The Experts below are selected from a list of 84180 Experts worldwide ranked by ideXlab platform
D. Gorinevsky - One of the best experts on this subject based on the ideXlab platform.
-
An approach to parametric nonlinear least square optimization and application to task-level learning control
IEEE Transactions on Automatic Control, 1997Co-Authors: D. GorinevskyAbstract:This paper considers a parametric nonlinear least square (NLS) optimization problem. Unlike a classical NLS problem statement, we assume that a nonlinear optimized system depends on two arguments: an input Vector and a Parameter Vector. The input Vector can be modified to optimize the system, while the Parameter Vector changes from one optimization iteration to another and is not controlled. The optimization process goal is to find a dependence of the optimal input Vector on the Parameter Vector, where the optimal input Vector minimizes a quadratic performance index. The paper proposes an extension of the Levenberg-Marquardt algorithm for a numerical solution of the formulated problem. The proposed algorithm approximates the nonlinear system in a vicinity of the optimum by expanding it into a series of Parameter Vector functions, affine in the input Vector. In particular, a radial basis function network expansion is considered. The convergence proof for the algorithm is presented. The proposed approach is applied to task-level learning control of a two-link flexible arm. Each evaluation of the system in the optimization process means completing a controlled motion of the arm.
-
An algorithm for on-line parametric nonlinear least square optimization
Proceedings of 1994 33rd IEEE Conference on Decision and Control, 1994Co-Authors: D. GorinevskyAbstract:This paper considers a parametric nonlinear least square (NLS) optimization problem. In extension of a classical NLS problem statement, it is assumed that the nonlinear optimized system depends on two arguments: an input Vector and a Parameter Vector. The input Vector can be modified to optimize the system, while the Parameter Vector changes from one optimization iteration to another and is not controlled. The optimization process goal is to find a dependence of the optimal input Vector on the Parameter Vector, where the optimal input Vector minimizes a quadratic performance index. The paper proposes an extension of the Levenberg-Marquardt algorithm for numerical solution of the formulated problem. The proposed algorithm approximates the nonlinear system by an expansion into a series of the Parameter Vector functions, affine in the input Vector. In particular, a radial basis function network expansion is considered. The convergence proof for the algorithm is presented.
Nanning Zheng - One of the best experts on this subject based on the ideXlab platform.
-
ACCV (4) - Large-Scale bundle adjustment by Parameter Vector partition
Computer Vision – ACCV 2012, 2013Co-Authors: Shanmin Pang, Le Wang, Jianrue Xue, Nanning ZhengAbstract:We propose an efficient parallel bundle adjustment (BA) algorithm to refine 3D reconstruction of the large-scale structure from motion (SfM) problem, which uses image collections from Internet. Different from the latest BA techniques that improve efficiency by optimizing the reprojection error function with Conjugate Gradient (CG) methods, we employ the Parameter Vector partition strategy. More specifically, we partition the whole BA Parameter Vector into a set of individual sub-Vectors via normalized cut (Ncut). Correspondingly, the solution of the BA problem can be obtained by minimizing subproblems on these sub-Vector spaces. Our approach is approximately parallel, and there is no need to solve the large-scale linear equation of the BA problem. Experiments carried out on a low-end computer with 4GB RAM demonstrate the efficiency and accuracy of the proposed algorithm.
-
large scale bundle adjustment by Parameter Vector partition
Asian Conference on Computer Vision, 2012Co-Authors: Shanmin Pang, Le Wang, Nanning ZhengAbstract:We propose an efficient parallel bundle adjustment (BA) algorithm to refine 3D reconstruction of the large-scale structure from motion (SfM) problem, which uses image collections from Internet. Different from the latest BA techniques that improve efficiency by optimizing the reprojection error function with Conjugate Gradient (CG) methods, we employ the Parameter Vector partition strategy. More specifically, we partition the whole BA Parameter Vector into a set of individual sub-Vectors via normalized cut (Ncut). Correspondingly, the solution of the BA problem can be obtained by minimizing subproblems on these sub-Vector spaces. Our approach is approximately parallel, and there is no need to solve the large-scale linear equation of the BA problem. Experiments carried out on a low-end computer with 4GB RAM demonstrate the efficiency and accuracy of the proposed algorithm.
Jie Huang - One of the best experts on this subject based on the ideXlab platform.
-
Global adaptive output regulation for a class of nonlinear systems with iISS inverse dynamics using output feedback
Automatica, 2013Co-Authors: Jie Huang, Zhong-ping JiangAbstract:Abstract This paper generalizes our recent results on global robust output regulation of output feedback systems with integral input-to-state stable (iISS) inverse dynamics in two aspects. First, we consider the general relative degree case instead of the unity relative degree case. For this purpose, the one step approach proposed previously has to be generalized to a recursive approach. Second, we allow the exosystem to be uncertain. For this purpose, we need to introduce an adaptive control technique to estimate the unknown Parameter Vector in the exosystem. A convergence analysis of the estimated Parameter Vector will also be given.
-
Global output regulation for output feedback systems with an uncertain exosystem and its application
International Journal of Robust and Nonlinear Control, 2010Co-Authors: Jie HuangAbstract:This paper presents the solvability conditions for the global robust output regulation problem for a class of output feedback systems with an uncertain exosystem by using output feedback control. An adaptive control technique is used to handle the unknown Parameter Vector in the exosystem. It is shown that this unknown Parameter Vector can be exactly estimated asymptotically if a controller containing a minimal internal model is employed. The effectiveness of our approach has been illustrated by an asymptotic tracking problem of a generalized fourth-order Lorenz system. Copyright © 2009 John Wiley & Sons, Ltd.
