FIX: Correctly calculate masses even if no particles were predicted into a cell

TheCodez / dynamic-occupancy-grid-map

Implementation of "A Random Finite Set Approach for Dynamic Occupancy Grid Maps with Real-Time Application"

MIT License

278 stars 39 forks source link

FIX: Correctly calculate masses even if no particles were predicted into a cell #91

Closed TheCodez closed 2 years ago

TheCodez commented 2 years ago

If no particles where predicted into a cell, a simplified calculation was applied in the mass update. This lead to issues, such as no particle birth even though a cell received a measured occupancy mass.

See the discussion https://github.com/TheCodez/dynamic-occupancy-grid-map/discussions/87 for more details.

TheCodez commented 2 years ago

@cbachhuber any idea why the CI keeps failing?

cbachhuber commented 2 years ago

Unfortunately not. I don't understand why it fails in CI, and I can't reproduce locally, where clang-tidy just passes.

Feel free to disable clang-tidy in CI (or make it a step that may fail) and assign an issue to me for reactivating it.