fix(nebula_tests): correct scan cutting in VLS128 unit test ground truth #150
Conversation
This file contains 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
…(360deg). has been ~372deg before
mojomex
added a commit
that referenced
this pull request
May 22, 2024
* fix(hesai_hw_interface): add error handling * check PTC command error codes, throw exception if necessary * perform size checks before parsing responses * emit errors on too-large payload size * fix(hesai_decoder): print config instead of config address * refactor(nebula_ros): combine Hesai wrapper nodes into one This is step one of the single node refactoring of Nebula. In this step, only the three Hesai wrapper nodes were combined into one, and no optimizations have been done that are made possible by this refactoring. The next step is to do those optimizations (e.g. get rid of unnecessary pointcloud copying). * refactor: remove pandarScan pub/sub, decode one packet at a time * reword later * temporary progress * fix(hesai_ros_wrapper): remove changes wrongly merged during rebase * fix(hesai_ros_wrapper): increase packet buffering to stop internal packet loss * feat(nebula_common): add instrumentation tools * feat(hesai_ros_wrapper): instrument code on critical path * disable instrumentation * feat(hesai): move received buffer into ros message instead of copy * feat(hesai_ros_wrapper: add thread-safe queue between udp receiver and decoder, decode in separate thread * feat(nebula_common): more instrumentation tools * fix: update link to transport_drivers fork * fix(hesai_hw_monitor_ros_wrapper): fixed wrong range given for S/N copy * update GitHub PR view * fix(hesai): print uint8, uint16 as numbers * refactor(mt_queue): make variables more readable * perf(hesai_ros_wrapper): update queue capacity to alleviate packet drops on ECU * chore(hesai_ros_wrapper): remove unused pub/sub * refactor(hesai_ros_wrapper): clean up control flow, member variables * feat(hesai_ros_wrapper): publish/subscribe to legacy pandar_packets on demand * change launch file back to single-threaded container * attempt to make queue contention less bad * chore(hesai_ros_wrapper): reduce logging output * feat(hesai_ros_wrapper): print warning when connected to HW and receiving /pandar_packets * feat(nebula_launch.py): throw error when trying to launch Hesai sensor and refer to hesai_launch_all_hw.xml * chore: update cspell ignore * refactor(nebula_ros): split single wrapper file into 3 sub-wrappers * chore(velodyne_calibration_decoder): fix spelling * chore(nebula_common): remove debug code * chore(velodyne_calibration_decoder): fix spelling * fix(hesai_decoder): initialize last_phase to prevent empty pointcloud published on startup * fix(nebula_tests): make Hesai tests compile again * fix(nebula_examples): make Hesai examples compile again * chore(velodyne_calibration_decoder): fix spelling once and for all * chore(velodyne_calibration_decoder): fix spelling once and for all * fix(nebula_examples): fix test failure (remove ament_lint_common) * fix(hesai_hw_interface): add missing check for PTC error, make error type more readable * refactor(hesai): re-introduce parameter update mechanism * feat(hesai_ros): add watchdog timer for pointcloud output * fix(hesai): change to possibly more accurate high_resolution_clock * fix(hesai): fix crash on QT128 * refactor(hesai_hw_interface): refactor repeated error handling code * fix(hesai_launch_all_hw.xml): set valid RPM when AT128 is selected * refactor(hesai_decoder): remove redundant arguments for correction/calibration * refactor(nebula_ros): move mt_queue to common * refactor(velodyne)!: unify wrapper nodes. Currently WIP but compiling * fix(velodyne_decoders): after the last commit, point clouds weren't reset correctly. fixed in this commit * fix(velodyne_tests): make tests compile with the new decoder API * fix(velodyne_examples): make examples compile with new decoder API * feat(velodyne_hw_interface): synchronous, null- and thread-safe HTTP requests * fix(velodyne_launch_all_hw.xml): refactor to single-node * fix(velodyne): make hw interface wrapper work with new hw interface API * fix(velodyne): implement correct locking behavior in hw monitor wrapper * fix(nebula_tests): re-cut scans in existing reference data correctly (360deg). has been ~372deg before (#150) * chore(velodyne_scan_decoder.hpp): explicitly initialize some fields for clarity * feat(nebula_ros): add velodyne json schema Signed-off-by: amadeuszsz <[email protected]> * chore(nebula_ros): update config Signed-off-by: amadeuszsz <[email protected]> * chore(nebula_ros): update schema Signed-off-by: amadeuszsz <[email protected]> * chore(nebula): add json schema workflow Signed-off-by: amadeuszsz <[email protected]> * ci(nebula): use of autoware actions repo Signed-off-by: amadeuszsz <[email protected]> * chore(nebula_ros): add schema composition Signed-off-by: amadeuszsz <[email protected]> * chore(pre-commit): autoupdate hooks * fix(hesai_decoder): print config instead of config address * refactor(nebula_ros): combine Hesai wrapper nodes into one This is step one of the single node refactoring of Nebula. In this step, only the three Hesai wrapper nodes were combined into one, and no optimizations have been done that are made possible by this refactoring. The next step is to do those optimizations (e.g. get rid of unnecessary pointcloud copying). * refactor: remove pandarScan pub/sub, decode one packet at a time * reword later * temporary progress * fix(hesai_ros_wrapper): remove changes wrongly merged during rebase * fix(hesai_ros_wrapper): increase packet buffering to stop internal packet loss * feat(nebula_common): add instrumentation tools * feat(hesai_ros_wrapper): instrument code on critical path * disable instrumentation * feat(hesai): move received buffer into ros message instead of copy * feat(hesai_ros_wrapper: add thread-safe queue between udp receiver and