基于混沌遗传算法的柔性机械手滑模控制器优化设计

针对柔性机械手动力学方程的非最小相位特点,本文提出一种柔性机械手的终端滑模控制方法,将关节电机转角和柔性模态变量的线性组合定义为柔性机械手系统的输出.通过输入输出线性化,将系统分解为输入输出子系统和零动态子系统.设计终端滑模控制策略,使输入输出子系统在有限时间收敛到零;利用混沌遗传算法优化控制器的设计参数,使零动态了系统在甲衡点附近渐近稳定,从而保证整个系统的渐近稳定.本文提出的方法设计过程简单,易于实现.仿真结果证明了设计的有效性.     

The finite zeros of diminished systems

The assignment and preservation of the finite zeros of multivariable systems is considered when selected inputs and outputs are combined to form now input-output transferences. Conditions are given under which all of the finite zeros of the newly formed ‘ diminished system ’ can be assigned arbitrarily. The results have some features in common with the more widely known conditions under which all of the poles of a system can be relocated using constant output feedback. One consequence of the results is that further insight can be obtained into a well known zero assignment algorithm due to Kouvaritakis and MacFarlane (1976 a, b). The results have relevance to the problem of ‘ squaring down ’ a non-square system.     

Transformation of finite degree discrete Volterra systems withcross products to finite input-output forms

In this paper the problem of finite input-output representation is solved for a class of finite degree discrete Volterra systems containing cross products among inputs and outputs, via the so-called δϵ-operators. A corresponding algorithm is supplied. An application concerning the bilinear systems is also given     

参数不确定柔性机械手的终端滑模控制

针对参数不确定双臂柔性机械手系统, 提出一种基于遗传算法的终端滑模控制方法, 以实现其末端控制.基于输出重定义方法, 通过输入输出线性化, 将系统分解为输入输出子系统和内部子系统. 设计终端滑模控制策略,使输入输出子系统有限时间内收敛到零, 内部子系统变为零动态子系统; 采用遗传算法优化零动态子系统参数, 使其在平衡点附近渐近稳定. 根据Lyapunov稳定性理论算出末端输出位移的误差范围. 仿真结果证明该方法有效性.     

柔性机械手的鲁棒控制器设计

针对柔性机械手的动力学方程具有非最小相位的特点,运用重新定义的柔性机械手系统的输出,通过输入输出线性化,将系统分解为输入输出子系统和零动态子系统．考虑到柔性机械手系统存在的不确定性,设计终端滑模控制器,使输入输出子系统在有限时间内收敛到零．最优组合输出系数采用混沌遗传算法优化,以保证零动态子系统在平衡点附近渐近稳定,从而保证整个系统渐近稳定．仿真结果证明了设计方法的有效性．     

Residual models and stochastic realization in state-space identification

This paper presents theory and algorithms for validation in system identification of state-space models from finite input-output sequences in a subspace model identification framework. Our formulation includes the problem of rank-deficient residual covariance matrices, a case which is encountered in applications with mixed stochastic-deterministic input-output properties as well as for cases where outputs are linearly dependent. Similar to the case of prediction-error identification, it is shown that the resulting model can be decomposed into an input-output model and a stochastic innovations model. Using the Riccati equation, we have designed a procedure to provide a reduced-order stochastic model that is minimal with respect to system order as well as the number of stochastic inputs thereby avoiding several problems appearing in standard application of stochastic realization to the model validation problem.     

Adaptive control for infinite-dimensional systems

This paper deals with the design of a model reference adaptive controller for a distributed parameter system generated by a self-adjoint operator. The space of inputs and the space of outputs are both one dimensional. After introducing input-output representations, an adaptive control system is designed for the system with unknown input function and unknown output function. The input-output equations can be expressed using filtered values generated from the inputs and outputs     

A linear-quadratic-Gaussian control algorithm for sulphide ore grinding

The paper suggests an algorithm for the multivariable control of an industrial sulphide ore grinding plant. The algorithm is based on the linear-quadratic-Gaussian control theory with a moving time horizon. The dynamic process model with two inputs and two outputs is obtained by fitting simple low order linear time-invariant models to measured input-output pairs. Characteristic to the model are long time delays. The model is brought to the state space form, in which the time delays are described by additional state components. The inputs are replaced by input increments for the purpose of avoiding imbalances of stationary values and of assigning quadratic costs to input changes. Because of disturbances both in the process and in the measuring equipment, a Kalman filter is used for state estimation. The control algorithm was tested both by simulation runs and by implementation in the control of the plant. The test runs showed that the algorithm can be tuned rather easily, and satisfactory results were obtained. The steady-state interactions were almost completely compensated and the transient interactions were within acceptable limits.     

Self-tuning controller for drug delivery systems

A new adaptive control algorithm, the Control Advance Moving Average Controller (CAMAC), was implemented in a drug delivery system to control mean arterial pressure by the infusion of sodium nitroprusside. The CAMAC, which is designed to control non-minimum-phase systems with unknown or varying dead times, is presented here. The control law minimizes a cost function at a time advance which is equal to or greater than the dead time, based upon an assumption that is made on future inputs. By properly tuning the control advance and implementing an estimator with an extended B parameter matrix, the CAMAC can effectively control non-minimum-phase systems with unknown or varying deadtimes. By allowing the control advance to be selected separately (though not independently) for each output in a multivariable system, the CAMAC can decrease the output error for systems where the deadtime for individual input-output pairs may differ. In animal experiments the controller effectively lowered the mean arterial pressure and, despite the intentional addition of large disturbances by the administration of other vasoactive agents, the CAMAC maintained the mean arterial pressure near the setpoint. An example of multivariable control of mean arterial pressure and cardiac output by infusion of sodium nitroprusside and dobutamine is also presented.     

