<div align="center">
<h1>Trajectopy - Trajectory Evaluation in Python</h1>
<a href="https://github.com/gereon-t/trajectopy-core/releases"><img src="https://img.shields.io/github/v/release/gereon-t/trajectopy-core?label=version" /></a>
<a href="https://www.python.org/downloads/"><img src="https://img.shields.io/badge/python-3.8.2+-blue.svg" /></a>
<a href="https://github.com/gereon-t/trajectopy-core/blob/main/LICENSE"><img src="https://img.shields.io/github/license/gereon-t/trajectopy-core" /></a>
<a href="https://github.com/psf/black"><img alt="Code style: black" src="https://img.shields.io/badge/code%20style-black-000000.svg"></a>
<br />
<a href="https://github.com/gereon-t/trajectopy-core"><img src="https://img.shields.io/badge/Windows-0078D6?st&logo=windows&logoColor=white" /></a>
<a href="https://github.com/gereon-t/trajectopy-core"><img src="https://img.shields.io/badge/Linux-FCC624?logo=linux&logoColor=black" /></a>
<a href="https://github.com/gereon-t/trajectopy-core"><img src="https://img.shields.io/badge/mac%20os-000000?&logo=apple&logoColor=white" /></a>
#### Trajectopy is a toolbox for empirical trajectory evaluation.
</div>
## Key Features
Trajectopy offers a range of features, including:
- Absolute Trajectory Error (__ATE__) computation
- Relative Pose Error (__RPE__) computation with distance- or time-based pose-pair selection
- Trajectory __alignment__ using least squares adjustment theory and up to 11 parameters (i.e. similarity transformation, lever arm, time shift)
- Customizable __HTML report__ generation ([demo](https://htmlpreview.github.io/?https://github.com/gereon-t/trajectopy-core/blob/main/example_data/report.html)) (takes a few seconds to load)
## Table of Contents
- [Installation](#installation)
- [Exemplary Evaluation](#exemplary-evaluation)
- [Importing Trajectories](#importing-trajectories)
- [Processing Settings](#processing-settings)
## Installation
#### Create virtual environment (Optional but recommended)
Unix
```console
python3 -m venv .venv
```
Windows
```console
python -m venv .venv
```
Activate (Unix)
```console
source .venv/bin/activate
```
**or** (Windows)
```console
.\.venv\Scripts\activate
```
#### Installation via pip
Update pip:
```console
pip install --upgrade pip
```
Install trajectopy:
```console
pip install trajectopy-core
```
### Or using the repository:
```console
pip install git+https://github.com/gereon-t/trajectopy-core.git@main
```
## Exemplary Evaluation
This section shows how to use trajectopy to evaluate two trajectories. The example data can be found in the example_data folder. The full code can be found in the example_scripts folder.
### Absolute Trajectory Error (ATE)
```python
from trajectopy_core.evaluation.metrics import ate
from trajectopy_core.settings.processing import ProcessingSettings
from trajectopy_core.trajectory import Trajectory
# Import
gt_traj = Trajectory.from_file("./example_data/KITTI_gt.traj")
est_traj = Trajectory.from_file("./example_data/KITTI_ORB.traj")
# default settings
settings = ProcessingSettings()
ate_result = ate(trajectory_gt=gt_traj, trajectory_est=est_traj, settings=settings)
```
### Relative Pose Error (RPE)
```python
from trajectopy_core.evaluation.metrics import rpe
from trajectopy_core.settings.processing import ProcessingSettings
from trajectopy_core.trajectory import Trajectory
# Import
gt_traj = Trajectory.from_file("./example_data/KITTI_gt.traj")
est_traj = Trajectory.from_file("./example_data/KITTI_ORB.traj")
# default settings
settings = ProcessingSettings()
rpe_result = rpe(trajectory_gt=gt_traj, trajectory_est=est_traj, settings=settings)
```
## Importing Trajectories
Trajectory files must be ASCII files with a csv-like layout. The default column structure that can be read without any configuration is the following:
| time | position x | position y | position z | quaternion x | quaternion y | quaternion z | quaternion w |
| ---- | ---------- | ---------- | ---------- | ------------ | ------------ | ------------ | ------------ |
Columns are expected to be separated by commas by default.
It is recommended to provide a header at the beginning of the trajectory file. Header entries always begin with a "#".
Below you can find a table of all allowed header entries and their meaning.
