Evo-V3-Unity

Made by: Shahar Yair

A genetic algorithm built on the unity platform to allow simulating evolution of organisms (cells)

This repo is a separate continuance of: Playing-With-Spheres

3rd Version. Previous versions written in Python can be found at: V2+V1

To run the alorithm, clone the repository and run as a unity project.

Requirements:

• Unity version 2021.3.0f1
• 8 GB RAM

Interactive Simulation

While the simulation is running, you can travel and click on cells to get their current real-time data!

Travel using (Right-Click + WASD).

Speed-up by pressing (Shift).

Get data by clicking on the cells (Left-Click).

Output:

You may stop the simulation at any time.

The simulation streams the data in real-time to a csv file in its folder with the starting timestamp.

Example output of the simulation data converted to graph:

Simulation Concepts:

Cell Parameters:

• Speed: the speed in which the cell will move. Defined by: Globals.SpeedUnits * cell.SpeedPoints.
• SightRadius: the sight radius in which the cell can see. Defined by: Globals.SightRadiusUnits * cell.SightRadiusPoints.
• MaxFullness: maximum fullness the cell can reach. Defined by: Globals.MaxFullnessUnits * cell.MaxFullnessPoints.
• Size: the size of the cell. Defined by: Globals.SizeUnits * cell.SizePoints.
• FoodWorth: worth of each food unit consumed. Defined by: Globals.FoodWorthUnits * cell.FoodWorthPoints.
• ReplicationRate: the rate of replication of the cell. In ths version - is constant. Defined by: Globals.ReplicationRateUnits * cell.ReplicationRatePoints.

Cells Rules:

Food and Hunger:

1. A cell will look for food as long as it is hungry.
2. A cell is hungry if eating a single unit of food will make its fullness equal or lower than MaxFullness.
3. A cell can only see in its SightRadius.
4. Cell1 can eat Cell2 if Cell1.Size > Cell2.Size + 2. Eating another cell gains 3 times the regular food worth.

Movement and Speed:

5. A cell will move in a random direction unless it is hungry and there is no food in its sight radius, or if it spots a cell that can eat it.
6. A cell will always move at its Speed.

Replication:

7. Replication: a cell will divide when two conditions are met:
   7.1 The TimeToReplication (starting at ReplicationRate) has reached 0.
   7.2 Fullness > cell.CellMaxFullness / 2.
8. When a cell divides, its Fullness is cut by half and its TimeToReplication is reset to cell.ReplicationRate.
9. Mutation: On each replication, there is a chance of mutation: increasing one parameter's evolution point with another's (increasing one and decreasing the other).