GeneticSharp is a fast, extensible, multi-platform and multithreading C# Genetic Algorithm library that simplifies the development of applications using Genetic Algorithms (GAs).
Ask a question
Currently, I'm trying to implement a minimal Ai. The objective is to navigate on a 2D grid 1 time at a block to a destination. My problem is that since all the chromosome values are changing there is a way to set some gene values manually? (I need this because as I said the chromosome is can only move 1 block at a time. And I give the position as a chromosome value) I tried to use the ReplaceGene method but it caused a runtime error:/.
The First Value is the possible movement direction. (Left, Up, Right, Down) each represented as an int. The second and third numbers are the destination coordinates (X and Y coordinate). Clearly, the Ai doesn't have the necessary information. As I said I'm trying to give the ai more information about itself by storing its position in the chromosome as a value. (Those are missing in the image) But it raises the problem that those values cannot be changed at mutation and can only be changed by manually I think I'm on the wrong path and I need to use a totally different approach but I cannot find one.
This is my fitness function I convert the point distances to a negative value so the algorithm tries to find the smallest negative value and thus the best route.
Ask a question Currently, I'm trying to implement a minimal Ai. The objective is to navigate on a 2D grid 1 time at a block to a destination. My problem is that since all the chromosome values are changing there is a way to set some gene values manually? (I need this because as I said the chromosome is can only move 1 block at a time. And I give the position as a chromosome value) I tried to use the ReplaceGene method but it caused a runtime error:/.![error](https://user-images.githubusercontent.com/30203762/102698700-0a21d500-4248-11eb-9367-86f9f313ff38.png)