| Header | Description |
| ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| #name | The name provided here is displayed in plots of the trajectory |
| #epsg | [EPSG Code](https://epsg.io/) of the datum of the input positions. Required, if geodetic datum transformations are desired. Default: 0, meaning local coordinates without any known geodetic datum |
| #fields | Describes the columns of the ASCII trajectory file. Separated with commas. <table> <thead> <th>field name</th> <th>Meaning</th> </tr> </thead> <tbody> <tr> <td>t</td> <td>time</td> </tr> <tr> <td>l</td> <td>arc lengths in meters</td> </tr> <tr> <td>px</td> <td>position x / lat (degrees only)</td> </tr> <tr> <td>py</td> <td>position y / lon (degrees only) </td> </tr> <tr> <td>pz</td> <td>position z</td> </tr> <tr> <td>qx</td> <td>quaternion x</td> </tr> <tr> <td>qy</td> <td>quaternion y</td> </tr> <tr> <td>qz</td> <td>quaternion z</td> </tr> <tr> <td>qw</td> <td>quaternion w</td> </tr> </tr> <tr> <td>ex</td> <td>euler angle x</td> </tr> </tr> <tr> <td>ey</td> <td>euler angle y</td> </tr> </tr> <tr> <td>ez</td> <td>euler angle z</td> </tr> </tr> <tr> <td>vx</td> <td>speed x</td> </tr> </tr> <tr> <td>vy</td> <td>speed y</td> </tr> </tr> <tr> <td>vz</td> <td>speed z</td> </tr> </tr> </tbody> </table> Example: "#fields t,px,py,pz" Note: The only column that is allowed to appear multiple times is the "t" column. |
| #delimiter | Delimiter used to separate the columns within the file. Default: "," |
| #nframe | Definition of the navigation-frame the orientations of the trajectory refer to. Choices: "enu": East North Up or "ned": North East Down. Default: "enu" |
| #rot_unit | Unit of the orientations. Choices: "deg": Degree, "rad": Radians. Default: "rad" |
| #time_format | Format of the timestamps / dates. Choices: "unix": Unix timestamps (since 01-01-1970), "datetime": Human readable date-times. Default: "unix" |
| #time_offset | Offset in seconds that is applied to the imported timestamps. Default: 0.0 |
| #datetime_format | Format of the datetimes. Only relevant if "time_format" is "datetime". Default: "%Y-%m-%d %H:%M:%S.%f" |
| #datetime_timezone | Time zone of the timestamps. During import, all timestamps are converted to UTC considering the input time zone. Choices: [Time zone](https://en.wikipedia.org/wiki/List_of_tz_database_time_zones) or "GPS" |
| #sorting | Sorting of the input data. Choices: "chrono": Chronologically sorted data (usually the case), "spatial": Spatially sorted data, i.e. along the arc length. Default: "chrono" |
**New since 0.9.2**: Experimental ROS bag support for geometry_msgs/msg/PoseStamped messages. Files must have a ".bag" extension. Poses must have positions and orientations. One file can contain multiple trajectories published under different topics.
## Processing Settings
Trajectopy offers a range of processing options that can be applied to the imported trajectories. These options are:
| Option | Description |
| ---------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Alignment | Alignment of two trajectories using least squares adjustment. The implemented approach can handle a similarity transformation (translation, rotation, scale), a lever arm (3d vector), and a time shift (scalar). Each parameter can be included or exluded from the adjustment depending on the individual sensor modalities using the `AlignmentSettings`. In addition, preprocessing steps and stochastics can also be configured. |
| Matching | Matching of two trajectories to establish pose-to-pose correspondencies. After matching both trajectories will have the same number of poses. You can choose from different matching methods in the `MatchingSettings`. |
| Comparison | Comparison of two trajectories using absolute (ATE) and relative (RPE) metrics. The relative comparison can be configured using the `RelativeComparisonSettings`. |
## Alignment Settings
### Preprocessing Settings
- `min_speed` (float): Only poses with a speed above this threshold are considered for alignment.
- `time_start` (float): Only poses with a timestamp above this threshold are considered for alignment. The timestamp is given in seconds and is relative to the first common timestamp of both matched trajectories.
- `time_end` (float): Only poses with a timestamp below this threshold are considered for alignment. The timestamp is given in seconds and is relative to the first common timestamp of both matched trajectories.
### Estimation Settings
- `trans_x` (boolean): Enable or disable x-translation of the similarity transformation.
- `trans_y` (boolean): Enable or disable y-translation of the similarity transformation.
- `trans_z` (boolean): Enable or disable z-translation of the similarity transformation.
- `rot_x` (boolean): Enable or disable rotation around the X-axis of the similarity transformation.
- `rot_y` (boolean): Enable or disable rotation around the Y-axis of the similarity transformation.
- `rot_z` (boolean): Enable or disable rotation around the Z-axis of the similarity transformation.
- `scale` (boolean): Enable or disable scaling of the similarity transformation.
- `time_shift` (boolean): Enable or disable the estimation of time shift.
- `use_x_speed` (boolean): Enable or disable the use of X-axis speed for time shift estimation.
- `use_y_speed` (boolean): Enable or disable the use of Y-axis speed for time shift estimation.
- `use_z_speed` (boolean): Enable or disable the use of Z-axis speed for time shift estimation.
- `lever_x` (boolean): Enable or disable estimation of lever arm in the X-axis.
- `lever_y` (boolean): Enable or disable estimation of lever arm in the Y-axis.
- `lever_z` (boolean): Enable or disable estimation of lever arm in the Z-axis.
- `sensor_rotation` (boolean): Enable or disable estimation of sensor rotation. Independent of the least squares adjustment, a constant rotational offset can be computed between the rotations of both trajectories after the alignment.
### Stochastics Settings
- `std_xy_from` (float): Standard deviation of XY source position components in meters.
- `std_z_from` (float): Standard deviation of Z source position component in meters.
- `std_xy_to` (float): Standard deviation of XY target position components in meters.
- `std_z_to` (float): Standard deviation of Z target position component in meters.
- `std_roll_pitch` (float): Standard deviation of roll and pitch in radians.
