ABSTRACT
Driving behavior modeling is an essential component of Advanced Driver Assistance Systems (ADAS). Existing methods usually analyze driving behaviors based on generic driving data, which do not consider personalization and user privacy. In this paper, we propose pBEAM, a collaborative cloud-edge computation system for personalized driving behavior modeling. The driving behavior model is built on top of Generative Adversarial Recurrent Neural Networks (GARNN), which adapts to the dynamic change of normal driving. Transfer learning from cloud to edge improves the model performance and robustness on the edge. We prune the deep neural networks in the cloud in order to minimize the model transferring load while maximally preserve the original model performance. A personalized edge model is trained on top of the pruned model using CGARNN-Edge (Conditional GARNN), which considers drivers' personal or contextual information as additional conditions. User privacy is well protected as no personal data needs to be uploaded to the cloud. Experimental results on driving data from both real world and driving simulator show that the proposed CGARNN-Edge achieves the best performance among all the methods.
- 2019. Collaborative Learning on the Edges: A Case Study on Connected Vehicles. In 2nd USENIX Workshop on Hot Topics in Edge Computing (HotEdge 19). USENIX Association, Renton, WA. https://www.usenix.org/conference/hotedge19/presentation/luGoogle Scholar
- Reza Abbasi-Asl and Bin Yu. 2017. Structural Compression of Convolutional Neural Networks Based on Greedy Filter Pruning. arXiv preprint arXiv:1705.07356 (2017).Google Scholar
- Benjamin F Bowne, Nicholas R Baker, Duane Lee Marzinzik, Matthew Eric Riley, Nick U Christopulos, Brian Mark Fields, J Lynn Wilson, Bryan T Wilkerson, David W Thurber, et al. 2013. Methods to Determine a Vehicle Insurance Premium Based on Vehicle Operation Data Collected Via a Mobile Device. US Patent App. 13/763,231.Google Scholar
- Daniel Ramage Brendan McMahan. 2017. Federated Learning: Collaborative Machine Learning without Centralized Training Data. Retrieved Sep 20, 2019 from https://ai.googleblog.com/2017/04/federated-learning-collaborative.htmlGoogle Scholar
- Markus M Breunig, Hans-Peter Kriegel, Raymond T Ng, and Jörg Sander. 2000. LOF: identifying density-based local outliers. In ACM Sigmod Record, Vol. 29. ACM, ACM, New York, NY, USA, 93--104.Google ScholarDigital Library
- Xsens Technologies B.V. 2009. Xsens Mti-G-700 user manual and technical documentation. Retrieved Sep 20, 2019 from https://projects.asl.ethz.ch/datasets/lib/exe/fetch.php?media=hardware:tiltinglaser:mti-g_user_manual_and_technical_documentation.pdfGoogle Scholar
- Emily L Denton, Soumith Chintala, Rob Fergus, et al. 2015. Deep generative image models using a laplacian pyramid of adversarial networks. In Advances in Neural Information Processing Systems (NIPS). 1486--1494.Google Scholar
- Xiaohan Ding, Guiguang Ding, Jungong Han, and Sheng Tang. 2018. Auto-balanced filter pruning for efficient convolutional neural networks. In Proceedings of the 32nd AAAI Conference on Artificial Intelligence (AAAI).Google Scholar
- Ziqiang Feng, Shilpa George, Jan Harkes, Padmanabhan Pillai, Roberta Klatzky, and Mahadev Satyanarayanan. 2018. Edge-Based Discovery of Training Data for Machine Learning. In 2018 IEEE/ACM Symposium on Edge Computing (SEC). IEEE, IEEE, 145--158.Google Scholar
- Lex Fridman, Daniel E. Brown, Michael Glazer, et al. 2017. MIT Autonomous Vehicle Technology Study: Large-Scale Deep Learning Based Analysis of Driver Behavior and Interaction with Automation. CoRR abs/1711.06976 (2017). arXiv:1711.06976 http://arxiv.org/abs/1711.06976Google Scholar
- Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. In Advances in Neural Information Processing Systems (NIPS). 2672--2680.Google Scholar