-
output regulation for output feedback systems with an uncertain exosystem
Conference on Decision and Control, 2009Co-Authors: Jie HuangAbstract:This paper studies the global robust output regulation problem for a class of output feedback systems subject to an uncertain exosystem by using output feedback control. An adaptive control technique is used to handle the unknown Parameter Vector in the exosystem. It is shown that this unknown Parameter Vector can be exactly estimated asymptotically if the controller incorporates a minimal internal model. The effectiveness of our approach has been illustrated by an asymptotic tracking problem associated with a generalized fourth-order Lorenz system.
-
brief paper Parameter convergence and minimal internal model with an adaptive output regulation problem
Automatica, 2009Co-Authors: Zhiyong Chen, Jie HuangAbstract:The Parameter convergence of nonlinear adaptive control systems is an important yet not well addressed issue. In this paper, using the global robust output regulation problem of output feedback systems with unknown exosystems as a platform, we will show that the well known persistency of excitation (PE) condition still guarantees the convergence of the estimated Parameter Vector to the true Parameter Vector. Moreover, the PE condition will be satisfied if the internal model is minimal in certain sense.
-
CDC - Output regulation for output feedback systems with an uncertain exosystem
Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference, 2009Co-Authors: Jie HuangAbstract:This paper studies the global robust output regulation problem for a class of output feedback systems subject to an uncertain exosystem by using output feedback control. An adaptive control technique is used to handle the unknown Parameter Vector in the exosystem. It is shown that this unknown Parameter Vector can be exactly estimated asymptotically if the controller incorporates a minimal internal model. The effectiveness of our approach has been illustrated by an asymptotic tracking problem associated with a generalized fourth-order Lorenz system.
Y Selen - One of the best experts on this subject based on the ideXlab platform.
-
linear regression with a sparse Parameter Vector
IEEE Transactions on Signal Processing, 2007Co-Authors: Erik G Larsson, Y SelenAbstract:We consider linear regression under a model where the Parameter Vector is known to be sparse. Using a Bayesian framework, we derive the minimum mean-square error (MMSE) estimate of the Parameter Vector and a computationally efficient approximation of it. We also derive an empirical-Bayesian version of the estimator, which does not need any a priori information, nor does it need the selection of any user Parameters. As a byproduct, we obtain a powerful model ("basis") selection tool for sparse models. The performance and robustness of our new estimators are illustrated via numerical examples
-
linear regression with a sparse Parameter Vector
International Conference on Acoustics Speech and Signal Processing, 2006Co-Authors: Erik G Larsson, Y SelenAbstract:We consider linear regression under a model where the Parameter Vector is known to be sparse. Using a Bayesian framework, we derive a computationally efficient approximation to the minimum mean-square error (MMSE) estimate of the Parameter Vector. The performance of the so-obtained estimate is illustrated via numerical examples.
-
ICASSP (3) - Linear Regression with a Sparse Parameter Vector
2006 IEEE International Conference on Acoustics Speed and Signal Processing Proceedings, 1Co-Authors: Erik G Larsson, Y SelenAbstract:We consider linear regression under a model where the Parameter Vector is known to be sparse. Using a Bayesian framework, we derive a computationally efficient approximation to the minimum mean-square error (MMSE) estimate of the Parameter Vector. The performance of the so-obtained estimate is illustrated via numerical examples.
Shanmin Pang - One of the best experts on this subject based on the ideXlab platform.
-
ACCV (4) - Large-Scale bundle adjustment by Parameter Vector partition
Computer Vision – ACCV 2012, 2013Co-Authors: Shanmin Pang, Le Wang, Jianrue Xue, Nanning ZhengAbstract:We propose an efficient parallel bundle adjustment (BA) algorithm to refine 3D reconstruction of the large-scale structure from motion (SfM) problem, which uses image collections from Internet. Different from the latest BA techniques that improve efficiency by optimizing the reprojection error function with Conjugate Gradient (CG) methods, we employ the Parameter Vector partition strategy. More specifically, we partition the whole BA Parameter Vector into a set of individual sub-Vectors via normalized cut (Ncut). Correspondingly, the solution of the BA problem can be obtained by minimizing subproblems on these sub-Vector spaces. Our approach is approximately parallel, and there is no need to solve the large-scale linear equation of the BA problem. Experiments carried out on a low-end computer with 4GB RAM demonstrate the efficiency and accuracy of the proposed algorithm.
-
large scale bundle adjustment by Parameter Vector partition
Asian Conference on Computer Vision, 2012Co-Authors: Shanmin Pang, Le Wang, Nanning ZhengAbstract:We propose an efficient parallel bundle adjustment (BA) algorithm to refine 3D reconstruction of the large-scale structure from motion (SfM) problem, which uses image collections from Internet. Different from the latest BA techniques that improve efficiency by optimizing the reprojection error function with Conjugate Gradient (CG) methods, we employ the Parameter Vector partition strategy. More specifically, we partition the whole BA Parameter Vector into a set of individual sub-Vectors via normalized cut (Ncut). Correspondingly, the solution of the BA problem can be obtained by minimizing subproblems on these sub-Vector spaces. Our approach is approximately parallel, and there is no need to solve the large-scale linear equation of the BA problem. Experiments carried out on a low-end computer with 4GB RAM demonstrate the efficiency and accuracy of the proposed algorithm.
