Weird behaviour of raisim::world::rayTest

[mstoelzle]

@jhwangbo I experienced some weird behaviour for some cases when using the raisim::world::rayTest method for my raycasting. In this example I am raycasting from a vantage point one meter above ground in the center of the grid. In the first screenshot you can see the result I received with the code build on raisim. In the second screenshot you can see the result I would expect to see (the elevation of the rectangle in the upper-left is at 10m and the rest of the ground is flat at 0m) and which I received after writing my own rayTest code. I am not sure, maybe I used parameters for the raisim::world::rayTest method which failed in some edge-cases, but it was hard for me to debug without seeing the source code.

Result of raycasting using raisim::world::rayTest (https://github.com/mstoelzle/grid-map-raycasting/blob/raisim/src/main.cpp):

Result of raycasting using my own rayTest code (https://github.com/mstoelzle/grid-map-raycasting/blob/master/src/main.cpp):

[jhwangbo]

Can you share the raisimUnity screenshot as well?

[mstoelzle]

Hi @jhwangbo , I attached an screenshot of the height map in raisimUnity

[jhwangbo]

You also need to visualize the scan points as well. You can check rayDemo2

[mstoelzle]

I tried to run the rayDemo2.py using raisimPy to see the visualization of the scan points, but it fails with this error:

Assertion failed: (index >= 0 && index < size()), function operator[], file /usr/local/include/eigen3/Eigen/src/Core/DenseCoeffsBase.h, line 162.
zsh: abort      python raisimPy/examples/rayDemo2.py
``

[jhwangbo]

I fixed a bug in the demo file. you can run it.
Next time you should run the code in debug mode and tell me which line failed. Currently, I don't have time to resolve all issues.

[jhwangbo]

I'll close this issue until I see the visualization.