api.sensors package¶

Submodules¶

api.sensors.camera module¶

class api.sensors.camera.Camera(name, platform)¶

Bases: pioneer.das.api.sensors.sensor.Sensor

Camera sensor, expects ‘img’ datasource, intrinsics matrix and distortion coefficients

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

property camera_matrix¶

the 3x3 intrinsics matrix

Type: np.ndarray

property distortion_coeffs¶

Nx1 distortion coefficient, refer to opencv documentation

Type: np.ndarray

load_intrinsics(intrinsics_config)¶

Looks for a pickle file containing intrinsics information for this sensor, e.g. ‘eagle_tfc.pkl’

Parameters

intrinsics_config – path to folder containing this sensor’s intrinsics pickle file,
or relative to dataset path), e.g. '/nas/cam_intrinsics' or 'cam_intrinsics' ((absolute) –

api.sensors.carla_gps module¶

api.sensors.carla_imu module¶

class api.sensors.carla_imu.CarlaIMU(name, platform)¶

Bases: pioneer.das.api.sensors.sensor.Sensor

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

create_egomotion_provider()¶

api.sensors.egomotion_provider module¶

api.sensors.encoder module¶

class api.sensors.encoder.Encoder(name, platform)¶

Bases: pioneer.das.api.sensors.sensor.Sensor

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

create_egomotion_provider()¶

api.sensors.imu_egomotion_provider module¶

api.sensors.imu_sbg_ekinox module¶

class api.sensors.imu_sbg_ekinox.ImuSbgEkinox(name, platform)¶

Bases: pioneer.das.api.sensors.sensor.Sensor

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

create_egomotion_provider()¶

api.sensors.lca3 module¶

class api.sensors.lca3.LCA3(name, platform)¶

Bases: pioneer.das.api.sensors.lcax.LCAx

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

get_trace_smoothing_kernel()¶: Get the convolution kernel for trace processing for Eagle

api.sensors.lcax module¶

class api.sensors.lcax.LCAx(name, platform)¶

Bases: pioneer.das.api.sensors.sensor.Sensor

LeddarTech LCAx family of sensors, expects ‘ech’, ‘sta’ and ‘cfg’ datasources

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

apply_distance_corrections(timestamp, indices, distances)¶: Applies calibration and temperature-related distances corrections

apply_temperature_correction(timestamp, indices, distances)¶: Applies temperature-related distance corrections

property base_point_count¶

property binning¶

cache(specs: Dict[str, Union[int, float]]) → Dict[str, numpy.ndarray]¶

caches computation that are shared for all ‘ech’ samples.

Parameters: specs – a dict with keys, ‘v’: number of scan channels, ‘h’: the number of imaging channels, ‘v_fov’: scan field of view, in degrees, ‘h_fov’: imaging field of view, in degrees
Returns: A dict with keys ‘angles’ (the projection angles), ‘directions’ (the projection directions) and ‘quad_directions’ (the projection surface cloud directions)

Note

If ‘angle_chart’ is defined in this sensor’s intrinsics, the angle chart’s angles is used to compute ‘directions’, but not ‘quad_directions’. If ‘mirror_temp_compensation’ is defined in this sensor’s intrisincs, ‘angles_temp_lut’ and ‘directions_temp_lut’ will appear in cache and be used by apply_direction_corrections()

property distance_scaling¶: Get the distance in meters between two subsequent points in a trace

get_corrected_cloud(timestamp, cache, type, indices, distances, amplitudes=None, dtype=<class 'numpy.float64'>)¶

Returns the point-cloud (or quad-cloud) using the get_corrected_projection_data() method.

Parameters: type – ‘point_cloud’ or ‘quad_cloud’

get_corrected_projection_data(timestamp: Union[int, float], cache: dict, type: str = 'directions')¶

Returns temperature compensated projections directions or angles

Parameters

timestamp – the timestamp at which the sample is needed (use to obtain the right temperature)
cache – the cache to search in (obtained by calling LCAx.cache(…))
type – the type of directions (e.g) ‘directions’ or ‘quad_directions’. Must be present in the cache

Returns

the compensated directions if any, or the uncompensated direction otherwise

get_temperature_at(timestamp)¶

get_trace_smoothing_kernel()¶: Get the convolution kernel for trace processing for LCA2

load_intrinsics(intrinsics_config: Union[str, dict])¶

Load the LCAx intrinsics from a dictionnary like {‘v’: 8, ‘h’, 32, ‘v_fov’: 20, ‘h_fov’: 30}

Parameters: intrinsics_config – The intrinsics configuration

load_specs_from_cfg()¶

load_static_noise_from_cfg()¶

load_time_base_delays_from_cfg()¶

property oversampling¶

property trace_smoothing_kernel¶

api.sensors.motor_lidar module¶

class api.sensors.motor_lidar.MotorLidar(name, platform)¶

Bases: pioneer.das.api.sensors.sensor.Sensor

class TemperatureCompensation¶

Bases: enum.Enum

An enumeration.

Activated = 0¶

Deactivated = 1¶

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

apply_temperature_correction(timestamp, distances)¶: Applies temperature-related distance corrections

get_corrected_cloud(timestamp, pts, dtype)¶: It corrects the pts-cloud according to a temperature compensation (if any).

load_intrinsics(intrinsics_config: Union[str, dict])¶: load data from yml platform

api.sensors.mti module¶

api.sensors.pixell module¶

class api.sensors.pixell.Pixell(name, platform)¶

Bases: pioneer.das.api.sensors.lcax.LCAx

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

static angles_from_heads(distances, angles, head_positions)¶

cache(specs)¶

caches computation that are shared for all ‘ech’ samples.