- `std_yaw` (float): Standard deviation of yaw in radians.
- `std_speed_to` (float): Standard deviation of platform speed in (meters per second).
- `error_probability` (float): Probability of error used for stochastic testing.
- `variance_estimation` (boolean): Enable or disable the estimation of the variance factor for a-posteriori variance computation.
### Threshold Settings
Usually, these settings can be left at their default values.
- `metric_threshold` (float): Iteration threshold for the least squares adjustment regarding the metric parameters.
- `time_threshold` (float): Iteration threshold for the least squares adjustment regarding the time shift parameter.
## Matching Settings
- `method` (`MatchingMethod`): The method used for trajectory matching. Choices: `MatchingMethod.NEAREST_SPATIAL`, `MatchingMethod.NEAREST_TEMPORAL`, `MatchingMethod.INTERPOLATION`, `MatchingMethod.NEAREST_SPATIAL_INTERPOLATED`. The methods are described below.
- `max_time_diff` (float): Maximum allowed time difference when matching two trajectories using their timestamps.
- `max_distance` (float): Maximum allowed distance between matched positions during spatial matching.
- `k_nearest` (integer): The number of nearest neighbors to consider during spatial interpolation matching.
### Matching Methods
#### Nearest Spatial
This method matches two trajectories by finding the nearest pose in the target trajectory for each pose in the source trajectory. The distance between two poses is computed using the Euclidean distance between their positions.
#### Nearest Temporal
This method matches two trajectories using their timestamps by finding the nearest timestamp in the target trajectory for each timestamp in the source trajectory.
#### Interpolation
This method matches two trajectories by interpolating the timestamps of one trajectory to the timestamps of the other trajectory. The interpolation is linear for both positions and rotations (SLERP).
#### Nearest Spatial Interpolated
This method matches both trajectories spatially by requesting the nearest k positions from the reference trajectory for each pose in the test trajectory. Then, an interpolation is performed using a 3d line fit of the k nearest positions. After this operation, both trajectories will have the length of the test trajectory. This method does not support rotation matching.
## Relative Comparison Settings
- `pair_min_distance` (float): Minimum pose pair distance to be considered during RPE (Relative Pose Error) computation.
- `pair_max_distance` (float): Maximum pose pair distance to be considered during RPE computation.
- `pair_distance_step` (float): Step in which the pose pair distance increases.
- `pair_distance_unit` (`Unit`): Unit of the pose pair distance. Choices: `Unit.METER`, `Unit.SECOND`.
- `use_all_pose_pairs` (boolean): If enabled, overlapping pose pairs will be used for relative metrics calculation.
#### RPE Background
For this metric, relative pose-pair differences are compared. The distance between two poses can be specified by the user and can be either time- or distance-based. The comparison involves finding pose pairs separated by a specific distance or time interval, computing the relative translation and rotation between the reference and estimated pose pairs, and calculating the translational and rotational difference normalized by the distance or time that separated the poses.
1. Find pose pair separated by e.g. 100 m in reference trajectory. This pair represents the start and end poses of a sequence of size $N$.
2. Find the corresponding pose pair in estimated trajectory
3. Compute relative translation and rotation between the reference pose pair
$\Delta_{P~ref} = P_{ref, 1}^{-1} \cdot P_{ref, N}$
4. Compute relative translation and rotation between the estimated pose pair
$\Delta_{P~est} = P_{est, 1}^{-1} \cdot P_{est, N}$
5. Compute transformation between 3) and 4)
6. Compute translation and rotation error from 5)
7. Divide 6) by the distance or the time that separated both poses (e.g. 100 m).
This metric does not require both trajectories to be aligned. Units are m/m: %, deg/m for distance based comparison and m/s, deg/s for time-based comparison.
Example:
- Minimum pose distance: 100
- Maximum pose distance: 800
- Distance step: 100
- Distance unit: Meter
Results in pose distances: [100 m, 200 m, 300 m, 400 m, 500 m, 600 m, 700 m, 800 m]
Furthermore, the user can choose to either use consecutive pose pairs (non-overlapping) or all posible pairs (overlapping).
### Report Settings
#### Visualization Settings
- `scatter_max_std` (float): The upper colorbar limit is set to the mean plus this value times the standard deviation of the data. This is useful to prevent outliers from dominating the colorbar. Default value is 4.0.
- `ate_unit_is_mm` (bool): Indicates whether the unit of Absolute Trajectory Error (ATE) is millimeters. Default value is False.
- `directed_ate` (bool): Indicates whether the ATE is split into along-, horizontal-cross- and vertical-cross-track direction. Default value is True.
- `histogram_opacity` (float): The opacity of the histogram bars. Default value is 0.7.
- `histogram_bargap` (float): The gap between histogram bars. Default value is 0.1.
- `histogram_barmode` (str): The mode of displaying histogram bars. Default value is "overlay".
- `histogram_yaxis_title` (str): The title of the y-axis in the histogram. Default value is "Count".
- `plot_mode` (str): The mode of displaying plots. Default value is "lines+markers".
- `scatter_mode` (str): The mode of displaying scatter plots. Default value is "markers".
- `scatter_colorscale` (str): The colorscale for scatter plots. Default value is "RdYlBu_r".
