的灵活性有望得到更为有效 地利用.基于奇异值分解(SVD)方法,还提供一种高效并行算法,以利于在实际的机器人控制系统 中实施.仿真结果表明了所提方法比其它相关方法所具有的优点.     

多足机器人足端支反力求解方法比较

多足移动机器人运动过程中足端支反力的求解是一个难题,本文介绍并比较了两种计算方法,即伪逆法和优化方法。伪逆法将等式约束线性化处理并补充到平衡方程中,然后利用伪逆理论求解;优化方法则以各个关节力矩的平方和最小为目标函数,搜寻最优解。本文用伪逆法、基于常规规划理论的优化方法和遗传算法分别实现了某6足机器人足端反力的求解。通过对计算结果进行分析比较,得知:伪逆法效率高,但结果偏于保守,机器人行走时能耗偏高;优化方法得到的结果虽然使机器人行走时能耗更少,但计算速度慢,无法用于实时控制。     

一种基于多极裁剪的广义逆控制分配算法

广义逆控制分配法的分配效率取决于控制效率矩阵伪逆的选择,但是所有伪逆阵都不能实现转矩可达集的完全分配.为解决这一问题,本算法根据飞行包线内转矩向量的特点在转矩可达集空间的不同方向上对伪逆阵进行裁剪,使不同包线区域内的转矩向量经过不同的伪逆阵进行分配,从而来提高广义逆控制分配法的分配效率.为验证改进算法的有效性,利用某飞机的数据进行仿真验证,结果显示该算法可以对转矩可达集内的转矩实施有效的分配,显著提高了分配器的分配效率.     

一种基于零空间的再分配伪逆算法

由于伪逆法不考虑执行器的位置和速率的限制,因此利用伪逆法设计出的分配器会产生操纵面过早进入饱和的现象,从而使分配器对转矩可达集的分配效率较低.在伪逆法的分配基础上利用控制效率矩阵零空间的特性对分配结果进行二次调节,使伪逆解重新落入执行器的控制子空间,从而避免了分配器过早进入饱和的现象.并提高了分配效率.利用某飞机的数据进行仿真验证,结果显示与伪逆法相比,控制性能有明显改善,提高了分配器的分配效率,且可以对转矩可达集内的转矩实施有效的分配.     

基于区间自组织映射的α阶逆系统控制研究*

为了提高α阶逆系统在控制中的鲁棒性,提出了区间自组织映射模型。根据定义的损失函数,利用梯度下降法得到新的模型竞争学习算法,并证明了该竞争算法的收敛性。利用区间自组织映射良好的逼近性能辨识非线性系统的α阶逆系统,并将其串联在原系统之前得到复合伪线性系统。仿真结果表明,该逆系统有较高的精度,逆控制器有良好的跟踪效果和较强的鲁棒性。     

PID神经元网络之权值直接确定法研究

网络权值如何确定,是人工神经网络研究中的一个重要课题。传统PID神经元网络在该问题的研究上,大多数采用误差回传(BP)的思想,通过迭代训练而估算出该网络的连接权值。针对PID神经元网络,对其进行简单巧妙的转化,可提出一种基于矩阵伪逆表述的直接计算权值的方法,从而避免了冗长的迭代训练过程。计算机仿真结果表明,该权值直接确定方法不仅有更快的学习/计算速度,而且能达到更高的计算精度。     

Recursive computation of pseudo-inverse of matrices

Methods for recursive computation of the pseudo-inverse of a sequence of horizontal rectangular matrices in which the successive matrices differ by: (i) one column only, and (ii) one row only, are described. In both the above cases, the pseudo-inverse of the first matrix of the sequence is assumed to be known. The proposed methods avoid the direct inversion of matrices. These new methods can also be used to compute the pseudo-inverse of any arbitrary matrix iteratively. The two distinct cases mentioned above are considered in two parts in this paper. These situations arise in the fixed memory adaptive identification of discrete-data systems.     

广义系统Wiener状态滤波新算法

应用时域上的现代时间序列分析方法，基于ARMA新息模型和白噪声估计理论，由一种新的非递推最优状态估值器的递推变形，提出了广义系统Wiener状态滤波的一种新算法，它可统一处理滤波、平滑和预报问题，且具有渐近稳定性。同某些算法相比，它避免了求解Riccati方程和Diophantine方程，且避免了计算伪逆，因而减小了计算负担。仿真例子说明了其有效性。     

一种新型神经网络模型在数据挖掘中的应用

针对传统神经网络存在的问题,提出一种新型神经网络模型,即Legendre前向神经网络模型,进而在该模型的基础上提出一种基于伪逆的权值直接确定法(即一步确定法),并将其应用到数据挖掘中.仿真结果显示该方法不仅能更快地收敛,而且可以达到更高的工作精度.     