Interconnected linear multivariable systems with delays: System properties and input-output stability

We consider an arbitrary interconnection of linear, time-invariant, distributed, multivariable subsystems; each subsystem is represented either by a strictly proper rational transfer function or by a finite sum of pure delay terms. Such models occur in transportation systems, in chemical processes, in production systems, in communication systems, and in models of boilers in nuclear power plants. Allowing inputs to be applied to any interconnection point, we show that, except for a special case, the input-output transfer function of such a system is in a class extensively stuided in the literature. Next we obtain necessary and sufficient conditions for input-output stability. An example illustrates the application of the test.     

On input-dependent system identification by Monte Carlo approach

The problem of a weighted least-squares approximation of a memoryless system within a given class of models is discussed, and an identification algorithm leading to the best model when only input-output data are accessible is derived using a concept of random choice of inputs.     

Equivalence relations between learnability,output-dissipativity and strict positive realness

This paper was originally motivated by aiming at explicating why a simple iterative learning control scheme for complicated robot dynamics with strong non-linearities works well in acquiring any given desired motion over a finite or infinite time duration or any periodic motion. To gain a physical insight into the problem, a class of linear dynamical systems with specified input and output of the same dimension is treated by defining two properties: output-dissipativity and learnability. It is then shown that the former implies the latter and furthermore, for a class of linear systems with single input and single output, they are equivalent to each other and each of them is also equivalent to strict positive realness of input-output transfer function. For a class of MIMO (multiple inputs and multiple outputs) systems, it is possible to prove that each of these properties is equivalent to strict positive realness of the input-output transfer function matrix if it is strictly proper or otherwise its direct term from input to output satisfies an extra condtion.     

On the identification of polynomial input-output differential systems

A least-squares parameter identification technique is formulated for a class of deterministic nonlinear systems modeled by polynomial input-output differential equations. The basis of the technique is Shinbrot's method of moment functionals using trigonometric modulating functions. Given the input-output data over a single finite time interval for a one-shot estimate, or over a sequence of finite time intervals for sequential least squares, the underlying computations utilize a fast fourier transform algorithm on polynomials of the data without the need for estimating unknown initial or boundary conditions at the start of each finite time interval.     

Input-output representations of spectral systems and adaptive controls

In this paper two types of input-output representations (static ones and dynamic ones) are presented for the class of distributed parameter systems generated by spectral operators. The input-output equations can be expressed using filtered values generated from the inputs and outputs. Using the input-output representations, the design problem of an adaptive control system is also investigated for systems with unknown input and output distribution functions.     

Analysis and design of hierarchical fuzzy systems

In this letter, the hierarchical fuzzy systems are analyzed and designed. In the analysis part, we prove that the hierarchical fuzzy systems are universal approximators and analyze the sensitivity of the fuzzy system output with respect to small perturbations in its inputs. In the design part, we derive a gradient descent algorithm for tuning the parameters of the hierarchical fuzzy system to match the input-output pairs. The algorithm is simulated for two examples and the results show that the algorithm is effective and the hierarchical structure gives good approximation accuracy     

Input selection for nonlinear regression models

A simple and effective method for the selection of significant inputs in nonlinear regression models is proposed. Given a set of input-output data and an initial superset of potential inputs, the relevant inputs are selected by checking whether after deleting a particular input, the data set is still consistent with the basic property of a function. In order to be able to handle real-valued and noisy data in a sensible manner, fuzzy clustering is first applied. The obtained clusters are compared by using a similarity measure in order to find inconsistencies within the data. Several examples using simulated and real-world data sets are presented to demonstrate the effectiveness of the algorithm.     

输入输出均为时变函数的过程神经网络及应用

为了解决实际系统中输入、输出经常是时变连续函数的问题,提出了一类基于基函数展开的过程神经元网络模型.该模型利用过程神经元网络所具有的对时间变量的非线性映射能力,实现系统的输入、输出之间的连续映射关系.另外,还给出了一种学习算法.为了简化计算,选择正交函数作为基函数,并以油藏开发仿真为例,验证了模型和算法的有效性.     

Algorithms for deterministic balanced subspace identification

New algorithms for identification of a balanced state space representation are proposed. They are based on a procedure for the estimation of impulse response and sequential zero input responses directly from data. The proposed algorithms are more efficient than the existing alternatives that compute the whole Hankel matrix of Markov parameters. It is shown that the computations can be performed on Hankel matrices of the input-output data of various dimensions. By choosing wider matrices, we need persistency of excitation of smaller order. Moreover, this leads to computational savings and improved statistical accuracy when the data is noisy. Using a finite amount of input-output data, the existing algorithms compute finite time balanced representation and the identified models have a lower bound on the distance to an exact balanced representation. The proposed algorithm can approximate arbitrarily closely an exact balanced representation. Moreover, the finite time balancing parameter can be selected automatically by monitoring the decay of the impulse response. We show what is the optimal in terms of minimal identifiability condition partition of the data into “past” and “future”.     

输入为Markov序列时的多用户检测算法

考虑当用户序列存在时间相关性时的多用户检测，并假设这种相关性可以用Markov链描述，在传统的线性最大似然检测器中嵌入一个隐Markov模型估计过程。因为输入序列是Markov链，检测器的输出可以看成是被噪声污染的Markov序列，Markov模型估计子用于估计用户序列及其转移概率，而估计得到的用户序列用来更新检测器的估计。因此，检测器和用户序列可以通过迭代的方式求解。仿真结果显示本文算法能充分利用信道输入的时间相关性．效果优于传统的最大似然线性检测器。