Flow Programming

RobVision Trigger Node

This Node triggers an image capture with the camera and starts the calculation of grasping points in the background. However, it does not move the Robot to the grasping position.

Configuration

1. Select RobVision Device
   Choose the connected RobVision Device.
2. Select Solution
   Select the solution to use.

RobVision Picking Node

This Node moves the Robot to the calculated grasping position, which in this case is 2 cm above the grasping point. The actual grasping point can then be reached by performing a relative movement in the negative Z-direction once the At Item output is activated.

Configuration

• Select RobVision Device: Choose the connected RobVision Device.
• Select Solution: Select the appropriate solution.
• Edit Poses: Click on Edit Pose to adjust:
  • Approach Pose: The approach pose is a safe pose above the grasping area.
  • Approach Trajectory: The approach trajectory defines the path the Robot takes to reach the grasping pose. For instance, if you intend to approach the grasping pose from 100mm directly above, you would define the approach movement as x=0, y=0, z=-100mm relative to the Robot's Base Coordinate System. This means the Robot will move 100mm straight down along the Z-axis of its base coordinate system to reach the grasping pose.
• Batch Mode:
  If Batch Mode is enabled, the Robot will pick all detected objects before it finishes. If it is not enabled, the Robot will pick only one object, regardless of how many objects are detected. Disable Batch Mode if you need to trigger a new scan each time an object is detected, such as when the scene is likely to change during picking.
• Detection Only: Gets the results of the Solution without moving the Robot. Can be used to check whether an object is present or to further process the detected values in the Variables.

Outputs

• Before Item Approach
  This output is triggered after the Robot has moved to the Approach Pose. It can be used to open the gripper.
• At Item
  This output is triggered once the Robot has moved to the item to be picked. Typically, you would close the gripper here.
• Process XXX
  This output is triggered after the Robot has moved back to the Approach Pose. There is a Process XXX output for each object class.
• After completion
  This output is triggered once all detected items have been processed.
• No Detection
  No object was detected
• Error
  This output is triggered when there is an error during the scan started by the RobVision Trigger Node.

Note

• The object closest to the camera will be prioritized.
• If you have a pallet of objects and want to pick them layer by layer, Batch Mode must be enabled.

Variables

You can create RobFlow Variables and write the result of the detection in there. That makes it possible to use the result of the detection in different places in the Flow.

• Length and Width (Float Variable): Result of the length measurement, only works if length measurement is activated in the solution .
• Number of Detections (Integer Variable): Number of objects that got detected .
• Picking Pose (Cartesian Pose): Pose of the current detection**.**

When the Detection Only toggle is set this can be used to to build the picking process independently from the Node. Using the Variable Node, custom changes to the detected pose can be made, to build custom picking patterns.

RobVision Feedback Node

The RobVision Feedback Node can be connected to the output of the RobVision Picking Node. It can be used to tag certain images, so they are easily accessible in the RobVision result tab. A common use case would be to tag all results that lead to error, or lead to no detections

Configuration

• Select RobVision Device: Choose the connected RobVision Device.
• Select Solution: Select the appropriate solution.
• Feedback tag: Enter a tag, you want to find the images under in the Results Tab.