Binary Coded Layered Autonoma

An interactive cellular automaton that combines Langton's Ant, multi-color substrates, and Conway's Game of Life into a single layered system. Multiple ants traverse a colored grid, encoding their movement rules in binary, while a selectively-activated life simulation evolves on top of the substrate they create.

Open layered_automata.html in a browser to run the simulation — no build step or dependencies required.

Concept

The system layers three interacting subsystems:

1. Multi-Color Substrate (2–8 colors) — a grid of colored cells that ants modify as they move.
2. Binary-Coded Ants — Langton-style ants whose turn behavior (Left/Right) is encoded as a binary string indexed by the substrate color underneath them.
3. Selective Conway's Life — a Game-of-Life layer that only evolves on cells whose substrate color is enabled by an activation mask, with positive/negative modes that create or inhibit life.

The interplay between these layers produces highway structures, fractal boundaries, and emergent life colonies tuned to specific color patterns.

How It Works

Ant Movement

Each ant reads the substrate color at its current cell and consults its binary Ant Rule:

• Bit 0 → turn Left
• Bit 1 → turn Right

The ant then increments the cell's color (mod numColors), marks the cell, and steps forward. With numColors = 4 and rule 0101, this generalizes Langton's classic LRLR ant.

Life Activation Mask

A second binary string — the Activation Mask — determines which substrate colors are eligible to host life. Each color also has an Activation Mode:

• Positive (+): ant presence spawns life and clears inhibition
• Negative (−): ant presence kills life and creates an inhibition zone that suppresses spawning (decays at ~ 10%/generation)

Conway's Life Layer

Life evolves only on marked cells whose substrate color is mask-enabled and not inhibited. The rules are extended:

• Search radius (1–5) — neighborhood size for counting live neighbors
• Birth — exact neighbor count required to spawn life
• Survival min/max — neighbor range that keeps a live cell alive
• Mutual inhibition — neighbors only count if they share the same activation mode (+ or −) as the cell being evaluated

Negative-mode cells use slightly relaxed survival/birth thresholds, making them more fragile.

Multi-Ant System

Up to 8 ants can run simultaneously with configurable:

• Spawn modes: center, corners, edges, random, grid
• Synchronization:
  • synchronized — all ants share the same rule and mask
  • independent — each ant gets a randomized rule/mask
  • offset — each ant uses a rotated version of the base rule

Controls

Buttons

• Start / Stop — toggle continuous simulation
• Step — advance a single generation
• Reset — clear grids and respawn ants
• Randomize All — randomize rule, mask, and parameters
• Random Rule — randomize only the ant rule
• Random Mask — randomize only the activation mask & modes

Stats Panel

• Generation — total simulation steps
• Ant Steps — combined steps across all ants
• Marked Cells — cells visited at least once by an ant
• Live Cells — currently alive Conway cells
• Inhibited Cells — cells suppressed by negative activation
• Active Ants — number of ants currently running
• Current Mask — the active life-activation bitstring

Visualization

• Substrate is rendered using the configurable color palette (only marked cells are drawn).
• Live cells appear as bright green squares.
• Inhibited cells are tinted red.
• Ants are shown as colored squares with a yellow direction indicator and an ID label (when multiple ants are present).
• Click the canvas to enter fullscreen zoom mode; press Esc or click again to exit.

Tips for Interesting Patterns

• Start with numColors = 4, rule 0101 (classic Langton's ant) to observe the famous ~10,000-step highway emergence.
• Try rule 1100 or 0011 with 4 colors for symmetric expanding patterns.
• Mix positive and negative activation modes to create competing life regions with shifting boundaries.
• Use multiple ants with the offset synchronization to generate kaleidoscopic interference patterns.
• Larger life search radii (3–5) combined with higher birth thresholds produce slow, organic-looking growth.
• Use Randomize All repeatedly to discover novel rule combinations.

Files

• layered_automata.html — self-contained simulation (HTML + CSS + JS)
• README.md — this document

Implementation Notes

• Rendering uses a single ImageData buffer for substrate and life cells (fast bulk pixel writes), with ants drawn via standard canvas calls for glow effects.
• Grid wraps toroidally on all edges.
• Inhibition decays stochastically (10% chance per generation) so negative regions slowly heal.
• Neighbor counting respects activation-mode matching, producing emergent segregation between positive and negative life populations.