Overview of deep learning and mobile edge computing in autonomous driving

: pp. 208 - 218
Lviv Polytechnic National University, Department of artificial intelligence systems

In recent years, mobile edge computing and deep learning have attracted strong industry attention in the application scenario of autonomous driving. Mobile edge computing reduces the transmission delay of autonomous driving information by offloading computational tasks to edge servers to reduce the network load; deep learning can effectively improve the accuracy of obstacle detection, thereby enhancing the stability and safety of autonomous driving. This paper first introduces the basic concept and reference architecture of MEC and the commonly used model algorithms in deep learning, and then summarizes the applications of MEC and deep learning in autonomous driving from three aspects: target detection, path planning, and collision avoidance, and finally discusses and outlooks the problems and challenges in current research.

  1. Li Z. S., Xie R. C,. Sun L., et al. Overview of mobile edge computing. Telecommunication Science, 2018, 34(1): 87–101. DOI: 10.11959/j.issn.1000-0801.2018011. Chinese.
  2. Liu J. W., Liu Y., Luo X. L. Advances in deep learning research. Computer Applications Research., 2014(7): 1921–1930. DOI: 10.3969/j.issn.1001-3695.2014.07.001. Chinese.
  3. Li Z. Research on autonomous driving technology based on laser scanning radar. Beijing: Northern Polytechnic University, 2018. Chinese
  4. Muresan M., Fu L., Pan G. Adaptive traffic signal control with deep reinforcement learning an exploratory investigation. arXiv.org, 2019: 1–15. DOI: 10.48550/arXiv.1901.00960(in English).
  5. Zhou Z. H., Feng J. Deep Forest: towards an alternative to deep neural networks. Proceedings of the 26th International Joint Conference on Artificial Intelligence. 2017: 3553–3559. DOI: 10.48550/arXiv.1702.08835 (in English).
  6. Thanh L., Rose T., Hu Q. Y., et al. Mobility-aware edge caching and computing in vehicle networks: A deep reinforcement learning. IEEE Transactions on Vehicular Technology, 2018(11): 10190–10203. DOI: 10.1109/TVT.2018.2867191(in English).
  7. Ying H., Zhao N., Yin H. X. Integrated networking, caching, and computing for connected vehicles: A deep reinforcement learning approach. IEEE Transactions on Vehicular Technology, 2017(1): 44–55. DOI: 10.1109/TVT.2017.2760281(in English).
  8. Liu S., Li L., Tang J., et al. Creating autonomous vehicle systems. Synthesis Lectures on Computer Science, 2017, 6(1): 1–186. DOI: 10.2200/S00787ED1V01Y201707CSL009 (in English).
  9. Moranduzzo T., Melgani F. Automatic car counting method for unmanned aerial vehicle images. IEEE Trans Geosci Remote Sens, 2014, 52(3): 1635–1647. DOI: 10.1109/TGRS.2013.2253108 (in English).
  10. Ye Y. Y., Xiao L. H., Hou J. C. Lane detection method based on lane structural analysis and CNNs. IET Intelligent Transport Systems, 2018(6): 513–520. DOI: 10.1049/iet-its.2017.0143. (in English).
  11. Bai M., Mattyus G., Homayounfar N., et al. Deep multi-sensor lane detection. 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2018: 3102–3109. DOI: 10.1109/IROS.2018.8594388. (in English).
  12. Gansbekew V., Brabandere B. D., Neven D., Et al. End-to-end Lane Detection through Differentiable Least- Squares Fitting. 2019 IEEE / CVF International Conference on Computer Vision Workshop (ICCVW), 2019: 905–913. DOI: 10.48550/arXiv.1902.00293. (in English).
  13. Li J., Mei X., Prokhorov D., et al. Deep Neural Network for Structural Prediction and Lane Detection in Traffic Scene. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(3): 690–703. DOI: 10.1109/TNNLS.2016.2522428. (in English).
  14. Wang Q, Gao J., Yuan Y. Embedding structured contour and location prior in Siamese fully convolutional networks for road detection. IEEE Trans Intel, 2018: 230–241. DOI: 10.1109/TITS.2017.2749964 (in English).
