This dissertation investigates the effectiveness of the Segment Anything Model (SAM) in enhancing the perception and decision-making capabilities of robotics and autonomous vehicles, specifically addressing the critical issue of how SAM's segmentation accuracy and efficiency influence operational performance. Utilizing a mixed-methods approach, the study presents quantitative data on segmentation performance metrics, revealing a significant increase in accuracy and processing speed compared to conventional models. Additionally, qualitative analyses from real-world implementation scenarios demonstrate improved navigation and obstacle recognition, which are crucial for autonomous system functionality. The findings indicate that SAM not only enhances the technical capabilities of robotics and autonomous vehicles but also holds substantial implications for their deployment in healthcare settings, where precision and swift decision-making are paramount for improving patient outcomes. By facilitating the integration of advanced perception systems in medical transport and robotic-assisted surgeries, this research underscores the transformative potential of SAM in advancing healthcare technologies. Ultimately, the study contributes to the understanding of how innovative segmentation models can drive the evolution of smart systems in healthcare, paving the way for further research and development aimed at optimizing operational efficiency and safety in critical environments.