ROS Answers: Open Source Q&A Forum - RSS feedhttps://answers.ros.org/questions/Open source question and answer forum written in Python and DjangoenROS Answers is licensed under Creative Commons Attribution 3.0Mon, 17 Oct 2016 03:59:24 -0500Is kld_z in amcl parameters mapped with standard statistical z table?https://answers.ros.org/question/245832/is-kld_z-in-amcl-parameters-mapped-with-standard-statistical-z-table/Hi, In [AMCL.cfg](https://github.com/ros-planning/navigation/blob/jade-devel/amcl/cfg/AMCL.cfg) file the kld_z parameter has a range from **0** to **1** and the default value of **0.99**. According to the [paper on KLD sampling](http://www.robots.ox.ac.uk/~cvrg/hilary2005/adaptive.pdf) pg. 15 the probability 1-delta (the error between the true posterior and the sample based approximation is less than kld_err) has value **0.99**. This information shows that the kld_z is a probabilty.
In [pf.c of amcl](https://github.com/ros-planning/navigation/blob/jade-devel/amcl/src/amcl/pf/pf.c) this kld_z probabilty parameter is directly feed in the function named **pf_resample_limit()** as variable pf->pop_z. while the [Paper on KLD sampling on page 13](http://www.robots.ox.ac.uk/~cvrg/hilary2005/adaptive.pdf) suggest that this kld_z probability should be mapped with standard statistical z table. But i cannot find any such mapping in the code.Manish SaroyaMon, 17 Oct 2016 03:59:24 -0500https://answers.ros.org/question/245832/