ci(pre-commit): autofix

calibrators/mapping_based_calibrator/README.md

@@ -27,7 +27,6 @@ It is recommended to select the lidar that has the highest resolution and best F
 
 After the mapping has been completed, some processing steps such as data synchronization, data selection and data preprocessing are required before estimating the transformations between lidars. In the explanation below, we assume that there is only a `calibration lidar` that needs to be calibrated with the `mapping lidar`.
 
-
 ##### Data synchronization
 
 As the tool aims to apply registration algorithms on the pointclouds from both the `mapping lidar` and the `calibration lidar` to estimate the transformation between them, it's crucial to ensure those pointclouds share the same timestamp. However, the `calibration lidar` may not be synchronized with the `mapping lidar`, meaning these pointclouds cannot be directly used together during movement. To address this, the tool first needs to interpolate the pose of the `mapping lidar` at the timestamp of the `calibration lidar` in the map.
@@ -39,7 +38,6 @@ In the data selection process, the tool initially pairs `mapping lidar` keyframe
 - The pairs have low time difference and low interpolation error (such as speed, estimated acceleration, etc.) between the keyframe and the `calibration lidar` frame.
 - The pointcloud in `calibration lidar` frame has enough features for calibration, which the tool filters out the pointcloud that has low variance in the z-axis that is mostly a plane.
 
-
 ##### Data preprocessing
 
 For the `calibration lidar` pointcloud to have more points to apply the registration algorithm, the calibrator would need a high-resolution pointcloud from the `mapping lidar`. Therefore, we augment the keyframe's pointcloud with all the frames' pointcloud near it. As the resulting pointcloud has redundant information, we downsample the pointcloud to quite a fine-grained resolution, which is called augmented pointcloud.
@@ -66,7 +64,6 @@ To estimate the transformation between the `mapping lidar` and the `base_link`,
 
 The transformation between the lidar and the ground pose is calculated by utilizing the normal vector and a point on the ground plane. To estimate the transformation between the ground pose and the `base_link`, the tool first determines the initial ground-pose-to-base-link using the initial lidar-to-base-link and lidar-to-ground-pose transformations. Then, the tool projects this initial ground-pose-to-base-link transformation onto the xy plane to estimate the transformation between the ground pose and the `base_link`.
 
-
 ## ROS Interfaces
 
 ### Input
@@ -132,7 +129,6 @@ The transformation between the lidar and the ground pose is calculated by utiliz
 | `min_calibration_pointcloud_size`        | `int`                 | `500`          | The minimum size of the pointcloud that is necessary for estimating transformation.                                                                                                            |
 | `mapping_lost_timeout`                   | `double`              | `1.0`          | Sensor's timeout in seconds to consider the mapping process is failed.                                                                                                                         |
 
-
 ### Mapping Parameters
 
 | Name                                 | Type     | Default Value | Description                                                                                                                        |
@@ -238,7 +234,6 @@ The transformation between the lidar and the ground pose is calculated by utiliz
 - To build the map accurately, drive your vehicle at the lowest feasible speed, such as 2 km/h. Driving too fast can distort the point cloud, negatively impacting the map's accuracy.
 - The surroundings of the calibration area are crucial for creating an accurate map and estimating transformations. Therefore, it is essential to ensure that the environment is rich in natural landmarks suitable for registration-based mapping and calibration in all directions, as shown in the image below. This richness in natural landmarks helps the lidar capture sufficient details beyond simple features like lane surfaces or walls, thereby enhancing the accuracy and reliability of the mapping and calibration processes.
 
-
 <p align="center">
     <img src="../docs/images/mapping_based_calibrator/mapping_based_vis.svg" alt="radar_reflector" width="900">
 <p align="center">