decoder, decode in separate thread * feat(nebula_common): more instrumentation tools * fix: update link to transport_drivers fork * update GitHub PR view * refactor(mt_queue): make variables more readable * perf(hesai_ros_wrapper): update queue capacity to alleviate packet drops on ECU * chore(hesai_ros_wrapper): remove unused pub/sub * refactor(hesai_ros_wrapper): clean up control flow, member variables * fix(hesai_hw_interface): add error handling * check PTC command error codes, throw exception if necessary * perform size checks before parsing responses * emit errors on too-large payload size * fix(hesai_hw_monitor_ros_wrapper): fixed wrong range given for S/N copy * fix(hesai): print uint8, uint16 as numbers * feat(hesai_ros_wrapper): publish/subscribe to legacy pandar_packets on demand * change launch file back to single-threaded container * attempt to make queue contention less bad * chore(hesai_ros_wrapper): reduce logging output * feat(hesai_ros_wrapper): print warning when connected to HW and receiving /pandar_packets * feat(nebula_launch.py): throw error when trying to launch Hesai sensor and refer to hesai_launch_all_hw.xml * refactor(nebula_ros): split single wrapper file into 3 sub-wrappers * chore: update cspell ignore * chore(velodyne_calibration_decoder): fix spelling * chore(nebula_common): remove debug code * chore(velodyne_calibration_decoder): fix spelling * fix(hesai_decoder): initialize last_phase to prevent empty pointcloud published on startup * fix(nebula_tests): make Hesai tests compile again * fix(nebula_examples): make Hesai examples compile again * chore(velodyne_calibration_decoder): fix spelling once and for all * fix(nebula_examples): fix test failure (remove ament_lint_common) * fix(hesai_hw_interface): add missing check for PTC error, make error type more readable * refactor(hesai): re-introduce parameter update mechanism * feat(hesai_ros): add watchdog timer for pointcloud output * fix(hesai): change to possibly more accurate high_resolution_clock * fix(hesai): fix crash on QT128 * refactor(hesai_hw_interface): refactor repeated error handling code * fix(hesai_launch_all_hw.xml): set valid RPM when AT128 is selected * refactor(hesai_decoder): remove redundant arguments for correction/calibration * refactor(nebula_ros): move mt_queue to common * chore(nebula_examples): change output rosbag format to NebulaPackets, clean up code * chore(hesai/decoder_wrapper): clarify watchdog behavior in decoder * chore(hesai/hw_monitor_wrapper): fix typo in diagnostics name * chore(hesai_ros_decoder_test): fix unclear naming in console output * chore(nebula_ros): add schema composition for robosense Signed-off-by: amadeuszsz <[email protected]> * chore: run pre-commit and implement suggested fixes (not including copyright yet) * fix(expected.hpp): revert explicit constructors * chore: remove erroneously added node_modules folder * chore(gitconfig): exclude node_modules from git * fix(nebula_tests): missing serialization format Signed-off-by: amadeuszsz <[email protected]> --------- Signed-off-by: amadeuszsz <[email protected]> Signed-off-by: amadeuszsz <[email protected]> Co-authored-by: Max SCHMELLER <[email protected]> Co-authored-by: Abraham Monrroy Cano <[email protected]> Co-authored-by: Maximilian Schmeller <[email protected]> Co-authored-by: Max Schmeller <[email protected]>
mojomex
added a commit
to mojomex/nebula
that referenced
this pull request
May 23, 2024
…(360deg). has been ~372deg before (tier4#150)
mojomex
added a commit
to mojomex/nebula
that referenced
this pull request
May 23, 2024
…(360deg). has been ~372deg before (tier4#150)
mojomex
added a commit
that referenced
this pull request
May 24, 2024
…#148) * refactor(velodyne)!: unify wrapper nodes. Currently WIP but compiling * fix(velodyne_decoders): after the last commit, point clouds weren't reset correctly. fixed in this commit * fix(velodyne_tests): make tests compile with the new decoder API * fix(velodyne_examples): make examples compile with new decoder API * feat(velodyne_hw_interface): synchronous, null- and thread-safe HTTP requests * fix(velodyne_launch_all_hw.xml): refactor to single-node * fix(velodyne): make hw interface wrapper work with new hw interface API * fix(velodyne): implement correct locking behavior in hw monitor wrapper * chore(velodyne_scan_decoder.hpp): explicitly initialize some fields for clarity * fix(nebula_tests): re-cut scans in existing reference data correctly (360deg). has been ~372deg before (#150) * fix: fix mistakes made during rebase * chore(velodyne_ros_wrapper): fix PandarScan mention * chore(velodyne): apply pre-commit (except copyright) * chore(velodyne_scan_decoder): moved magic numbers to constants
5 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
PR Type
Related Links
PR that uncovered this bug:
Description
The VelodyneScan messages in VLS128's ground truth rosbag used for its unit tests contained more than 260deg worth of packets, in fact it was close to 372deg:
This PR re-cuts the scans in the VLS128's ground truth rosbag and PCD file. This means that the VelodynePackets from before are still there, but are distributed differently across VelodyneScans. The scan messages now start/end at 0deg (the scan phase), except for the 1st pointcloud, which starts at 131deg since the first packet in the rosbag was at 131deg.
Review Procedure
Check that unit tests still pass, look at code and rosbag metadata. Confirm the PCD file looks good (e.g. via VS code plugin
pcd-viewer).Remarks
The last 0-183deg worth of packets were discarded as they do not contribute to any scan (scans complete at 0deg).
Pre-Review Checklist for the PR Author
PR Author should check the checkboxes below when creating the PR.
Checklist for the PR Reviewer
Reviewers should check the checkboxes below before approval.
Post-Review Checklist for the PR Author
PR Author should check the checkboxes below before merging.
CI Checks