perf(perception): reuse local surface cell observations

src/k1link/compute/lidar_local_surface_geometry.py

@@ -166,6 +166,16 @@ def update_cache(
     profile: K1LocalSurfaceProfile,
 ) -> None:
     keys, points = cloud_cell_observations(cloud, profile)
+    update_cache_observations(cache, keys, points, session_seconds)
+
+
+def update_cache_observations(
+    cache: dict[tuple[int, int], tuple[float, float]],
+    keys: npt.NDArray[np.int64],
+    points: npt.NDArray[np.float64],
+    session_seconds: float,
+) -> None:
+    """Update a surface cache from already aggregated cell observations."""
     for key, point in zip(keys, points, strict=True):
         cache[(int(key[0]), int(key[1]))] = (float(point[2]), session_seconds)
 

src/k1link/compute/lidar_local_surface_shadow.py

@@ -46,7 +46,7 @@ from .lidar_local_surface_geometry import (
     step_candidate_keys as _step_candidate_keys,
 )
 from .lidar_local_surface_geometry import (
-    update_cache as _update_cache,
+    update_cache_observations as _update_cache_observations,
 )
 from .live_perception import LatestWinsQueue
 
@@ -489,7 +489,7 @@ class K1LocalSurfaceShadowEstimator:
             position,
             self.profile,
         )
-        _, current_cell_points = _cloud_cell_observations(
+        current_cell_keys, current_cell_points = _cloud_cell_observations(
             local_cloud,
             self.profile,
         )
@@ -499,11 +499,11 @@ class K1LocalSurfaceShadowEstimator:
             position,
             self.profile,
         )
-        _update_cache(
+        _update_cache_observations(
             self._cache,
-            local_cloud,
+            current_cell_keys,
+            current_cell_points,
             value.session_seconds,
-            self.profile,
         )
         cell_keys, cell_points, cell_times = _local_cache_records(
             self._cache,

tests/test_lidar_local_surface.py

@@ -26,6 +26,11 @@ from k1link.compute.lidar_local_surface import (
 from k1link.compute.lidar_local_surface_geometry import (
     DEFAULT_K1_LOCAL_SURFACE_PROFILE as WORKER_LOCAL_SURFACE_PROFILE,
 )
+from k1link.compute.lidar_local_surface_geometry import (
+    cloud_cell_observations,
+    update_cache,
+    update_cache_observations,
+)
 from k1link.web.lidar_api import build_lidar_router
 from k1link.web.lidar_local_surface_service import K1LocalSurfaceReadService
 
@@ -48,6 +53,26 @@ def test_replay_and_minimal_worker_pin_the_same_local_surface_profile() -> None:
     assert WORKER_LOCAL_SURFACE_PROFILE.to_dict() == REPLAY_LOCAL_SURFACE_PROFILE.to_dict()
 
 
+def test_precomputed_local_surface_observations_preserve_cache_update() -> None:
+    cloud = np.asarray(
+        [
+            [0.1, 0.1, 0.8],
+            [0.2, 0.2, 0.2],
+            [1.1, 0.1, 0.5],
+            [-0.2, -0.2, -0.1],
+        ],
+        dtype=np.float64,
+    )
+    baseline: dict[tuple[int, int], tuple[float, float]] = {}
+    optimized: dict[tuple[int, int], tuple[float, float]] = {}
+
+    update_cache(baseline, cloud, 4.25, WORKER_LOCAL_SURFACE_PROFILE)
+    keys, points = cloud_cell_observations(cloud, WORKER_LOCAL_SURFACE_PROFILE)
+    update_cache_observations(optimized, keys, points, 4.25)
+
+    assert optimized == baseline
+
+
 def _source_pack(root: Path) -> Path:
     frame_count = 8
     available = np.asarray([True, True, True, True, True, True, True, False])