Sample Tube Detection Skill
Medical Sample Tube Handling
Last updated
Medical Sample Tube Handling
Last updated
This skill enables the detection and sorting of sample tubes standing in a rack according to size and color. It uses a two-finger parallel gripper so that it can grip the sample tubes properly. This information needs to be stored in the skill, because the collision analysis with the corresponding tubes is examined. This ensures that no damage occurs to other tubes when samples are picked up. The optimal gripping candidate is displayed with a blue axis in the image - further candidates are drawn in yellow on the tube and detected collisions are marked with a red line. The selected gripping axis of the AI Skill is displayed next to the detected cap color (marked with a circle), as well as the position and orientation are displayed as usual in the Web GUI. This information can be used during process development to sort the tubes by color and/or cap diameter. A very clear application example can be found on the page.
Position
Position of the grasp point relative to the robot’s coordinate system (in m)
Orientation
Rotation of the grasp point relative to the robot’s coordinate system (in m)
Width
Determined grip width (in m)
Color Information
(Red, Pink, Blue, Yellow, Purple, Orange, Green)
Indicates the number of detected samples by color.
Minimum instance quality
min_score
float
Minimum allowed quality score for detected masks, allows to filter out detections with poor confidence. The impact of this parameter can differ between AI models.
Finger dimensions along grasp axis
finger_on_axis
float
The finger dimension along the grasp axis. This value should be adapted to the geometry of the gripper.
Finger dimensions perpendicular to grasp axis
finger_off_axis
float
The finger dimension perpendicular the grasp axis. This value should be adapted to the geometry of the gripper.
Selection of the class ID for detections
det_cls
int
Used to select a specific class ID for detections, detections with other class IDs are rejected. Valid IDs differ between models. The default value of -1 enables all classes.
Z offset
grasp_depth
float
Offset for the z-component of the pick position
Grasp safety
grasp_safety
float
An offset for the grasp to avoid collisions between the gripper and the detected item.
depth_safety
float
The minimum distance between gripper fingers and items below. This parameter is used to improve the robustness of collision detection.
Optimize gripper orientation
orientation_optimization
boolean
Enables the optimization of the gripper orientation. If enabled the pick pose might be rotated by 180° around the approach axis to minimize the robot rotation between the capture and pick pose.
Enforce GPU usage
force_gpu
boolean
Enforces GPU usage for this skill (could disable GPU usage for other skills or cause performance problems.)
pose (Transformation)
Pick pose for the gripper
name (str)
Color name for the detected test tube
width (float)
Grasp width in meters
optimization_angle (float)
Only present if the parameter optimization_orientation is set to true. The parameter represents the rotation angle in radians that was applied to the original pick pose during the orientation-optimization step.
Conditions
Camera mount:
Dynamic
Static
Supported colors:
Blue
Orange
Green
Pink
Yellow
Purple
Red
colors can be adjusted
Specs
Avg. recognition time:
< 2 seconds
Supported grippers:
Parallel
Features
Measures
Collision detection
Multiple instance recognition
Color detection
