This study addresses the issue of suboptimal intelligent effects in the field of environmental art design at vocational colleges. Under the concept of virtual collaboration, it proposes an environmental art design system platform that achieves faster detection efficiency and higher accuracy in object collision detection. A hybrid collision detection technique combining bounding boxes with the fruit fly optimization algorithm is employed. Through iterative fruit fly optimization algorithm calculations, the minimum feature distance is selected. A hybrid collision detection algorithm combining bounding boxes and TCO-FOA is then used to conduct collision detection experiments in virtual scenes. Finally, 62 students from Class 2 of the Environmental Art Design program at a vocational high school in Zhejiang Province were selected as the research subjects. The students’ grades followed a normal distribution, with an average score of (64.57 ± 1.24). This method effectively enhances students’ interest, attitude, and academic performance in environmental art design courses.
