Getting Started

This page walks through a first configuration: one camera, one object detection, one robot, exported for LumiScan VGR. The window and its panels are described on the User Interface page; what each element contains is described under Configuration Elements.

Before you begin

  • LumiScan VGR Engineering is installed and started (Installation).

  • The camera is connected, reachable over the network and set up according to its manual - the preview needs it.

  • You have the files the configuration refers to at hand, for example the .hdvcal calibration file of the camera and the robot description (.urdf) with the hand-eye calibration.

1. Add a camera

  1. In the graph overview at the bottom of the window, type a unique ID into the id field of the Cameras column - for example cam_main - and click Add. IDs name the element in the exported settings, so choose names you will recognise later.

  2. Click the new element. The editor panel on the right shows its settings, the preview panel on the left shows what the camera sees.

  3. Enter the settings of the camera. If the settings refer to a file, such as the .hdvcal calibration file, upload it with Upload in the configuration panel (or drag it onto the buttons) and select it in the editor.

  4. Check the preview. Switch between the image and the point cloud with the buttons in the top-right corner of the preview; rotate, pan and zoom the point cloud with the mouse. The preview follows every change of a setting that affects the camera, so adjust until image and point cloud look right.

2. Add an object detection and connect it to the camera

  1. Type an ID into the id field of the ObjectDetections column - for example detection_network - and click Add.

  2. Drag from the camera element to the new object detection element. The connection tells LumiScan VGR which camera’s image this detection works on. A camera can feed several detections.

  3. Select the object detection and configure it in the editor: which network and which objects are to be detected, and how. The preview shows the element with the current settings.

3. Add a robot and connect it to the detection

  1. Type an ID into the id field of the Robots column - for example robot_arm - and click Add.

  2. Drag from the object detection element to the robot element: the detection result is used to control this robot.

  3. Select the robot and configure it in the editor - its description, the hand-eye calibration that places it relative to the camera, and the gripper.

4. Save and export

  1. Click Save in the configuration panel. Save keeps the current state in LumiScan VGR Engineering; Reload brings the last saved state back and discards unsaved changes.

  2. Click Download. You receive a .zip file with the configuration in two formats: the format of LumiScan VGR Engineering, to continue working on it later, and the settings for LumiScan VGR or other compatible software.

  3. Install the settings on the LumiScan VGR system as described in the LumiScan VGR manual.

Continuing later

Upload the .zip file you downloaded (Upload, or drag and drop onto the configuration panel): the configuration is restored and can be edited, saved and downloaded again.

Tips

  • Give every element an ID that says what it is (cam_bags, network_bags, robot_arm) - the IDs appear in the graph overview and in the exported settings.

  • Remove an element with the X in its top-right corner (visible when it is selected); remove a connection by double-clicking it.

  • If you prefer to see the settings as a whole, switch the editor to Raw JSON. The form and the JSON view show the same data; a change in one is a change in the other.