GEESE: Grammatical Evolution Algorithm for Evolution of Swarm Behaviors

Aadesh Neupane

Swarms - Bees

Swarms - Ants

Problem

- Design collective behaviors is cumbersome

- Requires experts for modeling

Solution

- Evolutionary computation to generate collective behaviors

$$\frac{dR}{dt} = -aR + vnD , R(0) = R_0$$

$$\frac{dO}{dt} = aR -bO +cE, O(0) = O_0$$

$$\frac{dE}{dt} = q(D)bO - cE, E(0) = E_0$$

$$\frac{dA}{dt} = p(D)bO - mA + wnD$$

$$\frac{dD}{dt} = mA - nD, D(0) = D_0$$

Reference: Stability of choice in the honey bee nest-site selection process

GEESE

Hello! Neighbour!

How are you doing with the phenotype in this environment?

Hello!

I collected 35 oz of water with the phenotype.

Take my genome and perform magic using genetic operators !

Santa Fe Trail

  • Objective
    • Find all food using maximum of 600 moves
  • 32 x 32 cells
  • Optimal trail
    • 144 cells
    • 89 food
    • 55 gaps
  • Actions :
    • Left
    • Right
    • Forward

Results

  • Solve Santa Fe Trail in 324 steps

  • Outperforms all other algorithms. Closest was 331 steps by Novelty GE

Results

Single Source Foraging problem

  • Better than hand-coded and standard GE

  • Collects 83 units of food in 284 time-steps

Thank You!

Made with Slides.com