NOC: Speculating an Asteroid Mining Rocket Genetic Algorithm
For this week's assignment I am speculating the use of a genetic algorithm for designing the best rocket for asteroid mining.
Scenario
Asteroid mining is expected to become a very lucrative industry in the near future because of the abundance in rare earth minerals and other resources in asteroids, the main obstacle stopping asteroid mining from becoming a popular industry right now is our technology - specifically the ability to scale down the cost of space travel/transportation. Below is a Kurzgesagt video about asteroid mining.
One of the main elements that makes space travel too expensive for the average consumer/company right now is the high ratio of fuel cost against the amount of objects transported to space (SpaceX charges around $1 Million for 200KG of payload, which is already much cheaper than other options). Because of this, the ideal asteroid mining ship design will have very efficient fuel consumption rate while having a very big capacity for transporting asteroids back to Earth. Additionally, the path which the spaceship takes can also be taken into consideration in the algorithm to find the optimal route the ship should take to save the most fuel.
Environment
The environment of such a genetic algorithm will be a simulation of the space between Earth and the asteroid. Gravitational force simulation can be used in the vicinity of Earth and the asteroid to help the algorithm discover the optimal path for the spaceships to take.
The population
The population of the algorithm will be individual space ships which will have a mass determined by the mount of materials used in their design. The mass of the spaceship along with gravitational forces of the planetary bodies will determine the amount of force needed to navigate the spaceship, which will be used to determine the fuel consumption.
Fitness & Evolution The goal of the algorithm is to create the optimal spaceship shape and navigation path which uses the least amount of fuel while being able to transport the highest amount of minerals in its storage. In a simple version of this algorithm (which would still be too complicated for me to try to create), different sized spaceships and travel routes will be the 2 main elements that changes, but in more complicated versions of this algorithm, things like air friction can be simulated as well and allow different shaped spaceships to be designed and judged.
In summary, the elements which evolve will be:
-spaceship size
-navigation path
-spaceship shape (in more complicated version)
It's also possible to set a fixed navigation path and see how ships of different sizes and shapes perform in a preset path (this sketch might resemble the smart rockets example sketch where many rockets simply can't make it to the destination: https://editor.p5js.org/codingtrain/sketches/o5PwECj42 )
Fitness
The ratio of fuel consumption vs amount of storage capacity (less fuel & more storage the better) will be used to determine the fitness of each spaceship. Additionally, in the version of the algorithm where navigation path is part of the simulation, only spaceships which made it to the asteroid and back will be considered for the next generation.
--------------------------------------------------------------
And that's all so far for my vision of an asteroid mining rocket genetic algorithm, don't forget to add me as a shareholder when you've built a successful asteroid mining conglomerate using this algorithm 🥺