- Song Han, Jeff Pool, John Tran, and William Dally. 2015. Learning both weights and connections for efficient neural network. In Advances in Neural Information Processing Systems (NIPS). 1135--1143.Google Scholar
- Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long short-term memory. Neural computation 9, 8 (1997), 1735--1780.Google Scholar
- Jin-Hyuk Hong, Ben Margines, and Anind K Dey. 2014. A smartphone-based sensing platform to model aggressive driving behaviors. In Proceedings of the 32nd annual ACM conference on Human factors in computing systems (CHI). ACM, ACM, New York, NY, USA, 4047--4056.Google ScholarDigital Library
- Daniel Jiwoong Im, Chris Dongjoo Kim, Hui Jiang, and Roland Memisevic. 2016. Generating images with recurrent adversarial networks. arXiv preprint arXiv:1602.05110 (2016).Google Scholar
- Sinan Kaplan, Mehmet Amac Guvensan, Ali Gokhan Yavuz, and Yasin Karalurt. 2015. Driver behavior analysis for safe driving: A survey. IEEE Transactions on Intelligent Transportation Systems 16, 6 (2015), 3017--3032.Google ScholarDigital Library
- Jakub Konečnỳ, H Brendan McMahan, Felix X Yu, Peter Richtárik, Ananda Theertha Suresh, and Dave Bacon. 2016. Federated learning: Strategies for improving communication efficiency. arXiv preprint arXiv:1610.05492 (2016).Google Scholar
- A. Kuefler, J. Morton, T. Wheeler, and M. Kochenderfer. 2017. Imitating driver behavior with generative adversarial networks. In 2017 IEEE Intelligent Vehicles Symposium (IV). 204--211.Google Scholar
- Liangkai Liu, Xingzhou Zhang, Mu Qiao, and Weisong Shi. 2018. Safeshareride: Edge-based attack detection in ridesharing services. In 2018 IEEE/ACM Symposium on Edge Computing (SEC). IEEE, 17--29.Google ScholarCross Ref
- Chunmei Ma, Xili Dai, Jinqi Zhu, Nianbo Liu, Huazhi Sun, and Ming Liu. 2017. DrivingSense: Dangerous Driving Behavior Identification Based on Smartphone Autocalibration. Mobile Information Systems 2017 (2017).Google Scholar
- Yunlong Mao, Shanhe Yi, Qun Li, Jinghao Feng, Fengyuan Xu, and Sheng Zhong. 2018. Learning from differentially private neural activations with edge computing. In 2018 IEEE/ACM Symposium on Edge Computing (SEC). IEEE, IEEE, 90--102.Google ScholarCross Ref
- Mehdi Mirza and Simon Osindero. 2014. Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784 (2014).Google Scholar
- Olof Mogren. 2016. C-RNN-GAN: Continuous recurrent neural networks with adversarial training. arXiv preprint arXiv:1611.09904 (2016).Google Scholar
- Naic. 2018. Usage-based insurance and telematics, https://www.naic.org. Retrieved Sep 3, 2018 from https://www.naic.org/cipr_topics/topic_usage_based_insurance.htmGoogle Scholar
- World Health Organization. 2015. Global status report on road safety 2015. World Health Organization.Google Scholar
- I Pavlidis, M Dcosta, S Taamneh, M Manser, T Ferris, R Wunderlich, E Akleman, and P Tsiamyrtzis. 2016. Dissecting driver behaviors under cognitive, emotional, sensorimotor, and mixed stressors. Scientific reports 6 (2016), 25651.Google Scholar
- Xukan Ran, Haoliang Chen, Xiaodan Zhu, Zhenming Liu, and Jiasi Chen. 2018. DeepDecision: A Mobile Deep Learning Framework for Edge Video Analytics. In IEEE International Conference on Computer Communications (INFOCOM).Google Scholar
- Protection Regulation. 2016. General data protection regulation. Official Journal of the European Union 59 (2016), 1--88.Google Scholar
- Mahadev Satyanarayanan. 2017. The emergence of edge computing. Computer 50, 1 (2017), 339.Google Scholar
- Weisong Shi, Jie Cao, Quan Zhang, Youhuizi Li, and Lanyu Xu. 2016. Edge Computing: Vision and challenges. IEEE Internet of Things Journal 3, 5 (2016), 637--646.Google ScholarCross Ref
- Surat Teerapittayanon, Bradley McDanel, and HT Kung. 2017. Distributed deep neural networks over the cloud, the edge and end devices. In Distributed Computing Systems (ICDCS), 2017 IEEE 37th International Conference on. IEEE, 328--339.Google ScholarCross Ref
