Parameter Adjustment in the Logistics Scenario
In the logistics (semantic segmentation) scenario, the Logistics_Seg_RGBSuction model should be used. This model can be used to obtain all pixels of the object in the image, and therefore obtain the contour and status (overlapped or not) of the object.
Parameter Adjustment Order
-
Set a Working Distance first to limit the range for the camera to capture the data.
-
Adjust the Contour Detection (Smart Placement, Special Shape) parameter to determine whether to process objects whose contours are detected with priority.
-
The objects in the logistics scenario are usually stacked randomly. The Overlap Detection can divide objects into overlapped objects and non-overlapped objects.
-
Adjust the Pose Processing parameter group to control the upward offset along the Z-axis to avoid damaging the object.
-
Adjust the Dynamic Place parameter group to adjust the speed of the robot.
-
Adjust the Priority Picking Setting and Pose Sorting Logic parameter group to determine the picking sequence.
Parameter Description
Working Distance
- Minimum Working Distance
-
Description: This parameter is used to specify the minimum distance (in millimeters) between the camera and the object in the scene. If a bin is used, this parameter should be the distance from the camera to the top edge of the bin. Please set the value according to the actual situation.
Default value: 0 mm
Value range: 0–3000 mm
- Maximum Working Distance
-
Description: This parameter is used to specify the maximum distance (in millimeters) between the camera and the object in the scene. If a bin is used, this parameter should be the distance from the camera to the bottom edge of the bin. Please set the value according to the actual situation.
Default value: 3000 mm
Value range: 0–3000 mm
Contour Detection (Smart Placement, Special Shape)
Overlap Detection
- Box Object Rectangularity Threshold
-
Description: This parameter is used to specify the rectangularity threshold for boxes. When the rectangularity of the box is lower than this threshold, the box will be considered to be overlapped, while a rectangularity higher than the threshold will be considered not overlapped.
Default value: 10%
Recommended value: 60%
Value range: 0–100%
- Picking Area Ratio Threshold for Boxes
-
The picking area refers to the region on the object’s surface predicted by deep learning to be suitable for picking. For box-shaped objects, if the ratio of the picking area to the object’s surface area in the camera’s field of view exceeds this threshold, the object is considered not overlapped. Otherwise, the object is considered overlapped.
Default value: 10%
Recommended value: 60%
Value range: 0–100%
- Envelope Object Rectangularity Threshold
-
Description: This parameter is used to specify the rectangularity threshold for envelope-shaped objects. When the rectangularity of the envelope-shaped object is lower than this threshold, the object will be considered to be overlapped, while a rectangularity higher than the threshold will be considered not overlapped.
Default value: 10%
Recommended value: 75%
Value range: 0–100%
- Picking Area Ratio Threshold for Envelopes
-
The picking area refers to the region on the object’s surface predicted by deep learning to be suitable for picking. For envelope-shaped objects, if the ratio of the picking area to the object’s surface area in the camera’s field of view exceeds this threshold, the object is considered not overlapped. Otherwise, the object is considered overlapped.
Default value: 10%
Recommended value: 75%
Value range: 0–100%
- Bag-Shaped Object Rectangularity Threshold
-
Description: This parameter is used to specify the rectangularity threshold for bag-shaped objects. When the rectangularity of the bag-shaped object is lower than this threshold, the object will be considered to be overlapped, while a rectangularity higher than the threshold will be considered not overlapped.
Default value: 10%
Recommended value: 70%
Value range: 0–100%
- Picking Area Ratio Threshold for Bags
-
The picking area refers to the region on the object’s surface predicted by deep learning to be suitable for picking. For bag-shaped objects, if the ratio of the picking area to the object’s surface area in the camera’s field of view exceeds this threshold, the object is considered not overlapped. Otherwise, the object is considered overlapped.
