Fix road boundary parsing for XODR#65
Closed
maximilianigl wants to merge 5 commits into
Closed
Conversation
…otobuf4, removal of map rasterization, etc).
* Add MADS dataset support Add the MADS dataset module, which provides: - MADSDataset: full dataset integration for loading MADS clipgt data including ego and obstacle agent parsing with cubic interpolation, overlap filtering, and map caching - populate_vector_map: constructs VectorMap instances from MADS parquet map data (lanes, road edges, traffic signs, wait lines) - MadsSceneRecord and env_utils registration for dataset discovery * Fix test * Update src/trajdata/dataset_specific/mads/mads_utils.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/dataset_specific/mads/mads_utils.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/utils/env_utils.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/dataset_specific/mads/mads_utils.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/dataset_specific/mads/mads_utils.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/dataset_specific/mads/mads_dataset.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/dataset_specific/mads/mads_dataset.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/dataset_specific/mads/mads_dataset.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update src/trajdata/dataset_specific/mads/mads_dataset.py Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com> * Update mads_dataset.py * fixes * Update env_utils.py * Update mads_utils.py --------- Co-authored-by: Boris Ivanovic <8534290+BorisIvanovic@users.noreply.github.com>
_extract_road_edges read lg.left_edge from the outermost left lane to get the left road boundary. In v1.4.3 the edge convention changed (left_edge = inner for all lanes), but _extract_road_edges was not updated, so the left boundary ended up at the centerline. Replace the indirect lookup with direct capture: after the edge-chaining loop finishes each side, current_edge is already the outermost road edge. Record it there and delete _extract_road_edges entirely. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Author
|
Mistakenly tried to merge into the wrong branch. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Updates the parsing of XODR to address (some of) the issues mentioned here: NVlabs/alpasim#61
The fundamental problem is that our XODR files don't have road boundaries. instead, we try to infer which lane-boundary is also a road boundary.
Before: Take the rightmost lane boundary in driving direction. It's not entirely clear to me why that didn't work reliably, but it didn't.
Now: Take the outermost lane-boundary on either side.
Guillermo did some analysis on the faulty scenes and it seems to fix most reported issues. The remaining ones might also be data problems: https://docs.google.com/document/d/1Lys_rCapKg0z1TLyBJ3WkAUNwy9-wKdZaXNOH1anl3Q/edit?tab=t.62pzl5e0asj