فهرست نشریات علمی دسترسی آزاد ایران

Millimeter-scale capsule robots are promising for minimally invasive biomedical applications, but their performance is often limited by insufficient displacement, low energy efficiency, and structural constraints. This study addresses these challenges by proposing a systematic optimization framework that combines Bayesian Optimization with the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) under Method based on the Removal Effects of Criteria (MEREC) and Entropy weighting schemes. A total of 1000 design configurations were generated and evaluated using a validated nonlinear dynamic model. The optimal configuration achieved a 5.03% increase in displacement, a 1.84% improvement in energy efficiency, and a 3.86% reduction in impact gap compared to the baseline design, resulting in a more compact and efficient capsule structure. Experimental validation confirmed the accuracy of the model and the effectiveness of the optimization strategy. The novelty of this work lies in integrating Bayesian Optimization with multi-criteria decision-making to simultaneously enhance locomotion efficiency, energy usage, and compactness, going beyond previous single-objective optimization approaches. This framework provides a robust pathway for advancing capsule robot design toward practical biomedical applications.