Data fitting with a spline using a real-coded genetic algorithm

To obtain a good approximation for data fitting with a spline, frequently we have to deal with knots as variables. The problem to be solved then becomes a continuous nonlinear and multivariate optimization problem with many local optima. Therefore, it is difficult to obtain the global optimum. In this paper, we propose a method for solving this problem by using a real-coded genetic algorithm. Our method can treat not only data with a smooth underlying function, but also data with an underlying function having discontinuous points and/or cusps. We search for the best model among candidate models by using the Bayes Information Criterion (BIC). With this, we can appropriately determine the number and locations of knots automatically and simultaneously. Five examples of data fitting are given to show the performance of our method.     

An integrated modelling environment for the embodiment of mechanical systems

The efficient, economical and often elegant transformation from a design schema to an embodied design solution containing real components is essential for the commercial success of a product. More often the ability of the designer to explore an optimum solution is severely frustrated by the analytically intensive and time-consuming aspects of embodying a concept. This paper presents a computer based system-modelling tool that is to be used by designers during the transformation of a concept to an embodied solution particularly to deal with standard components. This environment provides for the representation of conceptual schemas and subsequent embodiment from computer based component selection modules. These selection modules encompass the full range of electronic representations for standard mechanical components. The strategy for representing a mechanical system, its included components and their associated representations is discussed. The process of constructing a system model, specifying the desired performance characteristics and system resolution are also described.     

Controlling mechanical systems with backlash—a survey

Backlash is one of the most important non-linearities that limit the performance of speed and position control in industrial, robotics, automotive, automation and other applications. The control of systems with backlash has been the subject of study since the 1940s. This survey reveals that surprisingly few control innovations have been presented since the early path breaking papers that introduced the describing function analysis of systems with backlash. Promising developments are however taking place using adaptive and non-linear control strategies.     

Analysis of exclusively kinetic two-link underactuated mechanical systems

Analysis of exclusively kinetic two-link underactuated mechanical systems is undertaken in this paper. It is first shown that such systems are not full-state feedback linearizable around any equilibrium point. Also, the equilibrium points for which the system is small-time locally controllable (STLC) is at most a one-dimensional submanifold. A concept less restrictive than STLC, termed the small-time local output controllability (STLOC) is introduced, the satisfaction of which guarantees that a chosen configuration output can be controlled at its desired value. It is shown that the class of systems considered is STLOC, if the inertial coupling between the input and output is nonzero. Also, in such a case, the system is nonminimum phase. An example section illustrates all the results presented.     

Wiener models of direction-dependent dynamic systems

Analysis of systems with direction-dependent dynamics is currently limited to cases in which the dynamics in the two directions of the output are first order; results for such systems have been published for both pseudo-random maximum-length binary (MLB) and inverse-repeat maximum-length binary (IR-MLB) inputs. These relatively limited analytical results make it useful to examine alternative ways of modelling such systems and in this paper, Wiener models are considered for this purpose. Methods for optimising the Wiener model parameters by matching the system and model cross-correlation functions, outputs, and discrete Fourier transforms of the outputs are considered, and the results are compared. These methods are also applied to a first-order direction-dependent system with a maximum-length ternary (MLT) input, for which no analytical results are currently available, and to a second-order system with an IR-MLB input.     

Vote counting measures for ensemble classifiers

Various measures, such as Margin and Bias/Variance, have been proposed with the aim of gaining a better understanding of why Multiple Classifier Systems (MCS) perform as well as they do. While these measures provide different perspectives for MCS analysis, it is not clear how to use them for MCS design. In this paper a different measure based on a spectral representation is proposed for two-class problems. It incorporates terms representing positive and negative correlation of pairs of training patterns with respect to class labels. Experiments employing MLP base classifiers, in which parameters are fixed but systematically varied, demonstrate the sensitivity of the proposed measure to base classifier complexity.     

Convolution profiles for right inversion of multivariable non-minimum phase discrete-time systems

The problem of the non-causal inversion of linear multivariable discrete-time systems is analyzed in the geometric approach framework and is solved through the computation of convolution profiles which guarantee perfect tracking under the assumption of infinite-length preaction and postaction time intervals. It is shown how the shape of the convolution profiles is related to both the relative degree and the invariant zeros of the plant. A computational setting for the convolution profiles is derived by means of the standard geometric approach tools. Feasibility constraints are also taken into account. A possible implementation scheme, based on a finite impulse response system acting on a stabilized control loop, is provided.     

Fault-accommodation with intelligent sensors

Optimal sensor fault-accommodation is considered for faults modelled by an increase in measurement noise. This noise is taken to be bounded and its probabilistic properties unknown. It is assumed that intelligent (e.g. self-validating) instrumentation is in use and estimates of the noise bounds are available. The fault-tolerant controller is designed to optimize a noise rejection and a nominal reference tracking index and leads to a mixed norm minimization problem (l₁/l₂). We exploit known results and a particular feedback configuration to show that it is possible to optimize simultaneously without a trade-off the two performance indices. The results are applied to systems where the presence of auxiliary measurements allows for an optimal fault-accommodation strategy. Using properties of the optimal solution, we define a factorization for the optimal controller alternative to the Youla parametrization, leading to an algorithm which is optimal, transparent and efficient.