The Establishment and Evaluation Model of the Thematic Deep-Learning Teaching Module

In recent years, the application of artificial intelligence (AI) in industry has matured, requiring deeper learning and integration of existing technologies. This study started with technical education to improve the professional quality of human resources. The double-triangular fuzzy number and gray area testing methods in the fuzzy Delphi method (FDM) were used to evaluate expert consensus, plan technical capability indicators, and ensure the integrity and appropriateness of teaching materials. Based on these indicators, special subject teaching course units were designed and integrated into existing courses for experimental teaching and evaluation. The teaching module arrangement in this research used a virtual instrument control system with LabVIEW v2021 as the GUI and the myRIO controller. The proposed system integrates an artificial neural network (ANN) AI model built with Python v3.7 for data analysis and prediction, forming an embedded teaching module for a deep learning-oriented intelligent robotic environmental monitoring system. This study evaluated students’ acceptance of deep learning robotics teaching modules and their impact on improving their technical skills. The psychomotor scale established by the scholars was adopted and revised, including this study’s technical ability indicators. The test-retest reliability of the psychomotor scale was high. The results revealed that the post-test scores of the psychomotor scale were significantly better than those of the pre-test, indicating that students’ overall technical abilities improved. Students’ affective attitudes toward the four dimensions of teaching material and equipment, cognitive development, skills performance, and self-exploration were positive. Feedback revealed that students who participated in the teaching experiment responded positively on all levels of the affective scale, indicating increased motivation and willingness to continue learning. This study successfully constructed a teaching module and evaluation model for deep learning robotic environmental sensing and control. The teaching module and evaluation model established through this research contribute to the cultivation and effectiveness evaluation of relevant technical talents.