
http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
In the huge video game industry with billions of users, the cost of game testing is substantial enough to reach half, so automation studies are being conducted to reduce testing costs. However, in existing studies of test automation, manual work requiring human intervention such as writing scripts remains, resulting in high testing costs. Alternatively, it is implemented as a virtual environment of VGDL and GVG-AI, making it difficult to use for actual game testing. In this paper, we propose a tool to automate game testing for the purpose of detecting system defects using Unity and Unity ML-Agents for Unity platformer genre games. The proposed tool is based on an actual commercial game engine and analyzes the game itself without human intervention to build an automatic game testing environment. In addition, through examples of automating game testing for actual open source games, it is shown that the proposed tool can reduce test costs and improve quality and stability by performing automatic game analysis and testing environment construction.