Point-LIO: mirror FAST-LIO cleanups (no-YAML config, memory2 Recorder, frame scheme)

[jeff-hykin]

Mirrors onto the Point-LIO native module the same cleanups already landed for FAST-LIO in #2498, so the two modules stay consistent.

[codecov]

❌ 1 Tests Failed:

Tests completed	Failed	Passed	Skipped
2335	1	2334	74

View the full list of 1 ❄️ flaky test(s)

dimos.e2e_tests.test_dimsim_spatial_memory::test_go_to_the_bed

Flake rate in main: 21.15% (Passed 41 times, Failed 11 times)

Stack Traces | 559s run time

lcm_spy = <dimos.e2e_tests.lcm_spy.LcmSpy object at 0x7a26aa0df5c0>
start_blueprint = <function start_blueprint.<locals>.set_name_and_start at 0x7a26a6f3e5c0>
human_input = <function human_input.<locals>.send_human_input at 0x7a26a6f3e660>
dim_sim = <dimos.e2e_tests.dim_sim_client.DimSimClient object at 0x7a26a9336c60>
explore_house = <function explore_house.<locals>.explore at 0x7a26a6f3eb60>

    @pytest.mark.self_hosted_large
    def test_go_to_the_bed(lcm_spy, start_blueprint, human_input, dim_sim, explore_house) -> None:
        start_blueprint(
            "run",
            "unitree-go2-agentic",
            simulator="dimsim",
        )
        lcm_spy.save_topic(".../McpClient/on_system_modules/res")
        lcm_spy.wait_for_saved_topic(".../McpClient/on_system_modules/res", timeout=1200.0)
    
        explore_house()
    
        human_input("go to the bed")
    
>       lcm_spy.wait_until_odom_position(-3.567, -1.332, threshold=2, timeout=180)

dim_sim    = <dimos.e2e_tests.dim_sim_client.DimSimClient object at 0x7a26a9336c60>
explore_house = <function explore_house.<locals>.explore at 0x7a26a6f3eb60>
human_input = <function human_input.<locals>.send_human_input at 0x7a26a6f3e660>
lcm_spy    = <dimos.e2e_tests.lcm_spy.LcmSpy object at 0x7a26aa0df5c0>
start_blueprint = <function start_blueprint.<locals>.set_name_and_start at 0x7a26a6f3e5c0>

dimos/e2e_tests/test_dimsim_spatial_memory.py:32: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
dimos/e2e_tests/lcm_spy.py:182: in wait_until_odom_position
    self.wait_for_message_result(
        predicate  = <function LcmSpy.wait_until_odom_position.<locals>.predicate at 0x7a26a6f3ed40>
        self       = <dimos.e2e_tests.lcm_spy.LcmSpy object at 0x7a26aa0df5c0>
        threshold  = 2
        timeout    = 180
        x          = -3.567
        y          = -1.332
dimos/e2e_tests/lcm_spy.py:168: in wait_for_message_result
    self.wait_until(
        event      = <threading.Event at 0x7a26a93366f0: unset>
        fail_message = 'Failed to get to position x=-3.567, y=-1.332'
        listener   = <function LcmSpy.wait_for_message_result.<locals>.listener at 0x7a26a6f3ec00>
        predicate  = <function LcmSpy.wait_until_odom_position.<locals>.predicate at 0x7a26a6f3ed40>
        self       = <dimos.e2e_tests.lcm_spy.LcmSpy object at 0x7a26aa0df5c0>
        timeout    = 180
        topic      = '/odom#geometry_msgs.PoseStamped'
        type       = <class 'dimos.msgs.geometry_msgs.PoseStamped.PoseStamped'>
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

self = <dimos.e2e_tests.lcm_spy.LcmSpy object at 0x7a26aa0df5c0>

    def wait_until(
        self,
        *,
        condition: Callable[[], bool],
        timeout: float,
        error_message: str,
        poll_interval: float = 0.1,
    ) -> None:
        start_time = time.time()
        while time.time() - start_time < timeout:
            if condition():
                return
            time.sleep(poll_interval)
>       raise TimeoutError(error_message)
E       TimeoutError: Failed to get to position x=-3.567, y=-1.332

condition  = <bound method Event.is_set of <threading.Event at 0x7a26a93366f0: unset>>
error_message = 'Failed to get to position x=-3.567, y=-1.332'
poll_interval = 0.1
self       = <dimos.e2e_tests.lcm_spy.LcmSpy object at 0x7a26aa0df5c0>
start_time = 1782452239.9823167
timeout    = 180

dimos/e2e_tests/lcm_spy.py:105: TimeoutError

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}

[greptile-apps]