- `scatter_axis_order` (str): The order of the axes in scatter plots. Default value is "xy". If 3d plotting is desired, also specify "z".
- `scatter_marker_size` (int): The size of markers in scatter plots. Default value is 5.
- `scatter_detailed` (bool): Indicates whether to show scatter plots for each degree of freedom. Default value is False.
#### Mapbox Settings
These settings currently only apply to trajectory only plots without deviations.
- `scatter_mapbox` (bool): Indicates whether the trajectory should be plotted onto a map. Default value is False.
- `scatter_mapbox_style` (str): The style of the map. For some styles, a Mapbox token is required. Default value is "open-street-map".
- `scatter_mapbox_zoom` (int): The zoom level of the map. Default value is 15.
- `scatter_mapbox_token` (str): The Mapbox token. Default value is "".
The mapbox token can be obtained from [https://www.mapbox.com/](https://www.mapbox.com/) after creating a free account.
#### Position Units and Names
- `pos_x_name` (string): Name for the X-axis position. Default: "x".
- `pos_y_name` (string): Name for the Y-axis position. Default: "y".
- `pos_z_name` (string): Name for the Z-axis position. Default: "z".
- `pos_x_unit` (string): Unit for the X-axis position, Default: "m".
- `pos_y_unit` (string): Unit for the Y-axis position, Default: "m".
- `pos_z_unit` (string): Unit for the Z-axis position, Default: "m".
- `pos_dir_dev_x_name` (string): Name for the directed position deviation in along-track direction. Default: "along".
- `pos_dir_dev_y_name` (string): Name for the directed position deviation in horizontal cross-track direction. Default: "cross-h".
- `pos_dir_dev_z_name` (string): Name for the directed position deviation in vertical cross-track direction. Default: "cross-v".
#### Rotation Units and Names
- `rot_x_name` (string): Name for the roll rotation. Default: "roll".
- `rot_y_name` (string): Name for the pitch rotation. Default: "pitch".
- `rot_z_name` (string): Name for the yaw rotation. Default: "yaw".
- `rot_unit` (string): Unit symbol for rotation. Default: "°".
#### Export Settings
- `single_plot_export` (ExportSettings): The export settings for single plots. Default value is an instance of ExportSettings with width=800 and height=450.
- `two_subplots_export` (ExportSettings): The export settings for two subplots. Default value is an instance of ExportSettings with width=800 and height=540.
- `three_subplots_export` (ExportSettings): The export settings for three subplots. Default value is an instance of ExportSettings with width=800 and height=750.
- `single_plot_height` (int): The height of a single plot. Default value is 450.
- `two_subplots_height` (int): The height of two subplots. Default value is 540.
- `three_subplots_height` (int): The height of three subplots. Default value is 750.
#### Export Settings
- `format` (string): The export format. Choices: "png", "svg", "jpeg", "webp". Default: "png".
- `height` (integer): The export height in pixels. Default: 500.
- `width` (integer): The export width in pixels. Default: 800.
- `scale` (integer): The export scale. Default: 6.
Raw data
{
"_id": null,
"home_page": "https://github.com/gereon-t/trajectopy-core",
"name": "trajectopy-core",
"maintainer": "Gereon Tombrink",
"docs_url": null,
"requires_python": "<3.13,>=3.9",
"maintainer_email": "tombrink@igg.uni-bonn.de",
"keywords": "trajectory, evaluation, alignment, similarity, leverarm, epsg, robotics",
"author": "Gereon Tombrink",
"author_email": "tombrink@igg.uni-bonn.de",
"download_url": "https://files.pythonhosted.org/packages/93/b0/14478065202efa77e7317c74d891a38941dd86d1c03d1d2132a0d9e279f9/trajectopy_core-3.0.0.tar.gz",
"platform": null,
"description": "<div align=\"center\">\n <h1>Trajectopy - Trajectory Evaluation in Python</h1>\n <a href=\"https://github.com/gereon-t/trajectopy-core/releases\"><img src=\"https://img.shields.io/github/v/release/gereon-t/trajectopy-core?label=version\" /></a>\n <a href=\"https://www.python.org/downloads/\"><img src=\"https://img.shields.io/badge/python-3.8.2+-blue.svg\" /></a>\n <a href=\"https://github.com/gereon-t/trajectopy-core/blob/main/LICENSE\"><img src=\"https://img.shields.io/github/license/gereon-t/trajectopy-core\" /></a>\n <a href=\"https://github.com/psf/black\"><img alt=\"Code style: black\" src=\"https://img.shields.io/badge/code%20style-black-000000.svg\"></a>\n <br />\n <a href=\"https://github.com/gereon-t/trajectopy-core\"><img src=\"https://img.shields.io/badge/Windows-0078D6?st&logo=windows&logoColor=white\" /></a>\n <a href=\"https://github.com/gereon-t/trajectopy-core\"><img src=\"https://img.shields.io/badge/Linux-FCC624?logo=linux&logoColor=black\" /></a>\n <a href=\"https://github.com/gereon-t/trajectopy-core\"><img src=\"https://img.shields.io/badge/mac%20os-000000?