Cyber-Physical Security and Safety of Autonomous Connected Vehicles: Optimal Control Meets Multi-Armed Bandit Learning

Autonomous connected vehicles (ACVs) rely on intra-vehicle sensors such as camera and radar as well as inter-vehicle communication to operate effectively which exposes them to cyber and physical attacks in which an adversary can manipulate sensor readings and physically control the ACVs. In this paper, a comprehensive control and learning framework is proposed to thwart cyber and physical attacks on ACV networks. First, an optimal safe controller for ACVs is derived to maximize the street traffic flow while minimizing the risk of accidents by optimizing the ACV speed and inter-ACV spacing. It is proven that the proposed controller is robust to physical attacks which aim at making ACV systems unstable. Next, two data injection attack (DIA) detection approaches are proposed to address cyber attacks on sensors and their physical impact on the ACV system. The proposed approaches rely on leveraging the stochastic behavior of the sensor readings and on the use of a multi-armed bandit (MAB) algorithm. It is shown that, collectively, the proposed DIA detection approaches minimize the vulnerability of ACV sensors against cyber attacks while maximizing the ACV system’s physical robustness. Simulation results show that the proposed optimal safe controller outperforms the current state of the art controllers by maximizing the robustness of ACVs to physical attacks. The results also show that the proposed DIA detection approaches, compared to Kalman filtering, can improve the security of ACV sensors against cyber attacks and ultimately improve the physical robustness of an ACV system.

[1]  Panagiotis Papadimitratos,et al.  On the Performance of Secure Vehicular Communication Systems , 2011, IEEE Transactions on Dependable and Secure Computing.

[2]  Kaare Brandt Petersen,et al.  The Matrix Cookbook , 2006 .

[3]  Dong Hoon Lee,et al.  A Practical Wireless Attack on the Connected Car and Security Protocol for In-Vehicle CAN , 2015, IEEE Transactions on Intelligent Transportation Systems.

[4]  Yunpeng Wang,et al.  Long short-term memory neural network for traffic speed prediction using remote microwave sensor data , 2015 .

[5]  Stefania Santini,et al.  A collaborative approach for improving the security of vehicular scenarios: The case of platooning , 2018, Comput. Commun..

[6]  Tomas Olovsson,et al.  Security aspects of the in-vehicle network in the connected car , 2011, 2011 IEEE Intelligent Vehicles Symposium (IV).

[7]  Walid Saad,et al.  Robust Deep Reinforcement Learning for Security and Safety in Autonomous Vehicle Systems , 2018, 2018 21st International Conference on Intelligent Transportation Systems (ITSC).

[8]  Robert D. Kleinberg,et al.  Regret bounds for sleeping experts and bandits , 2010, Machine Learning.

[9]  Liam Kilmartin,et al.  Intra-Vehicle Networks: A Review , 2015, IEEE Transactions on Intelligent Transportation Systems.

[10]  Christian Laugier,et al.  Comparison of parametric and non-parametric approaches for vehicle speed prediction , 2014, 2014 American Control Conference.

[11]  Dirk Helbing,et al.  Enhanced intelligent driver model to access the impact of driving strategies on traffic capacity , 2009, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[12]  Kwok-Yan Lam,et al.  Wireless Communication and Security Issues for Cyber–Physical Systems and the Internet-of-Things , 2018, Proceedings of the IEEE.

[13]  Robert L. Williams,et al.  Linear State-Space Control Systems , 2007 .

[14]  Dipak Ghosal,et al.  Security vulnerabilities of connected vehicle streams and their impact on cooperative driving , 2015, IEEE Communications Magazine.

[15]  Huaiyu Dai,et al.  A Survey on Low Latency Towards 5G: RAN, Core Network and Caching Solutions , 2017, IEEE Communications Surveys & Tutorials.

[16]  P. Olver.  Nonlinear Systems , 2013 .

[17]  Wei Yu,et al.  On false data injection attacks against Kalman filtering in power system dynamic state estimation , 2016, Secur. Commun. Networks.

[18]  R. M. A. P. Rajatheva,et al.  Sleeping Multi-Armed Bandit Learning for Fast Uplink Grant Allocation in Machine Type Communications , 2018, IEEE Transactions on Communications.

