Overview
This project explores the emergent behavior of Physarum polycephalum, a slime mold known for its ability to find efficient paths between food sources without a central nervous system. Using agent-based modeling, this simulation recreates the mesmerizing, organic patterns formed by these organisms.
The simulation consists of thousands of independent "agents" that follow a simple set of rules. As they move, they leave behind a trail of pheromones, which diffuse and decay over time. Agents sense these trails and rotate toward areas of higher density, resulting in the complex, vein-like structures characteristic of slime molds.
Key Features
- Agent-Based Logic: High-performance particle movement.
- Pheromone Dynamics: Realistic trail diffusion and evaporation.
- Configurable Parameters: Real-time adjustment of sensor angles, decay rates, and agent speeds.
- Emergent Complexity: Watch simple rules evolve into complex biological networks.
How it Works
The algorithm follows three primary steps every frame:
- Sense: Each agent samples the trail map at three locations (front, left, and right).
- Move & Deposit: Agents rotate toward the strongest signal and move forward, depositing a new trail at their current position.
- Process Environment: The global trail map is blurred (diffusion) and slightly faded (decay).
function updateAgent(agent, trailMap) {
const sensorSignal = agent.sense(trailMap);
if (sensorSignal.left > sensorSignal.right) {
agent.angle -= agent.turnSpeed;
} else if (sensorSignal.right > sensorSignal.left) {
agent.angle += agent.turnSpeed;
}
agent.x += Math.cos(agent.angle) * agent.speed;
agent.y += Math.sin(agent.angle) * agent.speed;
}