&logo=apple&logoColor=white\" /></a>\n\n#### Trajectopy is a toolbox for empirical trajectory evaluation. \n</div>\n\n\n## Key Features\n\nTrajectopy offers a range of features, including:\n\n- Absolute Trajectory Error (__ATE__) computation\n\n- Relative Pose Error (__RPE__) computation with distance- or time-based pose-pair selection\n\n- Trajectory __alignment__ using least squares adjustment theory and up to 11 parameters (i.e. similarity transformation, lever arm, time shift)\n\n- Customizable __HTML report__ generation ([demo](https://htmlpreview.github.io/?https://github.com/gereon-t/trajectopy-core/blob/main/example_data/report.html)) (takes a few seconds to load)\n\n## Table of Contents\n- [Installation](#installation)\n- [Exemplary Evaluation](#exemplary-evaluation)\n- [Importing Trajectories](#importing-trajectories)\n- [Processing Settings](#processing-settings)\n\n\n## Installation\n\n#### Create virtual environment (Optional but recommended)\nUnix\n\n```console\npython3 -m venv .venv\n```\n\nWindows\n\n```console\npython -m venv .venv\n```\n\nActivate (Unix)\n\n```console\nsource .venv/bin/activate\n```\n\n**or** (Windows)\n```console\n.\\.venv\\Scripts\\activate\n```\n\n#### Installation via pip\nUpdate pip:\n\n```console\npip install --upgrade pip\n```\n\nInstall trajectopy:\n\n```console\npip install trajectopy-core\n```\n\n### Or using the repository:\n\n```console\npip install git+https://github.com/gereon-t/trajectopy-core.git@main\n```\n\n## Exemplary Evaluation\n\nThis section shows how to use trajectopy to evaluate two trajectories. The example data can be found in the example_data folder. The full code can be found in the example_scripts folder.\n\n### Absolute Trajectory Error (ATE)\n\n \n```python\nfrom trajectopy_core.evaluation.metrics import ate\nfrom trajectopy_core.settings.processing import ProcessingSettings\nfrom trajectopy_core.trajectory import Trajectory\n\n# Import\ngt_traj = Trajectory.from_file(\"./example_data/KITTI_gt.traj\")\nest_traj = Trajectory.from_file(\"./example_data/KITTI_ORB.traj\")\n\n# default settings\nsettings = ProcessingSettings()\n\nate_result = ate(trajectory_gt=gt_traj, trajectory_est=est_traj, settings=settings)\n\n```\n\n### Relative Pose Error (RPE)\n\n```python\nfrom trajectopy_core.evaluation.metrics import rpe\nfrom trajectopy_core.settings.processing import ProcessingSettings\nfrom trajectopy_core.trajectory import Trajectory\n\n# Import\ngt_traj = Trajectory.from_file(\"./example_data/KITTI_gt.traj\")\nest_traj = Trajectory.from_file(\"./example_data/KITTI_ORB.traj\")\n\n# default settings\nsettings = ProcessingSettings()\n\nrpe_result = rpe(trajectory_gt=gt_traj, trajectory_est=est_traj, settings=settings)\n\n```\n\n\n## Importing Trajectories\n\nTrajectory files must be ASCII files with a csv-like layout. The default column structure that can be read without any configuration is the following:\n\n| time | position x | position y | position z | quaternion x | quaternion y | quaternion z | quaternion w |\n| ---- | ---------- | ---------- | ---------- | ------------ | ------------ | ------------ | ------------ |\n\n\nColumns are expected to be separated by commas by default.\n\nIt is recommended to provide a header at the beginning of the trajectory file. Header entries always begin with a \"#\".\nBelow you can find a table of all allowed header entries and their meaning.\n\n| Header | Description |\n| ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| #name | The name provided here is displayed in plots of the trajectory |\n| #epsg | [EPSG Code](https://epsg.io/) of the datum of the input positions. Required, if geodetic datum transformations are desired. Default: 0, meaning local coordinates without any known geodetic datum |\n| #fields | Describes the columns of the ASCII trajectory file. Separated with commas. <table> <thead> <th>field name</th> <th>Meaning</th> </tr> </thead> <tbody> <tr> <td>t</td> <td>time</td> </tr> <tr> <td>l</td> <td>arc lengths in meters</td> </tr> <tr> <td>px</td> <td>position x / lat (degrees only)</td> </tr> <tr> <td>py</td> <td>position y / lon (degrees only) </td> </tr> <tr> <td>pz</td> <td>position z</td> </tr> <tr> <td>qx</td> <td>quaternion x</td> </tr> <tr> <td>qy</td> <td>quaternion y</td> </tr> <tr> <td>qz</td> <td>quaternion z</td> </tr> <tr> <td>qw</td> <td>quaternion w</td> </tr> </tr> <tr> <td>ex</td> <td>euler angle x</td> </tr> </tr> <tr> <td>ey</td> <td>euler angle y</td> </tr> </tr> <tr> <td>ez</td> <td>euler angle z</td> </tr> </tr> <tr> <td>vx</td> <td>speed x</td> </tr> </tr> <tr> <td>vy</td> <td>speed y</td> </tr> </tr> <tr> <td>vz</td> <td>speed z</td> </tr> </tr> </tbody> </table> Example: \"#fields t,px,py,pz\" Note: The only column that is allowed to appear multiple times is the \"t\" column. |\n| #delimiter | Delimiter used to separate the columns within the file. Default: \",\" |\n| #nframe | Definition of the navigation-frame the orientations of the trajectory refer to. Choices: \"enu\": East North Up or \"ned\": North East Down. Default: \"enu\" |\n| #rot_unit | Unit of the orientations. Choices: \"deg\": Degree, \"rad\": Radians. Default: \"rad\" |\n| #time_format | Format of the timestamps / dates. Choices: \"unix\": Unix timestamps (since 01-01-1970), \"datetime\": Human readable date-times. Default: \"unix\" |\n| #time_offset | Offset in seconds that is applied to the imported timestamps. Default: 0.0 |\n| #datetime_format | Format of the datetimes. Only relevant if \"time_format\" is \"datetime\". Default: \"%Y-%m-%d %H:%M:%S.