let populationSize = 100;

let agents = [];

let maxGenerations = 100;

let currentGeneration = 1;

let groundY;

let maxSteps = 1000;

​

// Define agent class

class Agent {

  constructor() {

    this.x = 50;

    this.y = 0;

    this.velY = 0;

    this.gravity = 0.2;

    this.walkForce = 1;

    this.genome = []; // Genome representing walking behavior

    this.fitness = 0; // Fitness score based on walking performance

    this.stepIndex = 0; // Current index in the genome

  }

​

  update() {

    this.applyWalkForce();

    this.applyGravity();

    this.updatePosition();

  }

​

  applyWalkForce() {

    let force = this.genome[this.stepIndex];

    this.velY += force * this.walkForce;

    this.stepIndex++;

    if (this.stepIndex >= this.genome.length) {

      this.stepIndex = 0;

    }

  }

​

  applyGravity() {

    this.velY += this.gravity;

  }

​

  updatePosition() {

    this.y += this.velY;

    if (this.y > groundY) {

      this.y = groundY;

      this.velY = 0;

    }

  }

​

  show() {

    fill(255);

    rect(this.x, this.y, 20, 40);

  }

}

​

// Initialize the population

function initializePopulation() {

  groundY = height - 50;

  for (let i = 0; i < populationSize; i++) {

    agents[i] = new Agent();

    // Initialize the genome with random walking behavior

    for (let j = 0; j < maxSteps; j++) {

      agents[i].genome[j] = random(-1, 1);

    }

  }

}

​

// Evaluate the fitness of each agent in the population

function evaluateFitness() {

  for (let i = 0; i < populationSize; i++) {

    let agent = agents[i];

    // Calculate the fitness based on the agent's walking performance

    agent.fitness = agent.x; // Example fitness calculation (distance traveled)

  }

}

​

// Perform selection, crossover, and mutation to create the next generation

function evolvePopulation() {

  let newAgents = [];

​

  // Selection

  for (let i = 0; i < populationSize; i++) {

    let parentA = selectParent();

    let parentB = selectParent();

    let child = crossover(parentA, parentB);

    mutate(child);

    newAgents[i] = child;

  }

​

  // Replace the existing population with the new offspring

  agents = newAgents;

}

​

// Select a parent based on fitness (roulette wheel selection)

function selectParent() {

  let totalFitness = agents.reduce((sum, agent) => sum + agent.fitness, 0);

  let rand = random(totalFitness);

​

  let partialSum = 0;

  for (let i = 0; i < populationSize; i++) {

    let agent = agents[i];

    partialSum += agent.fitness;

    if (partialSum >= rand) {

      return agent;

    }

  }

​

  return agents[populationSize - 1]; // Fallback option

}

​

// Perform crossover between two parents to create a child

function crossover(parentA, parentB) {

  let child = new Agent();

  let crossoverPoint = floor(random(parentA.genome.length));

​

  for (let i = 0; i < parentA.genome.length; i++) {

    if (i < crossoverPoint) {

      child.genome[i] = parentA.genome[i];

    } else {

      child.genome[i] = parentB.genome[i];

    }

  }

​

  return child;

}

​

// Introduce random mutations in the agent's genome

function mutate(agent) {

  for (let i = 0; i < agent.genome.length; i++) {

    if (random() < mutationRate) {

      agent.genome[i] += random(-mutationMagnitude, mutationMagnitude);

    }

  }

}

​

// Run the simulation

function draw() {

  background(0);

​

  // Update and render each agent in the population

  for (let i = 0; i < populationSize; i++) {

    agents[i].update();

    agents[i].show();

  }

​

  // Check if all agents have finished walking

  let allFinished = agents.every(agent => agent.x >= width - 20);

​

  if (allFinished) {

    // Evaluate the fitness of the population

    evaluateFitness();

​

    // Evolve the population

    evolvePopulation();

​

    // Increment the generation count

    currentGeneration++;

  }

​

  if (currentGeneration > maxGenerations) {

    noLoop();

    console.log("Training finished.");

  }

}

​

// Setup p5.js

function setup() {

  createCanvas(600, 400);

  initializePopulation();

  frameRate(30); // Adjust the frame rate as needed

}

​

// Start the simulation

function startSimulation() {

  loop();

}

​

// Stop the simulation

function stopSimulation() {

  noLoop();

}

​