基于Ray的服务于自动驾驶的远程分布式计算系统
A Ray-Based Remote Distributed Computing System for Autonomous Driving
DOI: 10.12677/CSA.2021.115135, PDF,    科研立项经费支持
作者: 邹星宇, 文 军：电子科技大学信息与软件工程学院，四川 成都；陈 波：智能终端四川省重点实验室，四川 宜宾
关键词: 深度学习；物联网；人工智能；自动驾驶；Ray；Deep Learning； Internet of Things； Artificial Intelligence； Autonomous Driving； Ray
摘要: 随着自动驾驶技术的飞速发展，自动驾驶系统需要采集和处理更多的信息，车载的计算平台难以支撑高级别自动驾驶任务的算力需求。本文提出一种使用Ray框架来处理自动驾驶任务的远程分布式计算系统设计方案，克服单一计算平台的算力限制，为更高级别的自动驾驶提供支持。实验表明，Ray框架实现的分布式计算系统可以大幅减少自动驾驶任务的处理时间，提高系统吞吐量，保证自动驾驶任务的实时性。
Abstract: With the high development of autonomous driving technology, autonomous driving systems need to collect and process more information, and the onboard computing platform has been unable to support the computing requirements of high-level autonomous driving tasks. Thus, we propose a remote distributed computing scheme that uses the Ray framework to implement such kinds of tasks to break through the computing power limitation of a single computing platform, and provide the support for higher-level autonomous driving. Experiments show that the distributed computing system implemented by the Ray framework can greatly reduce the processing time of autonomous driving tasks, increase system throughput, and ensure the real-time performance of autonomous driving tasks.
文章引用：邹星宇, 文军, 陈波. 基于Ray的服务于自动驾驶的远程分布式计算系统[J]. 计算机科学与应用, 2021, 11(5): 1334-1340. https://doi.org/10.12677/CSA.2021.115135

参考文献

[1] Geoffrey, H., Yann, L. and Yoshua, B. (2015) Deep Learning. Nature, 521, 436. [Google Scholar] [CrossRef] [PubMed]
[2] Jordan, M.I. and Mitchell, T.M. (2015) Machine Learning: Trends, Per-spectives, and Prospects. Science, 349, 255-260. [Google Scholar] [CrossRef] [PubMed]
[3] SAE International (2016) Taxonomy and Definitions for Terms Re-lated to Driving Automation Systems for On-Road Motor Vehicles. SAE International, Warrendale, PA.
[4] Geiger, A., Lenz, P. and Urtasun, R. (2012) Are We Ready for Autonomous Driving? The KITTI Vision Benchmark Suite. 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, RI, 16-21 June 2012, 3354-3361. [Google Scholar] [CrossRef] 
[5] Bojarski, M., Del Testa, D., Dworakowski, D., et al. (2016) End to End Learning for Self-Driving Cars. http://arxiv.org/pdf/1604.07316.pdf
[6] 张新钰, 高洪波, 赵建辉, 等. 基于深度学习的自动驾驶技术综述[J]. 清华大学学报(自然科学版), 2018, 58(4): 438-444.
[7] Zaharia, M., Chowdhury, M., Das, T., Dave, A., Ma, J., McCauley, M., Franklin, M.J., Shenker, S., and Stoica, I. (2012) Resilient Distributed Datasets: A Fault-Tolerant Ab-straction for In-Memory Cluster Computing. Proceedings of the 9th USENIX Conference on Networked Systems Design and Implementation (2012), USENIX Association, 15-28.
[8] Gabriel, E., Fagg, G.E., Bosilca, G., Angskun, T., Dongarra, J.J., Squyres, J.M., Sahay, V., Kambadur, P., Barrett, B., Lumsdaine, A., Castain, R.H., Daniel, D.J., Graham, R.L. and Woodall, T.S. (2004) Open MPI: Goals, Concept, and Design of a Next Generation MPI Implementation. Lec-ture Notes in Computer Science, 3241, 97-104. [Google Scholar] [CrossRef] 
[9] Dean, J. and Ghemawat, S. (2008) MapReduce: Simplified Data Processing on Large Clusters. Communications of the ACM, 51, 107-113. [Google Scholar] [CrossRef] 
[10] Moritz, P., Nishihara, R., Wang, S., et al. (2017) Ray: A Distrib-uted Framework for Emerging AI Applications. Proceedings of the 13th USENIX Symposium on Operating Systems De-sign and Implementation (OSDI’18), 8-10 October 2018, Carlsbad, CA, 561-577.
[11] Chang, K. (2015) Wireless Communications for Vehicular Safety. IEEE Wireless Communications, 22, 6-7. [Google Scholar] [CrossRef] 
[12] Xiong, K., Leng, S., Chen, X., Huang, C., Yuen, C. and Guan, Y.L. (2020) Communication and Computing Resource Optimization for Connected Autonomous Driving. IEEE Trans-actions on Vehicular Technology, 69, 12652-12663. [Google Scholar] [CrossRef] 
[13] Abadi, M., Barham, P., Chen, J., et al. (2016) TensorFlow: A System for Large-Scale Machine Learning. Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI), Savannah, 2-4 November 2016, 265-283.