New algorithm for robot localization based on BrunsVigia optimization algorithm | ||
| International Journal of Nonlinear Analysis and Applications | ||
| مقاله 48، دوره 14، شماره 1، فروردین 2023، صفحه 613-621 اصل مقاله (411.62 K) | ||
| نوع مقاله: Research Paper | ||
| شناسه دیجیتال (DOI): 10.22075/ijnaa.2021.20653.2188 | ||
| نویسندگان | ||
| Manizheh GhaemiDizaji؛ Chitra Dadkhah* | ||
| Faculty of Computer Engineering, K.N. Toosi University of Technology, Tehran, Iran | ||
| چکیده | ||
| Two problems with Particle Filters (PF) are particle impoverishment and degeneracy. Resampling is introduced to solve degeneracy problem which happens when the majority of the particles have very small weight and a few particles have large weights and sample’s weight variance is too high. Resampling ignores the less informative particles by replacing them with the better ones but it can results in sample impoverishment or diversity loss problem in the particles if there is no controlling mechanism. BrunsVigia Optimization Algorithm (BVOA) is applied in this paper as an extra step to Pf in order to avoid these problems. Operators of BVOA balance between exploration and exploitation and as the result the optimized PF will put much emphasize on more informative particles while keeping the diversity among them. The optimized PF using BVOA, namely BVOA\_PF, is tested in localization problem in a simulating environment. Application of BVOA\_PF in localization and comparing the simulation results with two well-known optimization algorithms as PSO and GWO verify the efficiency of BVOA in real applications like robot localization. | ||
| کلیدواژهها | ||
| Brunsvigia Optimization Algorithm؛ Evolutionary particle filter؛ Robot Localization | ||
|
آمار تعداد مشاهده مقاله: 17,552 تعداد دریافت فایل اصل مقاله: 12,399 |
||
| تعداد نشریات | 22 |
| تعداد شمارهها | 724 |
| تعداد مقالات | 10,417 |
| تعداد مشاهده مقاله | 72,972,409 |
| تعداد دریافت فایل اصل مقاله | 64,618,211 |