Radar¶

Radar Data¶

class py123d.datatypes.Radar[source]¶

Data structure for Radar point cloud data and associated metadata.

Public Data Attributes:

`radar_metadatas`	Returns the dictionary of per-radar metadata contained in this Radar's metadata.
`is_merged`	Returns True if this Radar contains merged data from multiple radars, False if it contains data from a single radar.
`metadata`	The `RadarMetadata` associated with this Radar recording.
`timestamp`	The timestamp associated with this Radar recording.
`point_cloud_3d`	Radar point cloud as an Nx3 numpy array, where N is the number of points, and the (x, y, z), indexed by `Point3DIndex`
`point_cloud_features`	The point cloud features as a dictionary of numpy arrays.
`xyz`	The point cloud as an Nx3 array of x, y, z coordinates.
`xy`	The point cloud as an Nx2 array of x, y coordinates.
`ids`	The point cloud as an Nx1 array of originating `RadarID` values, if available.
`timestamps`	The point cloud as an Nx1 array of timestamps in microseconds, if available.
`cluster_id`	The point cloud as an Nx1 array of cluster ids, if available.
`rcs`	The point cloud as an Nx1 array of radar cross section values (dBsm), if available.
`velocity`	The point cloud as an Nx2 array of raw (vx, vy) velocities in m/s, if available.
`velocity_comp`	The point cloud as an Nx2 array of ego-motion compensated (vx, vy) velocities in m/s, if available.

Inherited from BaseModality

`timestamp`	Returns the timestamp associated with this modality data, if available.
`metadata`	Returns the metadata associated with this modality data.
`modality_type`	Convenience property to access the modality type directly from the modality data.
`modality_id`	Convenience property to access the modality id directly from the modality data.
`modality_key`	Convenience property to access the modality key directly from the modality data.

Private Methods:

_feature(feature)

Returns the requested feature array cast to its canonical dtype, if available.

property modality_id: str | SerialIntEnum | None¶: Convenience property to access the modality id directly from the modality data.

property modality_key: str¶: Convenience property to access the modality key directly from the modality data.

property modality_type: ModalityType¶: Convenience property to access the modality type directly from the modality data.

property radar_metadatas: Dict[RadarID, RadarMetadata]¶

Returns the dictionary of per-radar metadata contained in this Radar’s metadata.

If the metadata is a RadarMergedMetadata, returns its internal dictionary. If the metadata is a single RadarMetadata, returns a dictionary with one entry.

property is_merged: bool¶: Returns True if this Radar contains merged data from multiple radars, False if it contains data from a single radar.

property metadata: RadarMetadata | RadarMergedMetadata¶: The RadarMetadata associated with this Radar recording.

property timestamp: Timestamp¶: The timestamp associated with this Radar recording.

property point_cloud_3d: ndarray[tuple[Any, ...], dtype[float32]]¶: Radar point cloud as an Nx3 numpy array, where N is the number of points, and the (x, y, z), indexed by Point3DIndex

property point_cloud_features: Dict[str, ndarray[tuple[Any, ...], dtype[_ScalarT]]] | None¶: The point cloud features as a dictionary of numpy arrays.

property xyz: ndarray[tuple[Any, ...], dtype[float32]]¶: The point cloud as an Nx3 array of x, y, z coordinates.

property xy: ndarray[tuple[Any, ...], dtype[float32]]¶: The point cloud as an Nx2 array of x, y coordinates.

property ids: ndarray[tuple[Any, ...], dtype[uint8]] | None¶: The point cloud as an Nx1 array of originating RadarID values, if available.

property timestamps: ndarray[tuple[Any, ...], dtype[int64]] | None¶: The point cloud as an Nx1 array of timestamps in microseconds, if available.

property cluster_id: ndarray[tuple[Any, ...], dtype[uint16]] | None¶: The point cloud as an Nx1 array of cluster ids, if available.

property rcs: ndarray[tuple[Any, ...], dtype[float32]] | None¶: The point cloud as an Nx1 array of radar cross section values (dBsm), if available.

property velocity: ndarray[tuple[Any, ...], dtype[float32]] | None¶: The point cloud as an Nx2 array of raw (vx, vy) velocities in m/s, if available.

property velocity_comp: ndarray[tuple[Any, ...], dtype[float32]] | None¶: The point cloud as an Nx2 array of ego-motion compensated (vx, vy) velocities in m/s, if available.

