CodingTrain / CodingTrain/Suggestion-Box
[Coding Challenge Suggestion] [Nature of Code] Antz!
- 主要言語
- 言語のデータがありません
- スター
- 570
- フォーク
- 85
- PR マージ指標
- 30日以内にマージされた PR はありません
説明
Here's a Coding Challenge Suggestion/Project:
The goal of "Antz" is to show that from a very simple set of rules, repeated by many agents (here virtual ants!), a complex task can be executed.
For "Antz" the basic rules are:
when not carrying food:
- search randomly (e.g. Gaussian randomness) ; they start from the nest.
- if some food is found then pick it up
- else if a 'strong enough' pheromone trail is found then try to follow it
when carrying food:
- go back to the nest in the shortest possible way
- leave a pheromone trail behind
- if the nest is reached, put the food down.
The pheromone trail fades away as the time goes by. I would make the trail not a single pixel wide line, but instead a few pixels wide with a gradient from the ant actual path. The pheromones from several ants accumulate.
A food source disappears when empty (several sources are generated when starting).
______________________________
In a further/2nd version, some obstacles could be added to show that the ants "choose" the shortest way around an obstacle, due to the fact that the short path has a more dense pheromone trail than the longest which eventually gets discarded. (This principle is studied in the data routing/networking field).
______________________________
My own amateur example can be found here : https://github.com/fphenix/Antz
Cheers!
Fred L.
コントリビューションガイド
このリポジトリのコントリビューションガイドは索引されていません
調査の方向性
リンクされているAntzの例と、記載されている運搬、探索、フェロモンのルールを確認することから始めます。食料源、薄れていく経路、蓄積する経路を含むシミュレーションの範囲を定義し、どの観測可能な動作が完了を示すのかを決めます。障害物の経路探索は明示的に後のバージョンで扱います。
索引モデルが issue の本文から書いたものです。
評価
- 領域
- ai
- issue の種類
- 機能追加
- 難易度
- 5/5
- 見積もり時間
- 1週間以上
- 活発さ
- 停滞
- 明瞭さ
- 説明が足りない
- 初心者へのやさしさ
- 25/100