  15. Tayarah, Kim G. S., Kil T. C. Vehicle detection and counting in high-resolution aerial images using convolutional regression neural network. IEEE Access, 2017 (6): 2220–2230. DOI: 10.1109/ACCESS.2017.2782260 (in English).
  16. Song H. S., Liang H. X., Li H. Y., et al. Vision-based vehicle detection and counting system using deep learning in highway scenes. European Transport Research Review, 2019, 11(1): 1–16. DOI: 10.1186/s12544-019-0390- 4 (in English).
  17. Melotti G., Premebida C., Gonçalves N. Multimodal deep-learning for object recognition combining camera and LIDAR data. In 2020 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC). 2020: 177–182. DOI: 10.1109/ICARSC49921.2020.9096138 (in English).
  18. Wang W., Zhou S., Li J., et al. Temporal pulses driven spiking neural network for fast object recognition in autonomous driving. arXiv preprint arXiv, 2020: 1–7. DOI: 10.48550/arXiv.2001.09220 (in English).
  19. Barba-Guaman L., Jose E. N., Anthony O. Deep learning framework for vehicle and pedestrian detection in rural roads on an embedded gpu. Electronics, 2020(4): 1–17. DOI: 10.3390/electronics9040589 (in English).
  20. Ucar A., Demir Y., Gjzeli C. Object recognition and detection with deep learning for autonomous driving applications. Electronics, 2017, 93(9): 759–769. DOI: 10.1177/0037549717709932 (in English).
  21. Wang S., Cheng J., Liu H., et al. Pedestrian detection via body part semantic and contextual information with DNN. IEEE Transactions on Multimedia, 2018, 130: 5642–5643. DOI: 10.1109/TMM.2018.2829602 (in English)..
  22. Liu S., Liu L., Tang J., et al. Edge computing for autonomous driving: opportunities and challenges. Proceedings of the IEEE, 2019(99): 1–20. DOI: 10.1109/JPROC.2019.2915983 (in English).
  23. Limmer M., Forster J., Baudach D., et al. Robust deep-learning-based road-prediction for augmented reality navigation systems at night. 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC) IEEE, 2016: 1888–1895. DOI: 10.1109/ITSC.2016.7795862 (in English).
  24. Hsin-Te W., Hsin-Hung C., Fan-Hsun T., et al. Optimal route planning system for logistics vehicles based on artificial intelligence. Journal of Internet Technology, 2020, 21(3): 757–764. DOI: 10.3966/160792642020052103013 (in English).
  25. Jin F., Sun S. Neural network multitask learning for traffic flow forecasting // 2008 IEEE International Joint Conference on Neural Networks, IEEE World Congress on Computational Intelligence, 2008: 1897–1901. DOI: 10.1109/IJCNN.2008.4634057 (in English).
  26. Huang W., Song G., Hong H., et al. Deep architecture for traffic flow prediction: deep belief networks with multitask learning. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(5): 2191–2201. DOI: 10.1109/TITS.2014.2311123 (in English).
  27. Long P. X., Liuwx, Pan J. Deep-learned collision avoidance policy for distributed multiagent navigation. IEEE Robotics and Automation Letters, 2017(2): 656–663. DOI: 10.1109/LRA.2017.2651371 (in English).
  28. Changwj, Chen L. B., Su Ky. DeepCrash: a deep learning-based internet of vehicles system for head-on and single-vehicle accident detection with emergency notification. IEEE Access, 2019(7): 148163–148175. DOI: 10.1109/ACCESS.2019.2946468 (in English).
  29. Yao Y., Xu M., Wang Y., et al. Unsupervised traffic accident detection in first-person videos. 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2019: 273–280. DOI: 10.1109/IROS40897.2019.8967556 (in English).
  30. Ren H., Song Y., Wang J., et al. A deep learning approach to the citywide traffic accident risk prediction. Computer and Society, 2018: 3346–3351. DOI: 10.1109/ITSC.2018.8569437(in English).
  31. Chen Q., Song X., Yamada H., et al. Learning deep representation from big and heterogeneous data for traffic  accident  inference.  AAAI  Conference  on  Artificial  Intelligence  Thirtieth  AAAI  Conference  on  Artificial Intelligence, 2016: 1–7 (in English).