Radar Metadata¶

class py123d.datatypes.RadarMetadata[source]¶

Metadata for Radar sensor, static for a given sensor.

Public Data Attributes:

`radar_name`	The name of the Radar sensor from the dataset.
`radar_id`	The ID of the Radar sensor.
`radar_to_imu_se3`	The extrinsic `PoseSE3` of the Radar sensor, relative to the IMU frame.
`modality_type`	Returns the type of the modality that this metadata describes.
`modality_id`	Optional identifier for the modality, e.g. sensor ID for sensor modalities.

Inherited from BaseModalityMetadata

`modality_type`	Returns the type of the modality that this metadata describes.
`modality_id`	Optional identifier for the modality, e.g. sensor ID for sensor modalities.
`modality_key`	Returns a unique key for this modality, combining type and id if applicable.

Public Methods:

`from_dict`(data_dict)	Construct the Radar metadata from a dictionary.
`to_dict`()	Convert the Radar metadata to a dictionary.

Inherited from BaseMetadata

`to_dict`()	Serialize the metadata instance to a plain Python dictionary.
`from_dict`(data_dict)	Construct a metadata instance from a plain Python dictionary.

property radar_name: str¶: The name of the Radar sensor from the dataset.

property radar_id: RadarID¶: The ID of the Radar sensor.

property radar_to_imu_se3: PoseSE3¶: The extrinsic PoseSE3 of the Radar sensor, relative to the IMU frame.

property modality_type: ModalityType¶: Returns the type of the modality that this metadata describes.

property modality_id: str | SerialIntEnum | None¶: Optional identifier for the modality, e.g. sensor ID for sensor modalities. Can be a string or a SerialIntEnum.

classmethod from_dict(data_dict)[source]¶

Construct the Radar metadata from a dictionary.

Parameters:: data_dict (dict) – A dictionary containing Radar metadata.
Return type:: RadarMetadata
Returns:: An instance of RadarMetadata.

to_dict()[source]¶

Convert the Radar metadata to a dictionary.

Return type:: Dict[str, Any]
Returns:: A dictionary representation of the Radar metadata.

property modality_key: str¶: Returns a unique key for this modality, combining type and id if applicable.

class py123d.datatypes.RadarMergedMetadata[source]¶

Public Data Attributes:

`modality_type`	Returns the type of the modality that this metadata describes.
`modality_id`	Optional identifier for the modality, e.g. sensor ID for sensor modalities.
`radar_metadatas`	Returns the dictionary of per-radar metadata contained in this merged metadata.

Inherited from BaseModalityMetadata

`modality_type`	Returns the type of the modality that this metadata describes.
`modality_id`	Optional identifier for the modality, e.g. sensor ID for sensor modalities.
`modality_key`	Returns a unique key for this modality, combining type and id if applicable.

Public Methods:

`to_dict`()	Serialize the metadata instance to a plain Python dictionary.
`from_dict`(data_dict)	Construct a metadata instance from a plain Python dictionary.

Inherited from BaseMetadata

`to_dict`()	Serialize the metadata instance to a plain Python dictionary.
`from_dict`(data_dict)	Construct a metadata instance from a plain Python dictionary.