- Pengyang Wang, Yanjie Fu, Jiawei Zhang, Pengfei Wang, Yu Zheng, and Charu Aggarwal. 2018. You are how you drive: Peer and temporal-aware representation learning for driving behavior analysis. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. ACM, ACM, New York, NY, USA, 2457--2466.Google ScholarDigital Library
- Yifan Wang, Shaoshan Liu, Xiaopei Wu, and Weisong Shi. 2018. CAVBench: A Benchmark Suite for Connected and Autonomous Vehicles. In 2018 IEEE/ACM Symposium on Edge Computing (SEC). IEEE, IEEE, 30-42.Google Scholar
- Wei Wen, Chunpeng Wu, Yandan Wang, Yiran Chen, and Hai Li. 2016. Learning structured sparsity in deep neural networks. In Advances in Neural Information Processing Systems (NIPS). 2074--2082.Google Scholar
- Xsens. 2018. Xsens North America Inc, https://www.xsens.com/products/mti-g-710/. Retrieved Jun 4, 2018 from https://www.xsens.com/products/mti-g-710/Google Scholar
- Chuang-Wen You, Nicholas D Lane, Fanglin Chen, Rui Wang, Zhenyu Chen, Thomas J Bao, Martha Montes-de Oca, Yuting Cheng, Mu Lin, Lorenzo Torresani, et al. 2013. Carsafe app: Alerting drowsy and distracted drivers using dual cameras on smartphones. In Proceeding of the 11th annual international conference on Mobile systems, applications, and services. ACM, ACM, New York, NY, USA, 13--26.Google Scholar
- Jing Yuan, Yu Zheng, Xing Xie, and Guangzhong Sun. 2011. Driving with knowledge from the physical world. In Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, ACM, New York, NY, USA, 316--324.Google ScholarDigital Library
- Qingyang Zhang, Yifan Wang, Xingzhou Zhang, Liangkai Liu, Xiaopei Wu, Weisong Shi, and Hong Zhong. 2018. OpenVDAP: An Open Vehicular Data Analytics Platform for CAVs. In Distributed Computing Systems (ICDCS), 2018 IEEE 38th International Conference on. IEEE. IEEE, 1310-1320.Google ScholarCross Ref
- Xingzhou Zhang, Yifan Wang, and Weisong Shi. 2018. pCAMP: Performance Comparison of Machine Learning Packages on the Edges. In USENIX Workshop on Hot Topics in Edge Computing (HotEdge 18). USENIX Association, Boston, MA. https://www.usenix.org/conference/hotedge18/presentation/zhangGoogle Scholar
- Michael Zhu and Suyog Gupta. 2017. To prune, or not to prune: exploring the efficacy of pruning for model compression. arXiv preprint arXiv:1710.01878 (2017).Google Scholar
Index Terms
- Collaborative cloud-edge computation for personalized driving behavior modeling
Recommendations
Driving behavior model considering driver's over-trust in driving automation system
AutomotiveUI '19: Proceedings of the 11th International Conference on Automotive User Interfaces and Interactive Vehicular Applications: Adjunct ProceedingsLevels one to three of driving automation systems (DAS) become more and more sophisticated, not only the driver's driving skills will be reduced, but also the problem of over-trust will become serious. To prevent the over-trust, this paper discusses the ...
Scheme of Autonomous Vehicle Abnormal Behavior Detection Technology Based on Edge Computing
HPCCT & BDAI '20: Proceedings of the 2020 4th High Performance Computing and Cluster Technologies Conference & 2020 3rd International Conference on Big Data and Artificial IntelligenceAs the development of automatic driving technology, more and more attention has been paid to the safety and standardization of autonomous vehicles. Although high-definition map[4] navigation has been used to assist the automatic driving of vehicles, ...
Quantitative Evaluation of Distracted Driving by Using a PrARX Model
SMC '13: Proceedings of the 2013 IEEE International Conference on Systems, Man, and CyberneticsThis research develops a metric for the evaluation of an automobile driver's distraction based on a mathematical driving behavior model. Driving data was collected in a driving simulator. The primary task was to maintain a constant following distance ...
Comments