Search for: All records

This paper addresses the challenge of deploying machine learning (ML)-based segmentation models on edge platforms to facilitate real-time scene segmentation for Autonomous Underwater Vehicles (AUVs) in underwater cave exploration and mapping scenarios. We focus on three ML models-U-Net, CaveSeg, and YOLOv8n-deployed on four edge platforms: Raspberry Pi-4, Intel Neural Compute Stick 2 (NCS2), Google Edge TPU, and NVIDIA Jetson Nano. Experimental results reveal that mobile models with modern architectures, such as YOLOv8n, and specialized models for semantic segmentation, like U-Net, offer higher accuracy with lower latency. YOLOv8n emerged as the most accurate model, achieving a 72.5 Intersection Over Union (IoU) score. Meanwhile, the U-Net model deployed on the Coral Dev board delivered the highest speed at 79.24 FPS and the lowest energy consumption at 6.23 mJ. The detailed quantitative analyses and comparative results presented in this paper offer critical insights for deploying cave segmentation systems on underwater robots, ensuring safe and reliable AUV navigation during cave exploration and mapping missions. 

Microservice architecture has become the leading design for cloud-native systems. The highly decentralized approach to software development consists of relatively independent services, which provides benefits such as faster deployment cycles, better scalability, and good separation of concerns among services. With this new architecture, one can naturally expect a broad range of advancements and simplifications over legacy systems. However, microservice system design remains challenging, as it is still difficult for engineers to understand the system module boundaries. Thus, understanding and explaining the microservice systems might not be as easy as initially thought. This study aims to classify recently published approaches and techniques to analyze microservice systems. It also looks at the evolutionary perspective of such systems and their analysis. Furthermore, the identified approaches target various challenges and goals, which this study analyzed. Thus, it provides the reader with a roadmap to the discipline, tools, techniques, and open challenges for future work. It provides a guide towards choices when aiming for analyzing cloud-native systems. The results indicate five analytical approaches commonly used in the literature, possibly in combination, towards problems classified into seven categories.