Modern automobiles have several processing elements called Electronic Control Units (ECUs) that control different functionalities in a vehicle. ECUs run various types of automotive applications such as anti-lock braking control, cruise control, and so on. Most of these automotive applications have strict timing (deadline) and latency constraints, and thus they are classified as hard real-time applications. The ECUs on which these applications execute are distributed across the vehicle and communicate with each other by exchanging messages. These messages can be classified as either time-triggered or event-triggered. Time-triggered messages are periodically generated messages originating from safety-critical software applications. In contrast, event-triggered messages
are generated asynchronously when a specific event occurs, typically by low-priority (e.g., maintenance) applications. The diverse nature of messages in automotive systems requires unique communication protocols and robust networks to support them. The advent of Advanced Driver Assistance Systems (ADAS) in vehicles has resulted in an increase in the number of ECUs, which in turn has increased the complexity of the in-vehicle network and the entire automotive system. It is projected that in the near future, improving ADAS effectiveness will require connecting to external systems using vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) protocols. This increased connectivity will make vehicle networks more vulnerable to sophisticated security attacks.
The research objective of this project is design novel scheduling, allocation, and adaptive management techniques to support secure, real-time, and fault-tolerant communication of messages across automotive networks. The project also involves aspects of anomaly detection using machine learning techniques, to counter attackers targeting automotive networks. The focus is on adapting and deploying these strategies on automotive network standards such as CAN, Flexray, and TTEthernet.
Selected Publications
S. V. Thiruloga, V. K. Kukkala, and S. Pasricha, “TENET: Temporal CNN with Attention for Anomaly Detection in Automotive Cyber-Physical Systems”, IEEE/ACM Asia & South Pacific Design Automation Conference (ASPDAC), Jan 2022. . (Best Paper Award Candidate)
V. K. Kukkala, S. V. Thiruloga, and S. Pasricha, “LATTE: LSTM Self-Attention based Anomaly Detection in Embedded Automotive Platforms”, ACM Transactions on Embedded Computing Systems (TECS), 2021.
V. K. Kukkala, S. V. Thiruloga, and S. Pasricha, “INDRA: Intrusion Detection using Recurrent Autoencoders in Automotive Embedded Systems”, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, (TCAD), 39(11), Nov 2020.
V. Kukkala, S. Pasricha, T. H. Bradley, “SEDAN: Security-Aware Design of Time-Critical Automotive Networks”, IEEE Transactions on Vehicular Technology (TVT), vol. 69, no. 8, Aug 2020
V. Kukkala, S. Pasricha, T. H. Bradley, “JAMS-SG: A Framework for Jitter-Aware Message Scheduling for Time-Triggered Automotive Networks”, ACM Transactions on Design Automation of Electronic Systems (TODAES), Vol. 24, Iss. 6, Nov 2019
V. K. Kukkala, S. Pasricha, T. Bradley, “JAMS: Jitter-Aware Message Scheduling for FlexRay Automotive Networks,” IEEE/ACM International Symposium on Networks-on-Chip (NOCS), Oct 2017.
V. K. Kukkala, T. Bradley, S. Pasricha, “Uncertainty Analysis and Propagation for an Auxiliary Power Module,” IEEE Transportation and Electrification Conference (TEC), 2017.
V. K. Kukkala, T. Bradley, S. Pasricha, “Priority-based Multi-level Monitoring of Signal Integrity in a Distributed Powertrain Control System,” 4th IFAC Workshop on Engine and Powertrain Control, Simulation and Modeling, Jul 2015.