Skip to main content

Campaign Designer: Building Experiments

1. Introduction​

Access: Automation -> Campaign Designer

The Campaign Designer is the workspace for creating Experiment Templates. In ARES, you design a single experiment defining the instructions for collecting one data point, and ARES OS automates the repetition of this template to form a Campaign.

2. Creating a New Campaign​

  1. Click the + button in the top right of the Campaigns list.
  2. Click the Edit (pencil icon) next to the default name to rename your campaign (e.g., "Close the Loop").
  3. Click Save.

3. Defining Planning Parameters​

The Parameter Designer tab defines the variables that the Planner will control.

  1. Click Create New Parameter.
  2. Parameter Name: Give it a unique ID (e.g., Temperature).
  3. Data Type: Select the type (e.g., Number).
  4. Metadata: Set the Unit (e.g., Degree Celsius), Minimum, and Maximum values.
  5. Has Initial Value: Enable this if the first experiment requires a specific starting point.
  6. Has Achieved Output: Toggle this if this parameter is a "goal" value that might not match what you actually achieve. For example, a furnace is told to heat to 150 degrees, but only actually achieves a temperature of 148 degrees.

4. Designing the Experiment Template​

The Experiment Template tab is where you build the sequential logic for a single iteration.

  1. Click Add Step to create a logical block.
  2. Inside a step, click Add Command.
  3. Select Device: Choose the hardware (e.g., Tube Furnace).
  4. Select Command: Choose the action (e.g., SetSetpoint).
  5. Mapping to Parameters: * To use a dynamic value from your Planner, toggle the Planned switch.
    • Select the corresponding parameter from the dropdown (e.g., Temperature).
  6. Capturing Output: To send data to the Analyzer, add a command like GetCurrentTemperature and ensure Provides Output is enabled.

5. Closing the Loop: Component Assignment​

Once the script is built, you must link the external services that will drive the automation.

A. Planning Tab​

ARES distinguishes between Planning Services and Planners:

  • Planning Service: A hosted service (e.g., a Python microservice) that can offer one or more planning algorithms.
  • Planner: A specific algorithm or logic type provided by that service (e.g., Gradual or Random).

To assign a planner:

  1. Select Service: Choose your registered Planning Service (e.g., My Planner (1.0.0)).
  2. Select Planner: Choose the specific logic type you wish to use for this campaign (e.g., Gradual Planner).
  3. ARES will automatically display the description and parameters managed by that specific selection.

B. Analyzer Inputs Tab​

  1. Select Analyzer: Choose your registered Analyzer (e.g., My Analyzer).
  2. Map Inputs: For each input required by the analyzer, select the corresponding Experiment Output from your script (e.g., mapping the Analyzer's "temperature" requirement to the Setpoint output of your furnace).

6. Saving and Validation​

  • Sequential Logic: Scripts are executed from top to bottom.
  • Validation: The Save button in the bottom right will persist your changes. If you reference a parameter in your script that hasn't been defined or mapped, the system will alert you.