lidarslam_ros2 v0.2.0

This is the public v2 beta release for lidarslam_ros2.

Date: 2026-03-25

Highlights

• the default workflow is now explicitly permissive-license only
• RKO-LIO + graph_based_slam is the recommended and dogfooded path
• the graph backend now has safer adjacent edges, loop dedup, robust kernels, multi-candidate validation, and safer snapshot-based state handling
• Autoware-compatible pointcloud-map export is part of the documented public path
• benchmark summary, HTML report, and release-readiness gate are part of the publication flow

Public Snapshot

• fixed Autoware entrypoint: bash scripts/run_autoware_quickstart.sh
• NTU VIRAL current default APE RMSE: 0.952 m
• NTU VIRAL best observed APE RMSE: 0.870 m
• MID360 current default cross-validation APE RMSE: 3.641 m
• MID360 best observed cross-validation APE RMSE: 3.590 m
• comparison page: docs/comparison.md

Supported Scope

• ROS 2 LiDAR SLAM with a permissive default path
• pointcloud-map generation for Autoware
• map_projector_info.yaml output as Local without GNSS and LocalCartesian with a stable GNSS origin
• release-readiness reports from collected metrics.json artifacts

Known Limits

• lanelet generation is out of scope
• MID360 evidence is currently cross-validation against GLIM, not ground truth
• the current release should be described as beta, not as fully production hardened across every dataset

Recommended Entry Points

bash scripts/run_autoware_quickstart.sh
bash scripts/run_rko_lio_graph_benchmark.sh
bash scripts/run_release_readiness_checks.sh --ape-threshold 0.10

Release Artifacts

For public distribution, publish at least:

• docs/comparison.md
• output/benchmark_summary.md
• output/latest_report.html
• output/v2_beta_readiness_20260324.md
• output/stress_validation_report_20260325.md