본문 바로가기
  • Home

Performance and Evolutionary Characteristics of Genetic Algorithm-based Optimal Pathfinding for Game NPCs in Unity Engine

  • Journal of Internet of Things and Convergence
  • Abbr : JKIOTS
  • 2026, 12(3), 7
  • Publisher : The Korea Internet of Things Society
  • Research Area : Engineering > Computer Science > Internet Information Processing
  • Received : April 9, 2026
  • Accepted : June 22, 2026
  • Published : June 30, 2026

MyounJae Lee 1

1백석대학교

Accredited

ABSTRACT

This paper analyzes the optimal path convergence performance and evolutionary characteristics of genetic algorithms to maximize the pathfinding efficiency of game NPCs. To this end, an evolutionary computation simulation spanning a total of 90 generations was conducted within the Unity engine environment, utilizing a fitness function designed to reflect the distance to the target, survival time, and successful arrival. The analysis results confirmed that as generations progressed, both the minimum and average distances exhibited downward stabilization, while the number of successful individuals continuously increased, ultimately converging to a stable optimal solution. This study demonstrates the excellent autonomous learning capabilities of genetic algorithms in complex environments and presents the potential for future expansion into adaptive parameter control models.

Citation status

* References for papers published after 2024 are currently being built.