detected_object_validation#
Purpose#
The purpose of this package is to eliminate obvious false positives of DetectedObjects.
References/External links#
- Obstacle pointcloud based validator
- Occupancy grid based validator
- Object lanelet filter
- Object position filter
Node Parameters#
object_lanelet_filter#
Name | Type | Description | Default | Range |
---|---|---|---|---|
filter_target_label.UNKNOWN | boolean | If true, unknown objects are filtered. | 1 | N/A |
filter_target_label.CAR | boolean | If true, car objects are filtered. | 0 | N/A |
filter_target_label.TRUCK | boolean | If true, truck objects are filtered. | 0 | N/A |
filter_target_label.BUS | boolean | If true, bus objects are filtered. | 0 | N/A |
filter_target_label.TRAILER | boolean | If true, trailer objects are filtered. | 0 | N/A |
filter_target_label.MOTORCYCLE | boolean | If true, motorcycle objects are filtered. | 0 | N/A |
filter_target_label.BICYCLE | boolean | If true, bicycle objects are filtered. | 0 | N/A |
filter_target_label.PEDESTRIAN | boolean | If true, pedestrian objects are filtered. | 0 | N/A |
polygon_overlap_filter.enabled | boolean | If true, objects that are not in the lanelet polygon are filtered. | 1 | N/A |
lanelet_direction_filter.enabled | boolean | If true, objects that are not in the same direction as the lanelet are filtered. | 0 | N/A |
lanelet_direction_filter.velocity_yaw_threshold | float | If the yaw difference between the object and the lanelet is greater than this value, the object is filtered. | 0.785398 | N/A |
lanelet_direction_filter.object_speed_threshold | float | If the object speed is greater than this value, the object is filtered. | 3 | N/A |
object_position_filter#
Name | Type | Description | Default | Range |
---|---|---|---|---|
filter_target_label.UNKNOWN | boolean | Filter for UNKNOWN label | 1 | N/A |
filter_target_label.CAR | boolean | Filter for CAR label | 0 | N/A |
filter_target_label.TRUCK | boolean | Filter for TRUCK label | 0 | N/A |
filter_target_label.BUS | boolean | Filter for BUS label | 0 | N/A |
filter_target_label.TRAILER | boolean | Filter for TRAILER label | 0 | N/A |
filter_target_label.MOTORCYCLE | boolean | Filter for MOTORCYCLE label | 0 | N/A |
filter_target_label.BICYCLE | boolean | Filter for BICYCLE label | 0 | N/A |
filter_target_label.PEDESTRIAN | boolean | Filter for PEDESTRIAN label | 0 | N/A |
upper_bound_x | float | Upper bound for X coordinate | 100 | N/A |
lower_bound_x | float | Lower bound for X coordinate | 0 | N/A |
upper_bound_y | float | Upper bound for Y coordinate | 10 | N/A |
lower_bound_y | float | Lower bound for Y coordinate | -10 | N/A |
obstacle_pointcloud_based_validator#
Name | Type | Description | Default | Range |
---|---|---|---|---|
min_points_num | array | The minimum number of obstacle point clouds in DetectedObjects | [10, 10, 10, 10, 10, 10, 10, 10] | N/A |
max_points_num | array | The max number of obstacle point clouds in DetectedObjects | [10, 10, 10, 10, 10, 10, 10, 10] | N/A |
min_points_and_distance_ratio | array | Threshold value of the number of point clouds per object when the distance from baselink is 1m, because the number of point clouds varies with the distance from baselink. | [800, 800, 800, 800, 800, 800, 800, 800] | N/A |
using_2d_validator | boolean | The xy-plane projected (2D) obstacle point clouds will be used for validation | False | N/A |
enable_debugger | boolean | Whether to create debug topics or not? | False | N/A |
occupancy_grid_based_validator#
Name | Type | Description | Default | Range |
---|---|---|---|---|
mean_threshold | float | The percentage threshold of allowed non-freespace. | 0.6 | N/A |
enable_debug | boolean | Whether to display debug images or not? | 0 | N/A |