%f\" |\n| #datetime_timezone | Time zone of the timestamps. During import, all timestamps are converted to UTC considering the input time zone. Choices: [Time zone](https://en.wikipedia.org/wiki/List_of_tz_database_time_zones) or \"GPS\" |\n| #sorting | Sorting of the input data. Choices: \"chrono\": Chronologically sorted data (usually the case), \"spatial\": Spatially sorted data, i.e. along the arc length. Default: \"chrono\" |\n\n**New since 0.9.2**: Experimental ROS bag support for geometry_msgs/msg/PoseStamped messages. Files must have a \".bag\" extension. Poses must have positions and orientations. One file can contain multiple trajectories published under different topics.\n\n## Processing Settings\n\nTrajectopy offers a range of processing options that can be applied to the imported trajectories. These options are:\n\n| Option | Description |\n| ---------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| Alignment | Alignment of two trajectories using least squares adjustment. The implemented approach can handle a similarity transformation (translation, rotation, scale), a lever arm (3d vector), and a time shift (scalar). Each parameter can be included or exluded from the adjustment depending on the individual sensor modalities using the `AlignmentSettings`. In addition, preprocessing steps and stochastics can also be configured. |\n| Matching | Matching of two trajectories to establish pose-to-pose correspondencies. After matching both trajectories will have the same number of poses. You can choose from different matching methods in the `MatchingSettings`. |\n| Comparison | Comparison of two trajectories using absolute (ATE) and relative (RPE) metrics. The relative comparison can be configured using the `RelativeComparisonSettings`. |\n\n## Alignment Settings\n\n### Preprocessing Settings\n\n- `min_speed` (float): Only poses with a speed above this threshold are considered for alignment.\n- `time_start` (float): Only poses with a timestamp above this threshold are considered for alignment. The timestamp is given in seconds and is relative to the first common timestamp of both matched trajectories.\n- `time_end` (float): Only poses with a timestamp below this threshold are considered for alignment. The timestamp is given in seconds and is relative to the first common timestamp of both matched trajectories.\n\n### Estimation Settings\n\n- `trans_x` (boolean): Enable or disable x-translation of the similarity transformation.\n- `trans_y` (boolean): Enable or disable y-translation of the similarity transformation.\n- `trans_z` (boolean): Enable or disable z-translation of the similarity transformation.\n- `rot_x` (boolean): Enable or disable rotation around the X-axis of the similarity transformation.\n- `rot_y` (boolean): Enable or disable rotation around the Y-axis of the similarity transformation.\n- `rot_z` (boolean): Enable or disable rotation around the Z-axis of the similarity transformation.\n- `scale` (boolean): Enable or disable scaling of the similarity transformation.\n- `time_shift` (boolean): Enable or disable the estimation of time shift.\n- `use_x_speed` (boolean): Enable or disable the use of X-axis speed for time shift estimation.\n- `use_y_speed` (boolean): Enable or disable the use of Y-axis speed for time shift estimation.\n- `use_z_speed` (boolean): Enable or disable the use of Z-axis speed for time shift estimation.\n- `lever_x` (boolean): Enable or disable estimation of lever arm in the X-axis.\n- `lever_y` (boolean): Enable or disable estimation of lever arm in the Y-axis.\n- `lever_z` (boolean): Enable or disable estimation of lever arm in the Z-axis.\n- `sensor_rotation` (boolean): Enable or disable estimation of sensor rotation. Independent of the least squares adjustment, a constant rotational offset can be computed between the rotations of both trajectories after the alignment.\n\n### Stochastics Settings\n\n- `std_xy_from` (float): Standard deviation of XY source position components in meters.\n- `std_z_from` (float): Standard deviation of Z source position component in meters.\n- `std_xy_to` (float): Standard deviation of XY target position components in meters.\n- `std_z_to` (float): Standard deviation of Z target position component in meters.\n- `std_roll_pitch` (float): Standard deviation of roll and pitch in radians.\n- `std_yaw` (float): Standard deviation of yaw in radians.\n- `std_speed_to` (float): Standard deviation of platform speed in (meters per second).\n- `error_probability` (float): Probability of error used for stochastic testing.\n- `variance_estimation` (boolean): Enable or disable the estimation of the variance factor for a-posteriori variance computation.\n\n### Threshold Settings\n\nUsually, these settings can be left at their default values.\n\n- `metric_threshold` (float): Iteration threshold for the least squares adjustment regarding the metric parameters.\n- `time_threshold` (float): Iteration threshold for the least squares adjustment regarding the time shift parameter.