Iterative Latin Hypercube Planner for ARES

An iterative Latin Hypercube Design of Experiments (DoE) planner built for the ARES OS 2.0 ecosystem. This package provides a service that dynamically populates hypercube bins based on experimental history, ensuring an even distribution of sampling points across the parameter space.

Features

• Iterative Binning: Automatically tracks parameter history to identify and fill empty hypercube bins.
• ARES Integration: Native support for PyAres and the ARES OS planning request/response protocol.
• Repeatable Results: Supports global random seeding for reproducible research.
• CLI Native: Runs instantly from your terminal with configurable port and network settings.

---

Installation

You can install the planner directly via pip. Ensure you have Python 3.10 or newer installed.

pip install ares-planner-iterative-latin-hypercube

Dependencies

• PyAres
• numpy >= 2.2.0

---

Command Line Interface (CLI)

Once installed, the planner service is automatically added to your system's PATH. You can start it from any terminal using the built-in command:

start-latin-hypercube

By default, the service starts on localhost and listens on port 1337.

CLI Arguments

You can customize the service behavior using the following optional arguments:

Argument	Short	Description
--port <int>	-p	Specifies the port number to run the service on. (Default: 1337)
--remote		Runs the service on all network interfaces (0.0.0.0) instead of strictly localhost. Required if ARES OS is on a different machine.
--help	-h	Displays the help menu and exits.

Usage Examples

**Run multiple

Distribution

Distribution A: Approved for public release; distribution unlimited. AFRL-2026-2820