本文關(guān)鍵詞：無人機(jī)跟蹤場景下的粒子濾波算法的改進(jìn)　出處：《計算機(jī)仿真》2017年02期 　論文類型：期刊論文

  更多相關(guān)文章： 粒子濾波 無人機(jī) 粒子初始化 粒子更新 粒子重采樣

【摘要】：為了提高粒子濾波跟蹤算法的效率,針對圖像的特點(diǎn)對粒子濾波算法進(jìn)行改進(jìn)。首先針對無人機(jī)跟蹤過程中,目標(biāo)始終在圖像中心位置附近晃動的特點(diǎn),結(jié)合高斯分布對粒子初始化過程進(jìn)行改進(jìn),保證粒子以較大概率集中在圖像中心位置附近;然后針對無人機(jī)姿態(tài)調(diào)整過程中,目標(biāo)運(yùn)動朝向圖像中心位置的特點(diǎn),采用高斯加權(quán)后的權(quán)值對粒子重采樣進(jìn)行改進(jìn),保證圖像中心區(qū)域的粒子有更大概率被保留。最后通過實(shí)驗(yàn)對改進(jìn)算法進(jìn)行分析和評價,驗(yàn)證了改進(jìn)算法的有效性。
[Abstract]:In order to improve the efficiency of the particle filter tracking algorithm, the particle filter algorithm is improved according to the characteristics of the image. Firstly, the target is always sloshing near the center of the image in the tracking process of UAV. The particle initialization process is improved by using Gao Si distribution to ensure that the particle is concentrated near the center of the image with a high probability. Then, aiming at the target moving towards the center of the image in the course of UAV attitude adjustment, the weighted weight of Gao Si is used to improve the particle resampling. Finally, the improved algorithm is analyzed and evaluated through experiments to verify the effectiveness of the improved algorithm.
【作者單位】： 清華大學(xué)電子工程系;
【基金】：國家自然科學(xué)基金項(xiàng)目(61172125,61132007) 國家自然科學(xué)基金-民航基金聯(lián)合資助(U1533132)
【分類號】：V279;TN713
【正文快照】： 1_ 隨著無人機(jī)技術(shù)的發(fā)展,無人機(jī)在各個領(lǐng)域的應(yīng)用越來越寬廣。無人機(jī)有著成本低、無人員傷亡風(fēng)險、生存能力強(qiáng)、機(jī)動性能好等優(yōu)點(diǎn),在民用和軍事等領(lǐng)域都有著廣闊的前景[1]。比如農(nóng)業(yè)中播撒農(nóng)藥,森林消防中對森林進(jìn)行巡視,災(zāi)難發(fā)生時對災(zāi)區(qū)情況探g�。随着无人机的飞行更�，

本文編號：1407917