diff --git a/modules/geolocation/rudimentary_geolocation.py b/modules/geolocation/rudimentary_geolocation.py
new file mode 100644
index 00000000..f566ac14
--- /dev/null
+++ b/modules/geolocation/rudimentary_geolocation.py
@@ -0,0 +1,71 @@
+"""
+Converts image space into world space in a more rudimentary sense
+"""
+
+import numpy as np
+
+from .. import detection_in_world
+from .. import merged_odometry_detections
+from .geolocation import Geolocation
+
+
+class RudimentaryGeolocation(Geolocation):
+    """
+    Converts image space into world space.
+    """
+
+    def run(
+        self, detections: merged_odometry_detections.MergedOdometryDetections
+    ) -> "tuple[bool, list[detection_in_world.DetectionInWorld] | None]":
+        """
+        Runs detections in world space
+        """
+        if detections.odometry_local.position.down >= 0.0:
+            self.__logger.error("Drone is underground")
+            return False, None
+
+        drone_position_ned = np.array(
+            [
+                detections.odometry_local.position.north,
+                detections.odometry_local.position.east,
+                detections.odometry_local.position.down,
+            ],
+            dtype=np.float32,
+        )
+
+        # Since camera points down, the rotation matrix will be this
+        # Calculated assuming pitch=-pi/2 and yaw=roll=0
+        drone_rotation_matrix = np.array(
+            [
+                [0.0, 0.0, -1.0],
+                [0.0, 1.0, 0.0],
+                [1.0, 0.0, 0.0],
+            ],
+            dtype=np.float32,
+        )
+
+        result, perspective_transform_matrix = self.__get_perspective_transform_matrix(
+            drone_rotation_matrix,
+            drone_position_ned,
+        )
+        if not result:
+            return False, None
+
+        # Get Pylance to stop complaining
+        assert perspective_transform_matrix is not None
+
+        # pylint: disable=duplicate-code
+        detections_in_world = []
+        for detection in detections.detections:
+            result, detection_world = self.__convert_detection_to_world_from_image(
+                detection,
+                perspective_transform_matrix,
+                self.__logger,
+            )
+            # Partial data not allowed
+            if not result:
+                return False, None
+            detections_in_world.append(detection_world)
+            self.__logger.info(detection_world)
+
+        return True, detections_in_world
diff --git a/tests/unit/test_rudimentary_geolocation.py b/tests/unit/test_rudimentary_geolocation.py
new file mode 100644
index 00000000..c35e7b34
--- /dev/null
+++ b/tests/unit/test_rudimentary_geolocation.py
@@ -0,0 +1,250 @@
+"""
+Test rudimentary_geolocation.
+"""
+
+import numpy as np
+import pytest
+
+from modules import detection_in_world
+from modules import detections_and_time
+from modules import merged_odometry_detections
+from modules.common.modules import orientation
+from modules.common.modules import position_local
+from modules.common.modules.logger import logger
+from modules.common.modules.mavlink import drone_odometry_local
+from modules.geolocation import camera_properties
+from modules.geolocation import rudimentary_geolocation
+
+FLOAT_PRECISION_TOLERANCE = 4
+
+
+# Test functions use test fixture signature names and access class privates
+# No enable
+# pylint: disable=protected-access,redefined-outer-name
+# pylint: disable=duplicate-code
+
+
+@pytest.fixture
+def basic_locator() -> rudimentary_geolocation.RudimentaryGeolocation:  # type: ignore
+    """
+    Forwards pointing camera.
+    """
+    result, camera_intrinsics = camera_properties.CameraIntrinsics.create(
+        2000,
+        2000,
+        np.pi / 2,
+        np.pi / 2,
+    )
+    assert result
+    assert camera_intrinsics is not None
+
+    result, camera_extrinsics = camera_properties.CameraDroneExtrinsics.create(
+        (0.0, 0.0, 0.0),
+        (0.0, 0.0, 0.0),
+    )
+    assert result
+    assert camera_extrinsics is not None
+
+    result, test_logger = logger.Logger.create("test_logger", False)
+    assert result
+    assert test_logger is not None
+
+    result, locator = rudimentary_geolocation.RudimentaryGeolocation.create(
+        camera_intrinsics,
+        camera_extrinsics,
+        test_logger,
+    )
+    assert result
+    assert locator is not None
+
+    yield locator  # type: ignore
+
+
+@pytest.fixture
+def detection_1() -> detections_and_time.Detection:  # type: ignore
+    """
+    Entire image.