Default value: 10%
Recommended value: 70%
Value range: 0–100%
Pose Processing
- Max Number of Output Poses
-
Description: This parameter is used to specify the maximum number of output poses.
Value range: 0–20
- Upward Offset along Z-Axis
-
Description: This parameter is used to specify the offset distance (in millimeters) along Z-axis above the pick point to prevent the robot from squeezing the object. The unit is millimeters.
Default value: 0
Value range: 0–10 mm
Tuning recommendation: Please set the parameter according to the actual situation.
Point Cloud Downsampling
- Point Cloud Downsampling
-
Description: This parameter is used to reduce the number of object point clouds and increase the processing speed of the Step. The larger the value, the fewer object point clouds, and the faster the Step processing speed.
Default value: 0.0010
Dynamic Place
This parameter group can be used to determine the height of the robot end tool when the robot places the object according to the height of the object, and therefore the following two situations can be avoided.
-
If the robot drops the object at a very high height, the object may roll on the fast-running conveyor belt.
-
If the robot places the object at a very low height, the object may collide with the conveyor belt, and both the object and the conveyor belt may be damaged.
When the robot picks the object, it needs to determine the weight of the object. The algorithm in this parameter estimates the object’s weight according to its volume. If the robot picks a heavy object, and the robot is moving at a fast speed, the following consequences may occur:
-
The object cannot be picked successfully.
-
The picked object is thrown out.
-
The package of the object is torn, but the object is not picked up successfully.
- 2D Area Threshold
-
Description: This parameter is used to detect the area of the detected object. If the area of the detected object is larger than this value, the robot speed will be reduced when placing the object. When the area of the detected object is smaller than this value, the robot will place the object with the Robot Placing Velocity.
Value range: 0–10000000 px
- Robot Placing Velocity
-
Description: This parameter is used to specify the speed when the robot places the object. When the area of the detected object is smaller than the set 2D Area Threshold, the robot will place the object with the speed set in this parameter; when the area of the detected object exceeds the set 2D Area Threshold, the robot reduces the placement speed.
Default value: 100%
Value range: 0–100%
Priority Picking Setting
- Priority Picking Direction
-
Description: This parameter is used to set the direction in which the object will be prioritized for picking.
Value list: Up, Down, Left, Right
Default setting: Right
Pose Sorting Logic
The detected poses in this Step will be sorted according to the pose score in descending order. Pose Score = Object Height × Pose Height Weight + Object Horizontal Position × Object Horizontal Position Weight + Rectangularity × Rectangularity Weight
- Pose Height Weight
-
Description: This parameter is used to set the weight of the pose height when the score of a pose is calculated. If this weight is set to a larger value, objects that are at a high height will be more likely to be picked first.
Default value: 5
Value range: 0–5
- Object Horizontal Position Weight
-
Description: This parameter is used to set the weight of the horizontal position when the score of a pose is calculated. If this weight is set to a larger value, objects that are close to the end in the Priority Picking Direction will be more likely to be picked first.
Default value: 1
Value range: 0–5
Visualization
|
The Visualization parameter group is only used for selecting what will be displayed in the Debug Output window and will not affect the adjustment of the above parameters. |
- Enable
-
Description: Once this option is selected, you can select the items you want to visualize in the Debug Output window.
Default value: selected.
- Visualization Options
-
Description: This parameter is used to specify the items you want to visualize.
Value list: Picking sequence, Pose height, 2D size, Object class, Rectangularity
-
Picking sequence: display the picking sequence as numbers in the central region of the pickable objects.
-
Pose height: display the object pose height in the central region of the pickable object.
-
2D size: display the number of pixels of the object in the image in the central region of the pickable object.
-
Object class: display the class of the object in the central region of the object.
-
Rectangularity: display the object’s rectangularity in the central region of the pickable object. Rectangularity = Size of the object/Size of its minimum bounding rectangle
Default value: Picking sequence
-
|
This parameter will only be displayed after the visualization is enabled. |