Analysis of the Efficacy of the Use of Inertial Measurement and Global Positioning System Data to Reverse Engineer Automotive CAN Bus Steering Signals

Overview

• Autonomous vehicle control is becoming more available in new cars, and there is a need to retrofit older vehicles with this capability.
• Automotive cybersecurity is a significant concern due to documented attacks on vehicles.
• Researchers have been exploring reverse engineering techniques to automate vehicle control and improve vehicle security and threat analysis.

Plain English Explanation

In this paper, the researchers describe a method for reverse-engineering a vehicle's steering wheel position by using data from the vehicle's internal sensors, such as the GPS and inertial measurement unit (IMU). This can be done without any prior knowledge of the specific vehicle's design.

The researchers build on previous work that used reverse engineering to identify a vehicle's accelerator and brake pedal control channels. However, the deceleration results were not as accurate as the acceleration results. By incorporating GPS data, the researchers show that they can significantly improve the accuracy of identifying the brake pedal control channels.

This research has potential applications in automated vehicle control and improving vehicle security. By reverse-engineering the vehicle's control systems, researchers can better understand how these systems work and identify potential vulnerabilities that could be exploited by attackers.

Technical Explanation

The researchers used a combination of GPS and IMU data to reverse-engineer the steering wheel position CAN (Controller Area Network) channels on a vehicle, without any prior knowledge of the vehicle's design. CAN is a communication protocol used in vehicles to connect various electronic control units (ECUs).

The researchers first identified the accelerator and brake pedal CAN channels using reverse engineering techniques, as in their previous work. They then incorporated GPS data to improve the correlation between the sensor data and the deceleration CAN channels, particularly for the brake pedal.

The researchers demonstrated the effectiveness of this approach through experiments on a real vehicle. They were able to achieve high correlation values between the sensor data and the identified CAN channels, showing the viability of this reverse engineering method.

Critical Analysis

The researchers acknowledge that their method relies on the availability of GPS and IMU data, which may not be present in all vehicles. They also note that the specific CAN message formats and vehicle-specific parameters would need to be determined for each individual vehicle.

Additionally, the researchers do not address the potential ethical and legal implications of reverse engineering vehicle control systems, particularly in the context of vehicle security. There could be concerns about the misuse of this technology by malicious actors, and the researchers could have discussed these issues in more depth.

Furthermore, the researchers could have explored the potential limitations of their approach, such as the impact of environmental conditions or the reliability of the sensor data. They could also have discussed potential avenues for future research, such as extending the method to other vehicle subsystems or exploring ways to automate the reverse engineering process further.

Conclusion

This research demonstrates the feasibility of using sensor data, such as GPS and IMU, to reverse-engineer a vehicle's steering wheel position CAN channels without prior knowledge of the vehicle's design. This has important implications for automotive cybersecurity and automated vehicle control, as it allows researchers to better understand and potentially improve the security and functionality of these critical vehicle systems.