+    """
+    result, detection = detections_and_time.Detection.create(
+        np.array([0.0, 0.0, 2000.0, 2000.0], dtype=np.float32),
+        1,
+        1.0 / 1,
+    )
+    assert result
+    assert detection is not None
+
+    yield detection  # type: ignore
+
+
+@pytest.fixture
+def detection_2() -> detections_and_time.Detection:  # type: ignore
+    """
+    Quadrant.
+    """
+    result, detection = detections_and_time.Detection.create(
+        np.array([0.0, 0.0, 1000.0, 1000.0], dtype=np.float32),
+        2,
+        1.0 / 2,
+    )
+    assert result
+    assert detection is not None
+
+    yield detection  # type: ignore
+
+
+class TestRudimentaryGeolocationRun:
+    """
+    Run.
+    """
+
+    def test_basic(
+        self,
+        basic_locator: rudimentary_geolocation.RudimentaryGeolocation,
+        detection_1: detections_and_time.Detection,
+        detection_2: detections_and_time.Detection,
+    ) -> None:
+        """
+        2 detections.
+        """
+        # Setup
+        result, drone_position = position_local.PositionLocal.create(
+            0.0,
+            0.0,
+            -100.0,
+        )
+        assert result
+        assert drone_position is not None
+
+        result, drone_orientation = orientation.Orientation.create(
+            0.0,
+            -np.pi / 2,
+            0.0,
+        )
+        assert result
+        assert drone_orientation is not None
+
+        result, drone_odometry = drone_odometry_local.DroneOdometryLocal.create(
+            drone_position,
+            drone_orientation,
+        )
+        assert result
+        assert drone_odometry is not None
+
+        result, merged_detections = merged_odometry_detections.MergedOdometryDetections.create(
+            drone_odometry,
+            [
+                detection_1,
+                detection_2,
+            ],
+        )
+        assert result
+        assert merged_detections is not None
+
+        result, expected_detection_1 = detection_in_world.DetectionInWorld.create(
+            # fmt: off
+            np.array(
+                [
+                    [ 100.0, -100.0],
+                    [ 100.0,  100.0],
+                    [-100.0, -100.0],
+                    [-100.0,  100.0],
+                ],
+                dtype=np.float32,
+            ),
+            # fmt: on
+            np.array(
+                [0.0, 0.0],
+                dtype=np.float32,
+            ),
+            1,
+            1.0,
+        )
+        assert result
+        assert expected_detection_1 is not None
+
+        result, expected_detection_2 = detection_in_world.DetectionInWorld.create(
+            # fmt: off
+            np.array(
+                [
+                    [ 100.0, -100.0],
+                    [ 100.0,    0.0],
+                    [   0.0, -100.0],
+                    [   0.0,    0.0],
+                ],
+                dtype=np.float32,
+            ),
+            # fmt: on
+            np.array(
+                [50.0, -50.0],
+                dtype=np.float32,
+            ),
+            2,
+            0.5,
+        )
+        assert result
+        assert expected_detection_2 is not None
+
+        expected_list = [
+            expected_detection_1,
+            expected_detection_2,
+        ]
+
+        # Run
+        result, actual_list = basic_locator.run(merged_detections)
+
+        # Test
+        assert result
+        assert actual_list is not None
+
+        assert len(actual_list) == len(expected_list)
+        for i, actual in enumerate(actual_list):
+            np.testing.assert_almost_equal(actual.vertices, expected_list[i].vertices)
+            np.testing.assert_almost_equal(actual.centre, expected_list[i].centre)
+            assert actual.label == expected_list[i].label
+            np.testing.assert_almost_equal(actual.confidence, expected_list[i].confidence)
+
+    def test_bad_direction(
+        self,
+        basic_locator: rudimentary_geolocation.RudimentaryGeolocation,
+        detection_1: detections_and_time.Detection,
+    ) -> None:
+        """
+        Bad direction.
+        """
+        # Setup
+        result, drone_position = position_local.PositionLocal.create(
+            0.0,
+            0.0,
+            -100.0,
+        )
+        assert result
+        assert drone_position is not None
+
+        result, drone_orientation = orientation.Orientation.create(
+            0.0,
+            0.0,
+            0.0,
+        )
+        assert result
+        assert drone_orientation is not None
+
+        result, drone_odometry = drone_odometry_local.DroneOdometryLocal.create(
+            drone_position,
+            drone_orientation,
+        )
+        assert result
+        assert drone_odometry is not None
+
+        result, merged_detections = merged_odometry_detections.MergedOdometryDetections.create(
+            drone_odometry,
+            [detection_1],
+        )
+        assert result
+        assert merged_detections is not None
+
+        # Run
+        result, actual_list = basic_locator.run(merged_detections)
+
+        # Test
+        assert not result
+        assert actual_list is None