Continuous-discrete gain transformation methods for linear feedback control

The problem of transforming between continuoustime state variable feedback gains and equivalent discrete gains suitable for digital implementation is considered. The concepts of state and control equivalence yield two simple transformation rules, a pseudo-inverse method and an average gain method, respectively. As the sampling interval δ→0, these methods are contrasted with existing Taylor series based approaches. The new transformation rules are also studied numerically using a ship course-keeping example. Transformed optimal continuous gains are compared with optimal discrete gains over a wide range of sampling intervals.     

Numerical concepts for rate-independent single crystal plasticity

We provide a new time integration algorithm for rate-independent single crystal plasticity. The algorithm is based on an augmented Lagrangian formulation of the principle of maximum plastic dissipation. It is a synthesis of Lagrangian multiplier and penalty formulations avoiding their drawbacks (active set search, pseudo-inverse and approximate fulfillment of the yield condition respectively). The algorithm is physically motivated and especially provides a clear concept how to determine the set of active slip systems. The method is stable and efficient, several representative numerical examples are shown.     

Inverse mapping of continuous functions using local and globalinformation

This paper presents a method for solving inverse mapping of a continuous function learned by a multilayer feedforward mapping network. The method is based on the iterative update of input vector toward a solution, while escaping from local minima. The input vector update is determined by the pseudo-inverse of the gradient of Lyapunov function, and, should an optimal solution be searched for, the projection of the gradient of a performance index on the null space of the gradient of Lyapunov function. The update rule is allowed to detect an input vector approaching local minima through a phenomenon called "update explosion". At or near local minima, the input vector is guided by an escape trajectory generated based on "global information", where global information is referred to here as predefined or known information on forward mapping; or the input vector is relocated to a new position based on the probability density function (PDF) constructed over the input vector space by Parzen estimate. The constructed PDF reflects the history of local minima detected during the search process, and represents the probability that a particular input vector can lead to a solution based on the update rule. The proposed method has a substantial advantage in computational complexity as well as convergence property over the conventional methods based on Jacobian pseudo-inverse or Jacobian transpose.     

Camera-robot transform for vision-guided tracking in a manufacturing work cell

The problem of camera calibration from the perspective of hand-eye integration (henceforth referred to as the Camera-Robot (CR) problem), is addressed in this paper. Mapping results obtained from a least-squares fit using pseudo-inverse technique and a three layer neural network are compared. The calibration matrix is developed to map the image coordinates of an IRI D256 vision processor equipped with a CCD camera directly on to the coordinates for an IBM 7540 SCARA manipulator. One transformation is obtained by performing a least-squares fit using pseudo-inverse technique on a set of one hundred data points which relates two-dimensional (2D) image coordinates to corresponding twodimensional robot coordinates. The CR problem is also approached by using the same data points on a neural network. The results not only demonstrate the ability of neural networks to ‘learn’ the transformation to a reasonable accuracy, but also from the basis for a relatively simple method of adaptive self-calibration of robot-vision systems. In a broader sense, the proposed method can be used to calibrate a variety of robotic sensors that are typically used in a flexible manufacturing environment.     

A multi-objective approach for the motion planning of redundant manipulators

Kinematic redundancy occurs when a manipulator possesses more degrees of freedom than those required to execute a given task. Several kinematic techniques for redundant manipulators control the gripper through the pseudo-inverse of the Jacobian, but lead to a kind of chaotic inner motion with unpredictable arm configurations. Such algorithms are not easy to adapt to optimization schemes and, moreover, often there are multiple optimization objectives that can conflict between them. Unlike single optimization, where one attempts to find the best solution, in multi-objective optimization there is no single solution that is optimum with respect to all indices. Therefore, trajectory planning of redundant robots remains an important area of research and more efficient optimization algorithms are needed. This paper presents a new technique to solve the inverse kinematics of redundant manipulators, using a multi-objective genetic algorithm. This scheme combines the closed-loop pseudo-inverse method with a multi-objective genetic algorithm to control the joint positions. Simulations for manipulators with three or four rotational joints, considering the optimization of two objectives in a workspace without and with obstacles are developed. The results reveal that it is possible to choose several solutions from the Pareto optimal front according to the importance of each individual objective.