Simulations are a powerful tool for understanding the world around us as it exists, and also as it could be. The creation of models that reflect a given context by adhering to selected rules that are followed in said context, show how all the constitutive parts behave and evolve over time, and by doing so offer huge insights about the context. This makes simulation an important tool for designers to explore and learn.
With that in mind, the objective of this project is to explore the use of simulations in a more creative way. Instead of trying to replicate an existing environment, a completely imaginary one is created, getting inspired by some existing rules of the physical world and creating some new non-existent rules. The project is also an opportunity to experiment by using multiple smaller simulations and concatenating their outputs, instead of creating an all encompassing simulation.


The project models a small solar system with two planets orbiting around a star. It takes into account the relative position and forces each of the celestial bodies have with each other and how they affect the body itself. For the case of the planets this relation to the other bodies has an effect on its climate/atmosphere. A small group of agents living on the surface of the planet that react to the dynamic climate are also simulated by the project.

Since the simulation covers such different scales instead of trying to solve all of them at once, the whole model is divided into 3 smaller simulations each of which manages a different scale, and that are performed sequentially from the one dealing with the larger scale to the one dealing with the smallest. This smaller simulation are, in the order that are performed the followings:

  • Solar System SIM: The simulation in charge of reproducing the movements of the celestial bodies. This simulation is  a physics simulation that uses mathematical formulas, in particular Newton’s gravity equation, to predict the outcome of the movement.
  • Planet System SIM: The simulation that controls the change in climate for the environment of the planets. The Planet System SIM, cannot be considered a physics simulation since it is not modelling a realistic climate, since the complexity of simulating one is too high. Instead it creates some simple rules that result in a simple atmosphere for the planet.
  • Alien Entity SIM: The simulation that runs the movement of the “Aliens” living on the surface of the planets. This is a Agent simulation that is focused on the movement of agents and that uses the context (in this case the planet and its climate) to constrain the behaviour of them.

Solar System SIM

The first simulation is the Solar System SIM. This is a physical model that follows closely the gravity equation to iteratively update the position of the celestial bodies. The starting inputs that are required to perform the simulation are the celestial bodies, which are an object that contains five important values (Type, Locked, Mass, Position, Velocity); the time step, which is a numeric value that determines the amount of time that passes in between each iteration of the simulation; and finally the gravity constant, which is a numeric value necessary to calculate the attraction between bodies. This simulation returns as an output the celestial bodies, but with a record of their movement and change in speed.

Starting Set Up

The initial setup for the simulation consisted of having one sun of 15 units of mass and no initial velocity, and two planets, one closest to the sun called Alpha, and a second one further away called Beta. Both of the planets have 0.5 units of mass and a tangential velocity to the sun.
For the rest of the inputs were set up like so: time step = 0.25 and gravity constant = 2.5.

Output Visualization
Video showing the movement of the celestial bodies (Sun in yellow, Alpha in light blue, Beta in darker blue) dictated by the forces of attraction.
Video showing the forces of attraction that each celestial body has towards the other bodies (Yellow), and the resulting force of attraction (Red).
Video showing the velocity (Red) and acceleration (Yellow) vectors of each celestial body.
Line plots showing the distance of the planet Alpha tho the other bodies over time (left), and the evolution of the planet’s velocity over time (right).
Line plots showing the distance of the planet Beta tho the other bodies over time (left), and the evolution of the planets velocity over time (right).

Planet System SIM

The next simulation is the Planet System SIM. As mentioned before, this is not truly a physical simulation as it does not adhere to real physics laws, instead a set of rules and logics are created to implement a simple climate system for the planets. This climate system keeps track of two main things, daylight cycle and the surface temperature of the planet. The daylight cycle is produced by evaluating which parts of the planet are facing the sun; this makes the cycle dependent on the speed of the axial rotation of the planet which is faster the further away from the sun the planet is (Given that when closer to the sun the gravitational pull slows the axial rotation down). On the other hand the surface temperature is obtained by dividing the temperature of the sun by the distance to the planet surface, and also by taking into account a thermal inertia that the planet atmosphere would provide. The surface temperature is closely related to the daylight cycle, as both depend on if a location is facing or not the sun. The Planet System SIM outputs the evolution of the surface temperature and the daylight cycle of the planet over time.

Starting Set Up

In the case of the Planet System SIM, some of the initial inputs are acquired from the previous simulation, that is the case of the celestial bodies’ relative positions, and other information. The rest of the initial variables are set up as follows: The atmosphere inertia is set up so that temperature takes half a planet rotation to fully dissipate. Meanwhile the rotation falloff or the speed of the axial rotation is set so that in distance closer or equal to 5 there is no rotation and from that every unit the distance grows the rotation is 1/600 of a revolution faster.

Output Visualization
Evolution over time of the Surface Temperature (left) and Daylight Cycle (Right) showed a mercator projection of the surface of the planet.
Line plot that shows the maximum (Top most dashed line), average (Middle line) and minimum (Bottom most dashed line) values of surface temperature of the planet over time.
Line plots comparing the Surface Temperature and Daylight Cycle over time, of four sample points on the surface of the planet.

Alien Entities SIM

The final simulation is the Alien Entities SIM. This is an Agent simulation that focuses on modelling the movement of agents on the surface of the planets, by setting ground rules that govern their behaviour and way they are allowed to move. In the case of this agent simulation, the rules were constructed to have the agents react to the change in the planet’s climate, and there are three of them. The first rule for the agents is that they must stay on land, the planets might have certain places with bodies of water, and the agents are not allowed to traverse them. The second rule dictates that if an agent is on a location in the planet where it isnt day the agent will sleep, and so it will not move until daylight starts again. The third and last rule has the agents move in response to the temperature change, each of them has a preferred temperature they would like to stay in, therefore they will move across the planet towards places that are closer to said preference.

Starting Set Up

The initial conditions for the Alien Entities SIM, like with the Planet System SIM, are given partly by the output of the previous simulation. In this case, both surface temperature and daylight cycle are the output of Planet System SIM. The other starting variables are set like so: The world seed is set randomly and it is used to create landmasses and waterbodies with the help of perlin noise. The last variable are the Agents of which they are fifty and start with a random initial movement vector, and a random temperature preference.

Output Visualization
Video showing the agent (Red square for awake agents, grey square for asleep ones) movement based on the Daylight Cycle, Surface Temperature and the planet’s Geography; on a mercator projection of the surface of the planet.
Line plots showing each agents “satisfaction” or how far they are from their preferred temperature over time (Dashed lines), and the average of satisfaction over all agents (Orange line).

Final Thoughts

Agents from Outer Space is an example of a possible way to go about building a simulation with the purpose of gaining understanding of an existing or a made up context. The project demonstrates how a simulation of a complex scenario can be the sum of smaller and more narrow/focused models, how a simulation can model the context in various different scales, and how different types of simulations like physics and agent simulations can be used together to more accurately describe a scenario.