robokudo.vis.visualizer_manager

Visualization management for RoboKudo pipelines.

This module provides a behavior tree node that manages visualization of pipeline data using different visualization backends. It coordinates:

• OpenCV-based visualization

• Open3D visualization

• ROS visualization

• Shared visualization state

• Pipeline data synchronization

Classes

VisualizationManager

Behavior tree node for managing pipeline visualizations.

Module Contents

class robokudo.vis.visualizer_manager.VisualizationManager(name: str)

Bases: py_trees.behaviour.Behaviour

Behavior tree node for managing pipeline visualizations.

This class coordinates visualization of pipeline data across multiple visualization backends. It handles:

• Pipeline data buffering and synchronization

• Visualizer creation and lifecycle management

• Shared visualization state

• Visualization timing and performance monitoring

Note

The manager operates in three phases:

1. Pre-tick: Prepare visualizers for new data

2. Tick: Update visualizations

3. Post-tick: Cleanup and synchronization

pipelines: typing_extensions.Dict[str, robokudo.pipeline.Pipeline]

Mapping of pipeline names to Pipeline objects

rk_logger: logging.Logger = None

Logger instance for this class

visualizer_types: typing_extensions.List[typing_extensions.Type]

List of available visualizer classes

visualizers: typing_extensions.Dict[str, typing_extensions.List[robokudo.vis.visualizer.Visualizer]]

Mapping of pipeline name to list of Visualizers

create_visualizers_for_pipeline(pipeline: robokudo.pipeline.Pipeline) → None

Create visualizer instances for a pipeline.

Parameters:

pipeline – Pipeline to create visualizers for

Note

Creates one instance of each visualizer type and associates them with the pipeline. All visualizers share a common visualization state.

static visualizer_instances() → typing_extensions.List[robokudo.vis.visualizer.Visualizer]

Get all active visualizer instances.

Returns:

List of all active Visualizer instances

initialise() → None

Initialize the behavior tree node.

This method:

• Clears existing pipeline references

• Finds all Pipeline nodes in the behavior tree

• Stores references to found pipelines

Note

The VisualizationManager should be placed one level below the top node (Parallel) in the behavior tree.

update() → py_trees.common.Status

Update visualizations for all pipelines.

This method:

1. Copies new data from buffer to visualization map

2. Creates visualizers for new pipelines

3. Notifies visualizers of new data

4. Executes pre-tick, tick, and post-tick phases

5. Checks for termination signals

Returns:

FAILURE if any visualizer indicates termination, RUNNING otherwise

Note

The update cycle ensures synchronized visualization across all backends while maintaining separation between data buffers and visualization state.