Workspaces

A Workspace refers to a defined area or coordinate system in the Robot's environment. This could be a table where the Robot performs tasks such as picking or placing parts. A Workspace is defined by three Positions in space, which are taught to the Robot.

Where to find it

Workspaces are Robot-level configuration, so they are not in the View Selector. Open the ⋮ menu in the top-right corner of the app, choose Robot Configuration, and pick Workspaces from the list of tabs:

A Robot with no Workspaces shows only Add Workspace. Because the Workspaces live with the Robot rather than with a Flow, every Flow on that Robot selects from the same list.

What a Workspace entry looks like

Adding one creates an entry in the middle column and opens it on the right:

Part of the entry	What it is for
Name	How the Workspace is identified when you select it in a Reference Group. Unnamed entries fall back to Workspace 1, Workspace 2, …
Select Cartesian Pose	An optional Cartesian Pose Variable that receives the transformation between the Initial and the Calibrated Positions
Initial / Calibrated switch	Switches the three Position fields below between the two sets — this is how you move between defining and calibrating the same Workspace
Position One, Two, Three	The three taught Positions, each with x, y and z in mm, an Apply current Pose button, a hold-to-move button, and Save as Variable

Selecting a Workspace previews it in 3D in the right-hand pane, which is the quickest way to confirm you taught the three Positions in the intended order — a Workspace with its x- and y-points swapped previews visibly rotated.

As with the Home Position fields, the small signed number under each axis is the difference between the stored value and where the Robot is right now, so you can see at a glance whether the Robot is standing at the Position you are about to overwrite.

note

A new Workspace arrives with its Initial Positions already locked — you have to press Unlock for editing before you can teach them. The confirmation spells out the consequence: unlocking will require re-calibration, and the Calibrated Positions have to be updated afterward. That is the protection described under Calibrating a Workspace working as intended, so unlock deliberately rather than out of habit.

Defining a Workspace

To define a Workspace, the Robot is taught three known Positions within the Workspace, such as the corners of a table or key positions within a machine. Those are the Initial Positions. These points allow calculating the Pose of the Workspace relative to the Robot itself.

Calibrating a Workspace

A Workspace may need to be recalibrated if the environment changes, such as when a table is moved, or if the Robot is repositioned. Calibration involves teaching the same three Positions again. Those are the Calibrated Positions. The Robot will calculate the transformation between the Initial Positions and Calibrated Positions. You can optionally save the resulting transformation in a Cartesian Pose Variable and use it for your own purposes inside a Flow.

Using Workspaces in Robot Programming

In order to apply a Workspace to a specific part of your Flow you can make use of the Reference Group. Select one of your Workspaces inside the Reference Group Node. Then place all Nodes inside the Reference Group that should be linked to that Workspace. Once the Workspace is calibrated, all movements within that Workspace are adjusted automatically. Note that currently only Cartesian Movements will be affected by the Workspace-Transformation, while Path Movements and Joint Movements will not be transformed.

Please follow the instructions below to define or calibrate a Workspace:

1. Define the three Initial Positions that describe the Workspace.
   The first Position is the origin of the local coordinate system, the second Position defines the direction of the x-axis, the third Position defines the direction of the y-axis.
2. At a later point you can calibrate the Workspace if it moves or if the Robot's positioning accuracy changes.
   Re-teach the same three Positions. Note that you can also use Cartesian Pose Variables here.