\n\n\n## Matching Settings\n\n- `method` (`MatchingMethod`): The method used for trajectory matching. Choices: `MatchingMethod.NEAREST_SPATIAL`, `MatchingMethod.NEAREST_TEMPORAL`, `MatchingMethod.INTERPOLATION`, `MatchingMethod.NEAREST_SPATIAL_INTERPOLATED`. The methods are described below.\n- `max_time_diff` (float): Maximum allowed time difference when matching two trajectories using their timestamps.\n- `max_distance` (float): Maximum allowed distance between matched positions during spatial matching.\n- `k_nearest` (integer): The number of nearest neighbors to consider during spatial interpolation matching.\n\n### Matching Methods\n\n#### Nearest Spatial\n\nThis method matches two trajectories by finding the nearest pose in the target trajectory for each pose in the source trajectory. The distance between two poses is computed using the Euclidean distance between their positions.\n\n#### Nearest Temporal\n\nThis method matches two trajectories using their timestamps by finding the nearest timestamp in the target trajectory for each timestamp in the source trajectory.\n\n#### Interpolation\n\nThis method matches two trajectories by interpolating the timestamps of one trajectory to the timestamps of the other trajectory. The interpolation is linear for both positions and rotations (SLERP).\n\n#### Nearest Spatial Interpolated\n\nThis method matches both trajectories spatially by requesting the nearest k positions from the reference trajectory for each pose in the test trajectory. Then, an interpolation is performed using a 3d line fit of the k nearest positions. After this operation, both trajectories will have the length of the test trajectory. This method does not support rotation matching.\n\n\n## Relative Comparison Settings\n\n- `pair_min_distance` (float): Minimum pose pair distance to be considered during RPE (Relative Pose Error) computation.\n\n- `pair_max_distance` (float): Maximum pose pair distance to be considered during RPE computation.\n\n- `pair_distance_step` (float): Step in which the pose pair distance increases.\n\n- `pair_distance_unit` (`Unit`): Unit of the pose pair distance. Choices: `Unit.METER`, `Unit.SECOND`.\n\n- `use_all_pose_pairs` (boolean): If enabled, overlapping pose pairs will be used for relative metrics calculation.\n\n\n#### RPE Background\n\nFor this metric, relative pose-pair differences are compared. The distance between two poses can be specified by the user and can be either time- or distance-based. The comparison involves finding pose pairs separated by a specific distance or time interval, computing the relative translation and rotation between the reference and estimated pose pairs, and calculating the translational and rotational difference normalized by the distance or time that separated the poses.\n\n1. Find pose pair separated by e.g. 100 m in reference trajectory. This pair represents the start and end poses of a sequence of size $N$.\n2. Find the corresponding pose pair in estimated trajectory\n3. Compute relative translation and rotation between the reference pose pair\n\n $\\Delta_{P~ref} = P_{ref, 1}^{-1} \\cdot P_{ref, N}$\n\n4. Compute relative translation and rotation between the estimated pose pair\n\n $\\Delta_{P~est} = P_{est, 1}^{-1} \\cdot P_{est, N}$\n\n5. Compute transformation between 3) and 4)\n\n6. Compute translation and rotation error from 5)\n\n7. Divide 6) by the distance or the time that separated both poses (e.g. 100 m).\n\nThis metric does not require both trajectories to be aligned. Units are m/m: %, deg/m for distance based comparison and m/s, deg/s for time-based comparison. \n\nExample:\n\n- Minimum pose distance: 100\n- Maximum pose distance: 800\n- Distance step: 100\n- Distance unit: Meter\n\nResults in pose distances: [100 m, 200 m, 300 m, 400 m, 500 m, 600 m, 700 m, 800 m]\n\nFurthermore, the user can choose to either use consecutive pose pairs (non-overlapping) or all posible pairs (overlapping).\n\n### Report Settings\n\n#### Visualization Settings\n\n- `scatter_max_std` (float): The upper colorbar limit is set to the mean plus this value times the standard deviation of the data. This is useful to prevent outliers from dominating the colorbar. Default value is 4.0.\n- `ate_unit_is_mm` (bool): Indicates whether the unit of Absolute Trajectory Error (ATE) is millimeters. Default value is False.\n- `directed_ate` (bool): Indicates whether the ATE is split into along-, horizontal-cross- and vertical-cross-track direction. Default value is True.\n- `histogram_opacity` (float): The opacity of the histogram bars. Default value is 0.7.\n- `histogram_bargap` (float): The gap between histogram bars. Default value is 0.1.\n- `histogram_barmode` (str): The mode of displaying histogram bars. Default value is \"overlay\".\n- `histogram_yaxis_title` (str): The title of the y-axis in the histogram. Default value is \"Count\".\n- `plot_mode` (str): The mode of displaying plots. Default value is \"lines+markers\".\n- `scatter_mode` (str): The mode of displaying scatter plots. Default value is \"markers\".\n- `scatter_colorscale` (str): The colorscale for scatter plots. Default value is \"RdYlBu_r\".\n- `scatter_axis_order` (str): The order of the axes in scatter plots. Default value is \"xy\". If 3d plotting is desired, also specify \"z\".