Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

SOME questions about parameters setting #19

Open
SeaEastXu opened this issue Oct 18, 2019 · 3 comments
Open

SOME questions about parameters setting #19

SeaEastXu opened this issue Oct 18, 2019 · 3 comments

Comments

@SeaEastXu
Copy link

Hi.
I want to modify the posecell network and launch my own experiment.I have read the paper "OpenRatSLAM",but I still don't know how to set some parameters:
pc_dim_xy: how to set up in correspondence to experiment environment
vt_active,pc_vt_restore

Could you give me some tips from your experiences?

THANK YOU

@scott--
Copy link
Collaborator

scott-- commented Nov 2, 2019

In Table 4 in the paper there are some values given for different environments for pc_dim_xy and pc_active_decay. Try the defaults for these vt_active_decay = 1.0, and pc_vt_restore = 0.05.

For pc_dim_xy, copy from the environments. Start by using pc_dim_xy = 30 for a bigger outdoor envionment, or pc_dim_xy = 11 for a smaller environment. If you get a lot of false positives, you might have to raise that parameter.

@SeaEastXu
Copy link
Author

3q!
As you said,pc_dim_xy is related to the size of environment, but there is another confusion about parameter pc_cell_x_size, illustrated as a scaling factor in paper, can be adjusted to suit the translational velocity range of the robot or sensor platform.
1.How to adjust pc_cell_x_size _?
vtrans /= PC_CELL_X_SIZE;
2."The normal operating range for this implementation of OpenRatSLAM is to limit the movement of the energy in the network to one cell per iteration".What this explanation means?
3.Can it be assumed that we use pc_dim_xy,pc_dim_xy to scale the real wolrd coodinate?

@SeaEastXu
Copy link
Author

SeaEastXu commented Nov 30, 2019

And another question is that I don't konw the principle to calculate the weight ?

  // % extend the pc.Posecells one unit in each direction (max supported at the moment)

 // % work out the weight contribution to the NE cell from the SW, NW, SE cells

 // % given vtrans and the direction

// % weight_sw = v * cos(th) * v * sin(th)         **why???**

// % weight_se = (1 - v * cos(th)) * v * sin(th)      **why???**

// % weight_nw = (1 - v * sin(th)) * v * sin(th)               **why???**


 // % weight_ne = 1 - weight_sw - weight_se - weight_nw     

// % think in terms of NE divided into 4 rectangles with the sides
// % given by vtrans and the angle
// pca_new=zeros(PARAMS.PC_DIM_XY+2);

I'm looking forward to your reply.
Thank u!!!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants