Caveline Detection at the Edge for Autonomous Underwater Cave Exploration and Mapping

Caveline Detection at the Edge for Autonomous Underwater Cave Exploration and Mapping
复制标题

DOI：
10.1109/icmla58977.2023.00210
复制
发表时间：
2023-12
期刊：
2023 International Conference on Machine Learning and Applications (ICMLA)
影响因子：
--
通讯作者：
Mohammadreza Mohammadi;Sheng-En Huang;T. Barua;Ioannis Rekleitis;Md Jahidul Islam;Ramtin Zand
Mohammadreza Mohammadi;Sheng-En Huang;T. Barua;Ioannis Rekleitis;Md Jahidul Islam;Ramtin Zand
中科院分区：
其他
文献类型：
--
作者：
Mohammadreza Mohammadi;Sheng-En Huang;T. Barua;Ioannis Rekleitis;Md Jahidul Islam;Ramtin Zand

文献摘要

相似文献

本文探讨了在边缘平台上部署基于机器学习（ML）的对象检测和分割模型的问题，以实现用于水下洞穴勘探和测绘的自主水下航行器（AUV）的实时洞穴线检测。我们具体研究了三种ML模型，即，U-Net、Vision Transformer（ViT）和YOLOv 8，部署在三个边缘平台上：Raspberry Pi-4、Intel Neural Compute Stick 2（NCS 2）和NVIDIA Jetson Nano。实验结果揭示了模型精度，处理速度和能耗之间的明确权衡。最准确的模型是U-Net，其F1得分为85.53，Intersection Over Union（IoU）值为85.38。同时，部署在Jetson Nano上的YOLOv 8模型分别在高功耗和低功耗模式下实现了最高的推理速度和最低的能耗。本文提供的全面定量分析和比较结果突出了重要的细微差别，可以指导水下机器人上的洞穴探测系统的部署，以确保水下洞穴勘探和测绘任务期间AUV导航的安全和可靠。
This paper explores the problem of deploying machine learning (ML)-based object detection and segmentation models on edge platforms to enable realtime caveline detection for Autonomous Underwater Vehicles (AUVs) used for under-water cave exploration and mapping. We specifically investigate three ML models, i.e., U-Net, Vision Transformer (ViT), and YOLOv8, deployed on three edge platforms: Raspberry Pi-4, Intel Neural Compute Stick 2 (NCS2), and NVIDIA Jetson Nano. The experimental results unveil clear tradeoffs between model accuracy, processing speed, and energy consumption. The most accurate model has shown to be U-Net with an 85.53 F1-score and 85.38 Intersection Over Union (IoU) value. Meanwhile, the highest inference speed and lowest energy consumption are achieved by the YOLOv8 model deployed on Jetson Nano operating in the high-power and low-power modes, respectively. The comprehensive quantitative analyses and comparative results provided in the paper highlight important nuances that can guide the deployment of caveline detection systems on underwater robots for ensuring safe and reliable AUV navigation during underwater cave exploration and mapping missions.