[19]  Elias Bou-Harb,et al.  Assessing Internet-wide Cyber Situational Awareness of Critical Sectors , 2018, ARES.

[20]  Li Sun,et al.  Hybrid secure beamforming and vehicle selection using hierarchical agglomerative clustering for C-RAN-based vehicle-to-infrastructure communications in vehicular cyber-physical systems , 2016, Int. J. Distributed Sens. Networks.

[21]  Walid Saad,et al.  Unmanned Aerial Vehicle With Underlaid Device-to-Device Communications: Performance and Tradeoffs , 2015, IEEE Transactions on Wireless Communications.

[22]  Walid Saad,et al.  Stochastic optimization and control framework for 5G network slicing with effective isolation , 2018, 2018 52nd Annual Conference on Information Sciences and Systems (CISS).

[23]  Panagiotis Papadimitratos,et al.  Secure vehicular communication systems: implementation, performance, and research challenges , 2008, IEEE Communications Magazine.

[24]  Peter Auer,et al.  Finite-time Analysis of the Multiarmed Bandit Problem , 2002, Machine Learning.

[25]  Calin Belta,et al.  Provably Safe Cruise Control of Vehicular Platoons , 2017, IEEE Control Systems Letters.

[26]  Walid Saad,et al.  Deep Learning for Signal Authentication and Security in Massive Internet-of-Things Systems , 2018, IEEE Transactions on Communications.

[27]  Li Pan,et al.  Predicting Short-Term Traffic Flow by Long Short-Term Memory Recurrent Neural Network , 2015, 2015 IEEE International Conference on Smart City/SocialCom/SustainCom (SmartCity).

[28]  Anupam Joshi,et al.  OBD_SecureAlert: An Anomaly Detection System for Vehicles , 2016, 2016 IEEE International Conference on Smart Computing (SMARTCOMP).

[29]  S. Liberty,et al.  Linear Systems , 2010, Scientific Parallel Computing.

[30]  Sandip Roy,et al.  Security concepts for the dynamics of autonomous vehicle networks , 2014, Autom..

[31]  Walid Saad,et al.  Deep Learning for Reliable Mobile Edge Analytics in Intelligent Transportation Systems: An Overview , 2017, IEEE Vehicular Technology Magazine.

[32]  Yeng Chai Soh,et al.  Optimal Kalman filtering with random sensor delays, packet dropouts and missing measurements , 2009, 2009 American Control Conference.

[33]  Walid Saad,et al.  Joint Communication and Control for Wireless Autonomous Vehicular Platoon Systems , 2018, IEEE Transactions on Communications.

[34]  Francesco Borrelli,et al.  A Learning-Based Framework for Velocity Control in Autonomous Driving , 2016, IEEE Transactions on Automation Science and Engineering.

[35]  Ryan M. Gerdes,et al.  A data trust framework for VANETs enabling false data detection and secure vehicle tracking , 2017, 2017 IEEE Conference on Communications and Network Security (CNS).

[36]  Walid Saad,et al.  Artificial Intelligence for Wireless Connectivity and Security of Cellular-Connected UAVs , 2018, ArXiv.

[37]  P. G. Gipps,et al.  A behavioural car-following model for computer simulation , 1981 .

[38]  S.J. Julier,et al.  Fusion of time delayed measurements with uncertain time delays , 2005, Proceedings of the 2005, American Control Conference, 2005..

[39]  Xin Zhang,et al.  VANET alert endorsement using multi-source filters , 2010, VANET '10.

[40]  Walid Saad,et al.  A colonel blotto game for interdependence-aware cyber-physical systems security in smart cities , 2017, SCOPE@CPSWeek.

[41]  Tareq Y. Al-Naffouri,et al.  Enhancing Secrecy With Multiantenna Transmission in Millimeter Wave Vehicular Communication Systems , 2016, IEEE Transactions on Vehicular Technology.

[42]  Justin M. Bradley,et al.  Optimization and Control of Cyber-Physical Vehicle Systems , 2015, Sensors.

[43]  Christoph Schmittner,et al.  The Need for Safety and Cyber-Security Co-engineering and Standardization for Highly Automated Automotive Vehicles , 2016 .