Time-harmonic problems arise in many important applications, such as eddy current optimally controlled electromagnetic problems. Eddy current modelling can also be used in non-destructive testings of conducting materials. Using a truncated Fourier series to approximate the solution, for linear problems the equation for different frequencies separate, so it suffices to study solution methods for the problem for a single frequency. The arising discretized system takes a two-by-two or four-by-four block matrix form. Since the problems are in general three-dimensional in space and hence of very large scale, one must use an iterative solution method. It is then crucial to construct efficient preconditioners. It is shown that an earlier used preconditioner for optimal control problems is applicable here also and leads to very tight eigenvalue bounds and hence very fast convergence such as for a Krylov subspace iterative solution method. A comparison is done with an earlier used block diagonal preconditioner.
10 December 2015 in International Journal for Numerical Methods in Engineering
This paper considers the differential game approach to the leaderless formation control problem of a linear dynamical multi-robot system with directed graph topology. The formation requirement is formulated in terms of a finite-horizon quadratic cost function for each robot through the use of graph Laplacian and optimal control theory. The motion equations of robots and formation cost functions are the state equations and cost functions of the differential game. In the differential game approach, each robot can select its neighbours and implement distributed controllers; however, there is no guarantee that the formation control exists. The main contribution of this paper is a differential game formulation to formation control admitting a unique Nash equilibrium that ensures the existence of the formation control.
While repairing industrial machines or vehicles, recognition of components is a critical and time-consuming task for a human. In this paper, we propose to automatize this task. We start with a Principal Component Analysis (PCA), which fits the scanned point cloud with an ellipsoid by computing the eigenvalues and eigenvectors of a 3-by-3 covariant matrix. In case there is a dominant eigenvalue, the point cloud is decomposed into two clusters to which the PCA is applied recursively. In case the matching is not unique, we continue to distinguish among several candidates. We decompose the point cloud into planar and cylindrical primitives and assign mutual features such as distance or angle to them. Finally, we refine the matching by comparing the matrices of mutual features of the primitives. This is a more computationally demanding but very robust method. We demonstrate the efficiency and robustness of the proposed methodology on a collection of 29 real scans and a database of 389 STL (Standard Triangle Language) models. As many as 27 scans are uniquely matched to their counterparts from the database, while in the remaining two cases, there is only one additional candidate besides the correct model. The overall computational time is about 10 min in MATLAB.