Skip to content

Kalman forward filter causing jaggedness in example notebook #122

@pavelkomarov

Description

@pavelkomarov

Something about #98 caused jaggedness in the Kalman results that I didn't notice until today. The unit test results aren't jagged, just the results on the simulation used for 1_basic_tutorial.ipynb.

Image

I've restored the old code and pasted in my new filter function, and the jaggedness appears, so it's not the backward step, not the "correction", not the code restructuring. It's likely I introduced a bug, although it's possible the old code was actually not right. I'm doing ye olde ablation testing on the old code to see exactly where modifying it into the new causes this symptom. That will be a clue.

Metadata

Metadata

Assignees

Labels

bugSomething isn't working

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions