<![CDATA[The rapid advancement of drone technology, particularly in its integration with artificial intelligence (AI), has had a significant impact on various sectors, one of which is the field of aircraft inspection and maintenance. This study aims to develop an AI-based script using the ChatGPT O1 and Microsoft Copilot platforms to control the DJI Tello EDU drone in automatically detecting cracks on aircraft bodies. The research was conducted through several stages, including collecting visual data via drone flights, processing and training the model using machine learning, developing the script with AI assistance, simulating on MATLAB Simulink, and finally implementing and testing directly on the physical drone. The results of the study indicate that ChatGPT O1 is capable of generating scripts that are more responsive, comprehensive, and easier to understand compared to Copilot, especially in interpreting natural language prompts. The generated scripts proved effective in both simulations and real-world flight tests, although there were limitations in the drone's sensors that affected the accuracy of altitude and distance measurements. The conclusion of this research is that AI plays a significant role in simplifying drone programming processes and enhancing work efficiency. This study contributes to the development of AI-based autonomous drone technology and opens up opportunities for broader and more efficient applications in other infrastructure inspection fields.]]>