Inherited from Mapping

`get`(k[,d])
`keys`()
`items`()
`values`()

get(k[, d]) → D[k] if k in D, else d. d defaults to None.¶

items() → a set-like object providing a view on D's items¶

keys() → a set-like object providing a view on D's keys¶

property modality_key: str¶: Returns a unique key for this modality, combining type and id if applicable.

values() → an object providing a view on D's values¶

property modality_type: ModalityType¶: Returns the type of the modality that this metadata describes.

property modality_id: str | SerialIntEnum | None¶: Optional identifier for the modality, e.g. sensor ID for sensor modalities. Can be a string or a SerialIntEnum.

property radar_metadatas: Dict[RadarID, RadarMetadata]¶: Returns the dictionary of per-radar metadata contained in this merged metadata.

to_dict()[source]¶

Serialize the metadata instance to a plain Python dictionary.

Return type:: Dict[str, Any]
Returns:: A dictionary representation using only default Python types.

classmethod from_dict(data_dict)[source]¶

Construct a metadata instance from a plain Python dictionary.

Parameters:: data_dict (Dict[str, Any]) – A dictionary containing the metadata fields.
Return type:: RadarMergedMetadata
Returns:: A metadata instance.

Radar IDs¶

class py123d.datatypes.RadarID[source]¶

Enumeration of Radar sensors, in multi-sensor setups.

RADAR_UNKNOWN = 0¶: Unknown Radar type.

RADAR_MERGED = 1¶: Merged sensor Radar type.

RADAR_FRONT = 2¶: Front-facing (center) Radar type.

RADAR_FRONT_LEFT = 3¶: Front-left corner Radar type.

RADAR_FRONT_RIGHT = 4¶: Front-right corner Radar type.

RADAR_SIDE_LEFT = 5¶: Left-side Radar type.

RADAR_SIDE_RIGHT = 6¶: Right-side Radar type.

RADAR_BACK_LEFT = 7¶: Back-left corner Radar type.

RADAR_BACK_RIGHT = 8¶: Back-right corner Radar type.

RADAR_BACK = 9¶: Back-facing (center) Radar type.

RADAR_REAR_LEFT = 10¶: Rear-left-facing Radar type (distinct from the back-left corner radar).

RADAR_REAR_RIGHT = 11¶: Rear-right-facing Radar type (distinct from the back-right corner radar).

Radar Features¶

class py123d.datatypes.RadarFeature[source]¶

Enumeration of common Radar point cloud features.

Aligned with the (rich) nuScenes radar schema; datasets populate the subset they provide. Extending is additive: append a member here and a dtype in RADAR_FEATURE_DTYPES.

IDS = 0¶: Per-point originating RadarID value. Used for merge/split, same role as IDS.

TIMESTAMPS = 1¶: Per-point timestamp feature index, in microseconds.

CLUSTER_ID = 2¶: Per-point cluster id feature index (the nuScenes id field).

DYN_PROP = 3¶: Dynamic property feature index (moving/stationary classification, dataset-native).

RCS = 4¶: Radar cross section feature index, in dBsm.

VELOCITY_X = 5¶: Raw radial velocity x feature index, in m/s.

VELOCITY_Y = 6¶: Raw radial velocity y feature index, in m/s.

VELOCITY_X_COMP = 7¶: Ego-motion compensated velocity x feature index, in m/s.

VELOCITY_Y_COMP = 8¶: Ego-motion compensated velocity y feature index, in m/s.

IS_QUALITY_VALID = 9¶: Quality-valid state flag feature index.

AMBIG_STATE = 10¶: Doppler ambiguity state feature index.

X_RMS = 11¶: x position RMS feature index.

Y_RMS = 12¶: y position RMS feature index.

INVALID_STATE = 13¶: Invalid state flag feature index.

PDH0 = 14¶: False-alarm probability feature index.

VX_RMS = 15¶: Velocity x RMS feature index.

VY_RMS = 16¶: Velocity y RMS feature index.

SNR = 17¶: Signal-to-noise ratio feature index, in dB.

EXIST_PROBABILITY = 18¶: Detection existence probability feature index.