Combine ACO and A* Algorithm in Game AI Pathfinding－As Sample with Tower Defense Game
| Autor: | HSU, CHIA-CHI, 徐嘉琦 |
|---|---|
| Rok vydání: | 2019 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 107 Mobile gaming has continued to soar in popularity, over recent years, with the development of mobile networks and tower defense games are one of the gaming genres that are best at killing time. In traditional tower defense games, once several starting and end points were set, a start point could be randomly selected and various pathfinding methods applied by the player and he/she only needed to establish defenses that were sufficiently strong, such as powerful towers or heroes, along the determined path to complete a stage without much effort, thereby, removing much of the challenge from the games. The purpose of this study is to use the A* pathfinding algorithm to calculate an enemy character’s shortest route and apply the Pheromone Theory of ACO in pathfinding. When a defense tower attacks an enemy, pheromone is produced at the point of attack. The value of this pheromone will impact pathfinding in the subsequent round and act as a mechanism to avoid the player’s strongest defenses. The experimental research utilized in this study is conducted in combination with the flow condition scale and the observational method to assess the game’s challenge and difficulty after the path correction. A total of 50 subjects were included in this study. The results of this study demonstrate that the possibility of integrating A* pathfinding and pheromones is viable and the results of the flow analysis show that while the significance of the challenge is not high, there is a clear increase in difficulty. Therefore, it is possible to force enemy characters to avoid the player’s strongest defenses and change their course of attack by using pheromone corrected A* pathfinding. The mechanisms designed in this study may possibly be applied to behavior trees or state machines to further enhance the diversity of enemy attack methods. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|