Domain: A simple 2D map navigation and resource gathering. More...
#include <harvester.h>
Public Member Functions | |
| bool | evaluatePopulation (darwin::Population *population) const override |
| Assigns fitness values to every genotype. More... | |
| size_t | inputs () const override |
| Number of inputs to a Brain. | |
| size_t | outputs () const override |
| Number of outputs from a Brain. | |
 Public Member Functions inherited from darwin::Domain | |
| virtual unique_ptr< core::PropertySet > | calibrateGenotype ([[maybe_unused]] const Genotype *genotype) const |
| Optional: additional fitness metrics (normally not used in the population evaluation, ie a test set) | |
Detailed Description
Domain: A simple 2D map navigation and resource gathering.
The robot navigates a randomly generated 2D map, represented as a grid where a cell can be one of the following:
- Empty
- Food (Green): Reaching the cell provides a fixed health increase
- Junk Food (Yellow): Provides a minor health increase
- Poison (Red): Fixed health loss
- Wall (Gray): Blocks movement
The robot starts with an initial health value, then at each step it can perform an action. Each action has a health "cost", and once the health drops to zero the experiment is terminated.
Inputs
The robot's vision is constructed from a number of rays (Config::vision_resolution) fanned within a field of view (Config::vision_fov). For each ray, two input values are calculated (distance and color)
| Input | Value |
|---|---|
| 2i | distance |
| 2i+1 | color |
Outputs
| Output | Value |
|---|---|
| 0 | move (magnitude determines the movement distance) |
| 1 | rotate (magnitude determines the rotation angle) |
Member Function Documentation
◆ evaluatePopulation()
|
overridevirtual |
Assigns fitness values to every genotype.
Having a good fitness function is a key part of evolutionary algorithms:
- Perhaps obvious, the fitness value should accurately estimate the quality of a particular solution
- A "smooth" distribution is preferable since it provides a gradient which can guide the incremental search in the solutions space. (ex. if most fitness values are 1.0 or 0.0 it's hard to know which genotypes are good candidates for reproduction)
- Returns
trueif the evolution goal was reached
Implements darwin::Domain.
The documentation for this class was generated from the following files:
- C:/Users/lemo/work/darwin/domains/harvester/harvester.h
- C:/Users/lemo/work/darwin/domains/harvester/harvester.cpp