Greptile Summary

This PR mirrors the FAST-LIO cleanups from #2498 onto Point-LIO, keeping the two native LiDAR modules consistent in their configuration pattern, recorder structure, and TF frame scheme.

• No-YAML config: default.yaml and config_path are deleted; all Point-LIO tuning parameters are now declared as typed Pydantic fields on PointLioConfig and passed to the C++ binary as plain --flag value CLI args. NativeModuleConfig.to_cli_args already handles list[float] fields (gravity, extrinsic vectors) by joining them with commas, matching the new parse_doubles helper in main.cpp.
• Unified recorder: The bespoke PointlioPoseRecorder (pose_recorder.py) and the standalone time-aligning PointlioRecorder are merged into a single PointlioRecorder that extends the shared memory2 Recorder base; @pose_setter_for setters are updated to async def throughout (FAST-LIO, Go2, Unitree blueprints).
• Frame scheme alignment: body_frame_id/body_start_frame_id are renamed to child_frame_id/frame_id; a new sensor_frame_id field stamps the published point-cloud header separately from the odometry frames. macOS unicast support is added to the virtual Mid-360 replayer and SDK config.

Confidence Score: 5/5

Safe to merge — changes are mechanical cleanups mirroring an already-landed FAST-LIO refactor with no novel logic.

The diff removes the YAML-loading path and replaces it with typed Pydantic fields whose CLI serialisation (including list[float] comma-joining for gravity/extrinsic vectors) is handled by the existing NativeModuleConfig.to_cli_args. The Recorder migration is structurally identical to the FAST-LIO change already in main, and the memory2/module.py async promotion is guarded by a runtime iscoroutinefunction check. The macOS unicast fix is a symmetrical change across Rust and C++ headers. No new algorithmic logic is introduced.

No files require special attention. The two pre-existing style observations in module.py and main.cpp (extrinsic vector validation and parse_doubles error surfacing) were noted in the prior review and remain unchanged.

Important Files Changed

Filename	Overview
dimos/hardware/sensors/lidar/pointlio/module.py	Replaces YAML config path with ~50 typed Pydantic fields on PointLioConfig; renames frame fields and adds sensor_frame_id; removes cli_exclude (no longer needed). List[float] fields serialise correctly via NativeModuleConfig.to_cli_args.
dimos/hardware/sensors/lidar/pointlio/cpp/main.cpp	Replaces YAML loading with direct CLI arg parsing for all PointLioParams fields; adds parse_doubles helper; adds sensor_frame_id; renames body_frame_id to child_frame_id. Silent-discard on bad CSV input is a known pre-existing P2.
dimos/hardware/sensors/lidar/pointlio/recorder.py	Drastically simplified: now extends Recorder base, removes bespoke sqlite/timestamp-alignment logic, uses async @pose_setter_for. Staleness tolerance check is intentionally dropped (mirrors FAST-LIO behaviour).
dimos/memory2/module.py	PoseSetter type and _resolve_pose updated to async; pose_setter_for now validates iscoroutinefunction at decoration time; on_msg in _subscribe_port switched to async def and dispatched via process_observable.
dimos/hardware/sensors/lidar/pointlio/scripts/pcap_to_db.py	Major refactor: removes --odom-stream-name/--lidar-stream-name/--force/--time-offset; adds per-field --tuning flags; --config is now a hidden YAML override file; adds LFS fetching for --pcap and --db; uses PointlioRecorder with remappings.
dimos/hardware/sensors/lidar/virtual_mid360/src/main.rs	macOS-specific unicast path for point/IMU packets — on macOS data_dest is host_ip instead of the multicast group, matching the SDK config change in livox_sdk_config.hpp.
dimos/hardware/sensors/lidar/common/livox_sdk_config.hpp	Platform-conditional multicast_ip: empty string on macOS (binds host_ip to receive unicasts) vs 224.1.1.5 on Linux. Correctly paired with the Rust-side change.
dimos/hardware/sensors/lidar/fastlio2/recorder.py	Pose setters updated to async def, matching the new Recorder contract; _lidar_pose uses getattr defensive fallback matching PointlioRecorder style.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    subgraph Hardware["Hardware / Replay"]
        MID360["Mid-360 LiDAR\n(real or VirtualMid360 PCAP)"]
    end

