کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
1138333 1489209 2007 14 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
GMG — A guaranteed global optimization algorithm: Application to remote sensing
موضوعات مرتبط
مهندسی و علوم پایه سایر رشته های مهندسی کنترل و سیستم های مهندسی
پیش نمایش صفحه اول مقاله
GMG — A guaranteed global optimization algorithm: Application to remote sensing
چکیده انگلیسی

We investigate the role of additional information in reducing the computational complexity of the global optimization problem (GOP). Following this approach, we develop GMG — an algorithm for finding the Global Minimum with a Guarantee. The new algorithm breaks up an originally continuous GOP into a discrete (grid) search problem followed by a descent problem. The discrete search identifies the basin of attraction of the global minimum after which the actual location of the minimizer is found upon applying a descent algorithm. The algorithm is first applied to the golf-course problem, which serves as a litmus test for its performance in the presence of both complete and degraded additional information. GMG is further assessed on a set of standard benchmark functions. We then illustrate the performance of the validated algorithm on a simple realization of the monocular passive ranging (MPR) problem in remote sensing, which consists of identifying the range of an airborne target (missile, plane, etc.) from its observed radiance. This inverse problem is set as a GOP whereby the difference between the observed and model predicted radiances is minimized over the possible ranges and atmospheric conditions. We solve the GOP using GMG and report on the performance of the algorithm.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Mathematical and Computer Modelling - Volume 45, Issues 3–4, February 2007, Pages 459–472
نویسندگان
, , , ,