上海交通大学随机模拟方法与应用课程论文 目录 摘要 3 1、遗传算法(GA) 3 2、粒子滤波(PF:Particle Filter). 3 2.1概念 3 2.2感性认识 .4 2.3传统的粒子滤波一一贝叶斯滤波.… 5 2.3.1模型 ..5 2.3.2预测过程p(x-Y)→p(x-) ..5 2.3.3更新过程pxY)→p(xY) 6 2.4权重函数 ..6 2.4.1概念… .6 2.4.2权重函数递推.… 2.4.3重要密度函数(qx)的选择. .8 2.5传统的随机模拟方法 3.论文的创新点 8 3.1传统粒子滤波(贝叶斯滤波)的不足 3.2遗传算法与粒子滤波的相似性 8 3.3GA-MCMC粒子重采样… .8 3.3.1重组 ..8 3.3.2变异 .9 3.3.3选择… .9 3.4实验结果及分析 9 4.结论 10 5.自己的看法 .10 2上海交通大学随机模拟方法与应用课程论文 2 目录 摘要 ...................................................................................................................................... 3 1、遗传算法(GA)..................................................................................................................3 2、粒子滤波(PF: Particle Filter)..................................................................................................3 2.1概念 ................................................................................................................................3 2.2感性认识 ........................................................................................................................4 2.3 传统的粒子滤波——贝叶斯滤波...................................................................................5 2.3.1 模型.............................................................................................................................5 2.3.2 预测过程 1 1 1 ( | ) ( | ) k k k k p x Y p x Y .......................................................................5 2.3.3 更新过程 1 ( | ) ( | ) k k k k p x Y p x Y ............................................................................6 2.4 权重函数 ......................................................................................................................6 2.4.1 概念.............................................................................................................................6 2.4.2 权重函数递推..............................................................................................................7 2.4.3 重要密度函数(qx)的选择.......................................................................................8 2.5传统的随机模拟方法 .....................................................................................................8 3. 论文的创新点 ...................................................................................................................8 3.1传统粒子滤波(贝叶斯滤波)的不足.............................................................................8 3.2 遗传算法与粒子滤波的相似性 ......................................................................................8 3.3GA-MCMC 粒子重采样.......................................................................................................8 3.3.1 重组.............................................................................................................................8 3.3.2 变异.............................................................................................................................9 3.3.3 选择.............................................................................................................................9 3.4 实验结果及分析 ...........................................................................................................9 4. 结论 ..................................................................................................................................10 5. 自己的看法 ......................................................................................................................10