    subgraph PointLIO["PointLio NativeModule (C++ subprocess)"]
        SDK["Livox SDK2 callbacks"]
        ALGO["Point-LIO IESKF estimator"]
        CLI["CLI args from PointLioConfig\n(no YAML)"]
        CLI --> ALGO
        SDK --> ALGO
        ALGO --> |"odometry (nav_msgs.Odometry)"|OUT_ODOM["Out: odometry"]
        ALGO --> |"lidar (sensor_msgs.PointCloud2)\nstamped with sensor_frame_id"|OUT_LIDAR["Out: lidar"]
        ALGO --> |"TF: frame_id → child_frame_id"|TF["TF bus"]
    end

    subgraph Recorder["PointlioRecorder (memory2 Recorder)"]
        IN_ODOM["In: pointlio_odometry\n(remapped ← odometry)"]
        IN_LIDAR["In: pointlio_lidar\n(remapped ← lidar)"]
        POSE_SETTER["@pose_setter_for async\n_odom_pose → caches _last_odom_pose\n_lidar_pose → returns cached pose"]
        DB["SQLite db\npointlio_odometry / pointlio_lidar tables\n(with pose anchor per row)"]
        IN_ODOM --> POSE_SETTER
        IN_LIDAR --> POSE_SETTER
        POSE_SETTER --> DB
    end

    MID360 -->|"UDP point/IMU\n(multicast Linux / unicast macOS)"| SDK
    OUT_ODOM -->|"autoconnect"| IN_ODOM
    OUT_LIDAR -->|"autoconnect"| IN_LIDAR

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
    subgraph Hardware["Hardware / Replay"]
        MID360["Mid-360 LiDAR\n(real or VirtualMid360 PCAP)"]
    end

    subgraph PointLIO["PointLio NativeModule (C++ subprocess)"]
        SDK["Livox SDK2 callbacks"]
        ALGO["Point-LIO IESKF estimator"]
        CLI["CLI args from PointLioConfig\n(no YAML)"]
        CLI --> ALGO
        SDK --> ALGO
        ALGO --> |"odometry (nav_msgs.Odometry)"|OUT_ODOM["Out: odometry"]
        ALGO --> |"lidar (sensor_msgs.PointCloud2)\nstamped with sensor_frame_id"|OUT_LIDAR["Out: lidar"]
        ALGO --> |"TF: frame_id → child_frame_id"|TF["TF bus"]
    end

    subgraph Recorder["PointlioRecorder (memory2 Recorder)"]
        IN_ODOM["In: pointlio_odometry\n(remapped ← odometry)"]
        IN_LIDAR["In: pointlio_lidar\n(remapped ← lidar)"]
        POSE_SETTER["@pose_setter_for async\n_odom_pose → caches _last_odom_pose\n_lidar_pose → returns cached pose"]
        DB["SQLite db\npointlio_odometry / pointlio_lidar tables\n(with pose anchor per row)"]
        IN_ODOM --> POSE_SETTER
        IN_LIDAR --> POSE_SETTER
        POSE_SETTER --> DB
    end

    MID360 -->|"UDP point/IMU\n(multicast Linux / unicast macOS)"| SDK
    OUT_ODOM -->|"autoconnect"| IN_ODOM
    OUT_LIDAR -->|"autoconnect"| IN_LIDAR

Loading

_{Reviews (5): Last reviewed commit: "-" | Re-trigger Greptile}