Assignment Week #8 – Predator Prey Simulation

Predator-Prey simulation

For this assignment, I wanted to recreate an underwater chase scene. My goal was to show simulation through autonomous agents, like a single predatory shark chasing a school of fish. The fish reacts to the shark’s actions by fleeing. Each agent has its own set of rules and behaviors that it follows in their 2D ocean.

Implementation:

• The Shark’s Path: To control the shark’s movement, I used a Perlin noise algorithm. This created a smooth, random path that mimics how unpredictable a real shark’s patrol through the water is.
• Fish Fleeing:  In this state, their behavior changes dramatically. They prioritize moving away from the shark over any other actions they might be taking, such as wandering 
• Boundary Constraints: I coded the fish to wrap around the edges of the canvas so they stay in the canvas
• Shark Movement: Instead of being directly controlled, the shark follows a path made by perlin noise. I’ve tuned this function to keep the shark from moving randomly and give its hunt a bit of randomness. Also, the shark slows down when it gets closer to its target.

Sketch:

Code: 

The following is the function for sharks seeking their target which is calculated by Perlin noise.

class Shark {
  
  seek(target) {
    let desired = p5.Vector.sub(target, this.position);
    let distance = desired.mag();
    desired.normalize();
    if (distance < 100) {
      let m = map(distance, 0, 100, 0, this.maxSpeed);
      desired.mult(m); // Slow down as it gets closer to the target
    } else {
      desired.mult(this.maxSpeed);
    }
    let steer = p5.Vector.sub(desired, this.velocity);
    steer.limit(this.maxForce);
    this.applyForce(steer);
  }
  
}

• The seek method in the Shark class steers the shark towards a target.
• It calculates the direction to the target and if the shark is within 100 pixels, it slows down proportionally to the distance to avoid overshooting.
• The map function adjusts the shark’s speed dynamically, allowing for a gradual stop as it reaches the target.
• This prevents the shark from ‘spazzing’ or jittering when it gets close to where it wants to go.

The following is the function for fish fleeing the shark

class Fish {  
  flee(predator) {
    let desired = p5.Vector.sub(this.position, predator.position);
    let d = desired.mag();
    if (d < 100) { // If the shark is within 100 pixels
      desired.setMag(this.maxSpeed); // Set maximum speed
      let steer = p5.Vector.sub(desired, this.velocity);
      steer.limit(this.maxForce); // Limit the force to be applied
      this.applyForce(steer);
    }
  }
  
}

• The flee function in the Fish class calculates the escape direction from the shark.
• If the shark is within a certain range, the fish accelerates to its top speed away from the shark.
• This quick response mimics a fish’s instinct to escape predators.

Challenges: 

• Precision in Shark Movement: Improving the shark’s movement to stop it from “spazzing” as it gets to its target was a difficult job that required slowing the velocity as it got closer to the target.
• Fish Behavior Realism: Making flocking behavior was hard to implement because you had to find a fine balance between random movement and organized group dynamics. After a shark chase, there were difficulties getting the fish to gather back into a proper formation. I think this aspect still needs some work to make it more realistic.

Future Improvements: 

• User Interaction Features: I want to look into ways for users to connect more directly with the simulation, such as using different inputs to change the path of the shark or the fish.
• Ecosystem Development: I like the idea of adding more species and environmental features, like currents or obstacles, to make the experience more interesting.
• Develop Flocking Behavior: To make the fish’s reaction seem real, the flocking and evasion algorithms have to be tweaked to match how fish move naturally.