简体繁体 English

用粒子滤波器回溯

[英]Backtracking with Particle Filter

原文 2016-07-10 13:37:43 3 1 python/ particle-filter

I have just implemented a particle filter for Indoor Tracking. 我刚刚为室内跟踪实现了一个粒子滤波器。 It looks good but at some points the particles go in a room and are trapped there. 它看起来不错但在某些时候颗粒进入房间并被困在那里。

What's a smart way to do backtracking? 什么是回溯的聪明方法？ I save the state of the particles for their last 10 movements. 我保存了最后10次运动的粒子状态。

Thank you 谢谢

1 个解决方案

It is completely normal that particles get distributed everywhere. 粒子到处分布是完全正常的。 Otherwise it is not a probabilistic approach. 否则，它不是一种概率方法。 In addition, note that the particles are sampled based on the posterior probability at time t-1 and the current motion distribution. 另外，请注意，基于时间t-1处的后验概率和当前运动分布对粒子进行采样。 However, even if it is not recommended in filtering but you can restrict your research space in the sampling step. 但是，即使不建议过滤，但您可以在采样步骤中限制研究空间。

For backtracking, you may use at each time t the same approach of fortracking, with changing only the direction of the velocity (on all axies). 为回溯，可以每次在使用t fortracking的相同的方法，只改变的速度（对所有轴距）的方向上。 You can start from the state which maximise the probability distribution. 您可以从最大化概率分布的状态开始。 Finaly, you compare the obtained trajectories (result of for/backtracking) and you decide based on the result which further filtering is needed to get the best result. 最后，您比较获得的轨迹（for / backtracking的结果），并根据结果确定需要进一步过滤以获得最佳结果。