\n- `scatter_marker_size` (int): The size of markers in scatter plots. Default value is 5.\n- `scatter_detailed` (bool): Indicates whether to show scatter plots for each degree of freedom. Default value is False.\n\n#### Mapbox Settings\n\nThese settings currently only apply to trajectory only plots without deviations.\n\n- `scatter_mapbox` (bool): Indicates whether the trajectory should be plotted onto a map. Default value is False.\n- `scatter_mapbox_style` (str): The style of the map. For some styles, a Mapbox token is required. Default value is \"open-street-map\".\n- `scatter_mapbox_zoom` (int): The zoom level of the map. Default value is 15.\n- `scatter_mapbox_token` (str): The Mapbox token. Default value is \"\".\n\nThe mapbox token can be obtained from [https://www.mapbox.com/](https://www.mapbox.com/) after creating a free account.\n\n#### Position Units and Names\n\n- `pos_x_name` (string): Name for the X-axis position. Default: \"x\".\n- `pos_y_name` (string): Name for the Y-axis position. Default: \"y\".\n- `pos_z_name` (string): Name for the Z-axis position. Default: \"z\".\n- `pos_x_unit` (string): Unit for the X-axis position, Default: \"m\".\n- `pos_y_unit` (string): Unit for the Y-axis position, Default: \"m\".\n- `pos_z_unit` (string): Unit for the Z-axis position, Default: \"m\".\n- `pos_dir_dev_x_name` (string): Name for the directed position deviation in along-track direction. Default: \"along\".\n- `pos_dir_dev_y_name` (string): Name for the directed position deviation in horizontal cross-track direction. Default: \"cross-h\".\n- `pos_dir_dev_z_name` (string): Name for the directed position deviation in vertical cross-track direction. Default: \"cross-v\".\n\n#### Rotation Units and Names\n\n- `rot_x_name` (string): Name for the roll rotation. Default: \"roll\".\n- `rot_y_name` (string): Name for the pitch rotation. Default: \"pitch\".\n- `rot_z_name` (string): Name for the yaw rotation. Default: \"yaw\".\n- `rot_unit` (string): Unit symbol for rotation. Default: \"\u00b0\".\n\n#### Export Settings\n\n- `single_plot_export` (ExportSettings): The export settings for single plots. Default value is an instance of ExportSettings with width=800 and height=450.\n- `two_subplots_export` (ExportSettings): The export settings for two subplots. Default value is an instance of ExportSettings with width=800 and height=540.\n- `three_subplots_export` (ExportSettings): The export settings for three subplots. Default value is an instance of ExportSettings with width=800 and height=750.\n- `single_plot_height` (int): The height of a single plot. Default value is 450.\n- `two_subplots_height` (int): The height of two subplots. Default value is 540.\n- `three_subplots_height` (int): The height of three subplots. Default value is 750.\n\n\n#### Export Settings\n\n- `format` (string): The export format. Choices: \"png\", \"svg\", \"jpeg\", \"webp\". Default: \"png\".\n- `height` (integer): The export height in pixels. Default: 500.\n- `width` (integer): The export width in pixels. Default: 800.\n- `scale` (integer): The export scale. Default: 6.",
"bugtrack_url": null,
"license": "MIT",
"summary": "Trajectory Evaluation in Python",
"version": "3.0.0",
"project_urls": {
"Homepage": "https://github.com/gereon-t/trajectopy-core",
"Repository": "https://github.com/gereon-t/trajectopy-core.git"
},
"split_keywords": [
"trajectory",
" evaluation",
" alignment",
" similarity",
" leverarm",
" epsg",
" robotics"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "b97d6f611e0c93418e000b1527d60de6b1e4dc3c407516eb1473fc7b3e881720",
"md5": "a6bcfc251593be3a8c31c59d71f354cc",
"sha256": "48046568807a1ddfe61520d23715a516a054201221a195a359a4daf422d37ab9"
},
"downloads": -1,
"filename": "trajectopy_core-3.0.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "a6bcfc251593be3a8c31c59d71f354cc",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.13,>=3.9",
"size": 133067,
"upload_time": "2024-04-23T15:13:27",
"upload_time_iso_8601": "2024-04-23T15:13:27.390493Z",
"url": "https://files.pythonhosted.org/packages/b9/7d/6f611e0c93418e000b1527d60de6b1e4dc3c407516eb1473fc7b3e881720/trajectopy_core-3.0.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "93b014478065202efa77e7317c74d891a38941dd86d1c03d1d2132a0d9e279f9",
"md5": "e91cef065b8cc3f2b0dba11e1f375909",
"sha256": "41a4f0b817d91930d5e424540c903892ed2455886c277687aa36d7d4af0b228f"
},
"downloads": -1,
"filename": "trajectopy_core-3.0.0.tar.gz",
"has_sig": false,
"md5_digest": "e91cef065b8cc3f2b0dba11e1f375909",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.13,>=3.9",
"size": 111669,
"upload_time": "2024-04-23T15:13:29",
"upload_time_iso_8601": "2024-04-23T15:13:29.749224Z",
"url": "https://files.pythonhosted.org/packages/93/b0/14478065202efa77e7317c74d891a38941dd86d1c03d1d2132a0d9e279f9/trajectopy_core-3.0.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-04-23 15:13:29",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "gereon-t",
"github_project": "trajectopy-core",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"tox": true,
"lcname": "trajectopy-core"
}
JSON
