Modern autonomous underwater vehicles (AUVs) have advanced sensingcapabilities including sonar, cameras, acoustic communication, and diversebio-sensors. Instead of just sensing its environment and storing the data forpost-Mission inspection, an AUV could use the collected information to gain anunderstanding of its environment, and based on this understanding autonomouslyadapt its behavior to enhance the overall effectiveness of its mission. Manysuch tasks are highly computation intensive. This paper presents the results ofa case study that illustrates the effectiveness of an energy-aware, many-corecomputing architecture to perform on-board path planning within abatteryoperated AUV. A previously published path planning algorithm was portedonto the SCC, an experimental 48 core single-chip system developed by Intel.The performance, power, and energy consumption of the application were measuredfor different numbers of cores and other system parameters. This case studyshows that computation intensive tasks can be executed within an AUV thatrelies mainly on battery power. Future plans include the deployment and testingof an SCC system within a Teledyne Webb Research Slocum glider.
展开▼
机译：现代自动水下航行器（AUV）具有先进的传感能力，包括声纳，摄像机，声音通信和各种生物传感器。 AUV不仅可以感知其环境并为任务后的检查存储数据，还可以使用收集到的信息来了解其环境，并基于此理解自动适应其行为，以增强其任务的整体有效性。许多这样的任务是高度计算密集的。本文介绍了一个案例研究的结果，该案例说明了能量感知型多核计算体系结构在用电池操作的AUV中执行车载路径规划的有效性。先前发布的路径规划算法已移植到由英特尔开发的实验性48核单芯片系统SCC上，并针对不同数量的内核和其他系统参数测量了该应用程序的性能，功耗和能耗。此案例研究表明，可以在主要依靠电池电量的AUV内执行计算密集型任务。未来的计划包括在Teledyne Webb Research Slocum滑翔机内部署和测试SCC系统。
展开▼