Parameters: specs – a dict with keys, ‘v’: number of scan channels, ‘h’: the number of imaging channels, ‘v_fov’: scan field of view, in degrees, ‘h_fov’: imaging field of view, in degrees
Returns: A dict with keys ‘angles’ (the projection angles), ‘directions’ (the projection directions) and ‘quad_directions’ (the projection surface cloud directions)

Note

If ‘angle_chart’ is defined in this sensor’s intrinsics, the angle chart’s angles is used to compute ‘directions’, but not ‘quad_directions’. If ‘mirror_temp_compensation’ is defined in this sensor’s intrisincs, ‘angles_temp_lut’ and ‘directions_temp_lut’ will appear in cache and be used by apply_direction_corrections()

get_cloud(cache, type, indices, distances, amplitudes=None, dtype=<class 'numpy.float64'>)¶: Get cloud data (type_pts, type_amplitudes, type_indices) from a sensor.

get_corrected_cloud(timestamp, cache, type, indices, distances, amplitudes=None, dtype=<class 'numpy.float64'>)¶

Returns the point-cloud (or quad-cloud) using the get_corrected_projection_data() method.

Parameters: type – ‘point_cloud’ or ‘quad_cloud’

get_head_position_per_channel()¶

get_trace_smoothing_kernel()¶: Get the convolution kernel for trace processing for LCA2

load_intrinsics(intrinsics_config: Union[str, dict])¶: Basic intrinsics are brougth back under Sensor.

api.sensors.radar_ti module¶

class api.sensors.radar_ti.RadarTI(name, platform)¶

Bases: pioneer.das.api.sensors.sensor.Sensor

__init__(name, platform)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

get_corrected_cloud(timestamp, pts, dtype)¶

api.sensors.sensor module¶

class api.sensors.sensor.Sensor(name: str, pf, factories: Dict[str, Tuple[Any, Any]] = {}, call_start: Optional[Callable] = None, call_stop: Optional[Callable] = None)¶

Bases: object

A sensor encapsulate a sensor and its datasources. Important when you add a new derivation of class Sensor, don’t forget to add it to platform.SENSOR_FACTORY.

exception NoPathToReferential¶: Bases: Exception

__getitem__(key: str)¶

Returns the datasource

Parameters: key – the datasource type, e.g. ‘ech’

__init__(name: str, pf, factories: Dict[str, Tuple[Any, Any]] = {}, call_start: Optional[Callable] = None, call_stop: Optional[Callable] = None)¶

Constructor.

Parameters

name – the sensor’s name and position id, e.g. ‘eagle_tfc’
platform – the platform this sensor belongs to
factories – a dict containing one entry per datasource type, each entry’s value is a tuple containing (Sample-derived, optional_interpolator_function). For example, a factoryies dict could be {“ech”:(Echo, None), “sta”:(Sample, interpolators.linear_dict_of_float_interpolator)}
call_start – a callback to be called by Sensor.start() (e.g. to start an actual live sensor)
call_stop – similar to call_start

add_datasource(ds, ds_type: str, cache_size: int = 100)¶

Adds a datasouce to this sensor

Parameters

ds – a Filesource-derived instance or VirtualDatasource
ds_type – the datasource type, e.g. ‘ech’

create_egomotion_provider() → Optional[EgomotionProvider]¶

datasource_names() → List[str]¶: Returns this sensor’s datasource types

property extrinsics_dirty¶: The extrinsics dirty signal

invalidate_datasources_caches()¶: Invalidate sensor’s datasources caches

items()¶: Returns the datasource key,value iterable, implement dict API

keys()¶: Returns the datasource types, implement dict API

load_extrinsics(extrinsics_folder: str)¶

Looks for a pickle file containing extrinsics information for this sensor, named ‘From-To’ e.g. ‘flir_tfl-eagle_tfc.pkl’

Parameters

intrinsics_config – path to folder containing this sensor’s extrinsics pickle file
or relative to dataset path), e.g. '/nas/extrinsics' or 'extrinsics' ((absolute) –

load_intrinsics(intrinsics_config: str)¶

Looks for a pickle file containing intrinsics information for this sensor, e.g. ‘eagle_tfc.pkl’

Parameters

intrinsics_config – path to folder containing this sensor’s intrinsics pickle file,
or relative to dataset path), e.g. '/nas/cam_intrinsics' or 'cam_intrinsics' ((absolute) –

map_to(target: str) → numpy.ndarray¶

Returns a 4x4 tranform matrix mapping a point from this sensor’s referential to ‘target’ sensor’s referential

Parameters: target – the name of the target sensor in which referential we want to map to
Raises: Sensor.NoPathToReferential – if no mapping could be found

property platform¶

start()¶: Starts the live sensor wrapped by this Sensor instance

stop()¶: Stops the live sensor wrapped by this Sensor instance

api.sensors package¶

Submodules¶

api.sensors.camera module¶

api.sensors.carla_gps module¶

api.sensors.carla_imu module¶

api.sensors.egomotion_provider module¶

api.sensors.encoder module¶

api.sensors.imu_egomotion_provider module¶

api.sensors.imu_sbg_ekinox module¶

api.sensors.lca3 module¶

api.sensors.lcax module¶

api.sensors.motor_lidar module¶

api.sensors.mti module¶

api.sensors.pixell module¶

api.sensors.radar_ti module¶

api.sensors.sensor module¶

Module contents¶