Control Panel#

In Aeon, it is crucial to both monitor the state of an experiment as well as control the experiment’s hardware devices and parameters in real time. This is achieved through the construction of a ControlPanel, which constructs visualisers and a control panel for a given Aeon experiment. Full control panels for individual experiments can be found in the Extensions folder (and the toolbox in the Bonsai editor) on the specific experimental branch.

Nodes#

ControlPanel#

In Aeon, the ControlPanel node constructs visualisers and a control panel for an Aeon experiment. These are split into several categories, each implemented in a dedicated GroupWorkflow within the ControlPanel.

ControlPanelWorkflow

PatchActivity: Tracks the state of the foraging patches on the rig, including the current threshold, rolling and total distance travelled on the wheels of the foraging patches, information on the number and timing of food pellet deliveries and time since different subjects visited the different patches.
SubjectActivity: Constructs additional controls for the environment and exposed properties of the workflow, as well as additional visualiser windows. In Aeon, these additional windows are used to display tracking or pose information and monitoring of light levels in the experiment room.
EnvironmentState: Constructs a mashup of a general control panel displaying information and providing control dialogues for the foraging patches, the environment configuration and manual annotation and alert logs grouped into the “ExperimentMetadata” GroupWorkflow, as well as experimental subjects entered and removed from the habitat and subject weight measurements grouped into the “EnvironmentSubjectState” GroupWorkflow
CameraSelector: Displays a single image visualiser used to view any camera or other OpenCV.Net.IplImage, selected from a list of Subjects using the “SubjectActivity” control panel.

Additionally, the ControlPanel node enables manual control to start and stop the cameras on the setup through KeyDown nodes cast to ‘StartCameras’ and ‘StopCameras’ Subjects that trigger or halt acquisition of camera images by the CameraController. Finally, it defines and splits the timestamped entry and exit events of experimental subjects, defined by the SubjectDatabase inside ‘ExperimentMetadata’. These events are then utilised by the RepeatEverySubject workflow to begin and end subscription to the attached sequence on entry and exit of an experimental subject.

Visualisers and control panel are displayed when the workflow is run from the editor or the CLI. As with any other visualisers in Bonsai, the visibility, position, size and layout configuration are stored and loaded from the workflow’s .bonsai.layout file, saved with the workflow.

These control panels are largely supported and configured using in-built functionality provided by nodes in Bonsai.GUI, but some visualisers are constructed using custom auxiliary nodes for Aeon experiments in the Aeon.Environment namespace or as an extension in C#. These include:

and the external workflow:

SubjectDatabase

Usage#

PatchActivity#

The example below shows the “PatchActivity” GroupWorkflow used to construct a control panel for a single foraging patch, camera and RFID reader.

PatchActivityWorkflow

SubjectActivity#

The “SubjectActivity” GroupWorkflow uses an ExperimentProperties node to allow online access to exposed properties during an experiment, and a ButtonSource node (triggered through a shared Subject “ReloadEnvironment”) to generate a labelled button that reloads an environment file. Additionally, the results from pose tracking modules (“SubjectPoses”) and images from a camera monitoring the status of controlled light sources in the room (“CameraLightMonitor”) are subscribed to and organised into TableLayoutPanels, which then create the full visualiser window.

SubjectActivityWorkflow

EnvironmentState#

The “EnvironmentState” control panel or visualiser is created by combining the outputs of the “ExperimentMetadata” and “EnvironmentSubjectState” GroupWorkflows.

ExperimentMetadata#

The “ExperimentMetadata” GroupWorkflow creates a visualiser window that enables control of basic functions for each foraging patch in an experiment, through the control panels created by the correspondingPatchDispenser node. Patch dispenser events are parsed into “Reset” or manual “Deliver” command Subjects which are used by the UndergroundFeeder node to control the behaviour of the foraging patch hardware.

In addition, the AnnotationSource, EnvironmentState and LabelControl nodes create the full “ExperimentMetadata” control panel.

The example below shows the “ExperimentMetadata” GroupWorkflow used to construct a control panel for a single foraging patch.

ExperimentMetadataWorkflow

EnvironmentSubjectState#

CameraSelector#

The “CameraSelector” GroupWorkflow creates a simple image visualiser using a custom StreamSelector node connected to SubscribeSubjects carrying images from each of the cameras and/or other informative images, e.g. a heatmap constructed from tracked subject positions.

The StreamSelector has a single property, “SelectedStream” which enables switching between these images for display. This property is externalised to the highest level of the workflow, meaning it is accessible via the ExperimentProperties node. This enables the user to select and switch to specific cameras or images during acquisition.