基於邊緣計算的異常駕駛行為偵測平台之設計及實作

Design and Implementation of An Edge Computing Based Anomaly Detection Platform for Car Driving

廖家興(Jia-Xing Liao)

指導教授 ： 林風

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2021年)

https://doi.org/10.6342/NTU202004475

摘要

OBD-II是一個標準的車輛診斷連接埠，目前被廣泛用於車輛診斷及維護。在這篇論文中，我們設計並實作了一個基於邊緣計算的異常駕駛行為偵測平台。該平台透過OBD-II連接埠，從車輛的引擎電子控制單元(Engine ECU)蒐集OBD參數資料，並使用一個預先訓練好的駕駛行為模型進行異常偵測。該平台能夠被安裝到一個稱為樹莓派(Raspberry Pi)的小型單主機板電腦中，並提供低延遲的異常偵測服務。

關鍵字

OBD-II ； 異常偵測 ； 車輛 ； 駕駛行為

並列摘要

The OBD-II is a standard diagnostic port which has been widely used in the vehicle diagnostic and maintenance. In this thesis, we design and implement an edge-computing based anomaly detection platform (ECADP) for car driving, which collects the value of OBD parameters from the engine ECU of vehicle through OBD-II and detect anomalies with a pre-trained driving behavior model. The platform can be installed on a small single-board computer called Raspberry Pi. The anomaly detection service runs with low latency.

並列關鍵字

OBD-II ； anomaly detection ； car ； driving behavior

參考文獻

[1] M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat, G. Irving, M. Isard, et al. Tensorflow: A system for large­scale machine learning. In 12th USENIX symposium on operating systems design and implementation (OSDI 16), pages 265–283, 2016.

[2] Y.­S. Chou, Y.­C. Mo, J.­P. Su, W.­J. Chang, L.­B. Chen, J.­J. Tang, and C.­T. Yu. i­car system: A lora­based low power wide area networks vehicle diagnostic system for driving safety. In 2017 International Conference on Applied System Innovation (ICASI), pages 789–791, 2017.

[3] Elm Electronics Inc. ELM327 OBD to RS232 Interpreter [Online]. Avali­ able at https://www.elmelectronics.com/wp-content/uploads/2016/07/ ELM327DS.pdf (Accessed: January 2021).

[4] U. Flaig and A. Sieber. Electronic control units of bosch edc systems. Technical report, SAE Technical Paper, 1988.

[5] S. Hochreiter and J. Schmidhuber. Long short­term memory. Neural computation, 9(8):1735–1780, 1997.