The research paper titled "Anomaly Detection from Cyber Threats via Infrastructure to Automated Vehicle" explores the use of Infrastructure-to-Vehicle (I2V) information in autonomous driving scenarios with limited visibility and high-density traffic. The authors highlight the benefits of utilizing external sensors to enhance the vehicle's sensing capabilities. However, one of the major challenges associated with I2V communication is cyber security. To address this challenge, the researchers propose a method to increase the vehicle's self-awareness in intersections by introducing an anomaly detection algorithm. This algorithm, implemented on the automated vehicle, assesses the health of I2V communication against various cyber security attacks. The analysis is conducted in a simulation environment using cyber-attack scenarios from the Secredas Project, which focuses on Cyber Security for Cross Domain Reliable Dependable Automated Systems. The study provides valuable insights into the limitations faced by autonomous vehicles when encountering different types and amplitudes of I2V cyber attacks, particularly when sensor redundancy is lost. The results demonstrate that anomalies injected during these attacks can be robustly detected and mitigated by the autonomous vehicle. This capability allows for safer and more comfortable driving experiences while maintaining accurate object tracking in intersections. The paper is authored by Chris van der Ploeg, Robin Smit, Alexis Siagkris-Lekkos, Frank Benders, and Emilia Silvas and has been accepted and submitted for presentation at IEEE European Control Conference 2021. It consists of 7 pages and includes 8 figures illustrating key concepts.
- The research paper explores the use of Infrastructure-to-Vehicle (I2V) information in autonomous driving scenarios with limited visibility and high-density traffic.
- External sensors are highlighted as beneficial for enhancing the vehicle's sensing capabilities.
- Cyber security is a major challenge associated with I2V communication.
- The researchers propose an anomaly detection algorithm to increase the vehicle's self-awareness in intersections.
- The algorithm assesses the health of I2V communication against various cyber security attacks.
- The analysis is conducted in a simulation environment using cyber-attack scenarios from the Secredas Project.
- The study demonstrates that anomalies injected during these attacks can be robustly detected and mitigated by the autonomous vehicle.
- The paper has been accepted and submitted for presentation at IEEE European Control Conference 2021.
The research paper is about using special information in self-driving cars when it's hard to see and there are lots of other cars. External sensors can help the car sense things better. Cyber security means protecting the car from bad people trying to hack into it. The researchers made a special program that helps the car know if something strange is happening at intersections. They tested this program using fake cyber attacks. The program worked well and could find and fix problems caused by the attacks. The paper will be presented at a conference."
Definitions- Infrastructure-to-Vehicle (I2V): Information that helps self-driving cars in difficult driving situations.
- Autonomous: When something can work on its own without needing someone to control it.
- Sensors: Special tools that help machines sense or detect things around them.
- Cyber security: Protecting computers or machines from bad people who try to hack into them.
- Anomaly detection algorithm: A special program that looks for strange or unusual things happening.
- Self-awareness: When something knows what's happening around it and can react to it.
- Intersections: Places where roads meet or cross each other.
- Cyber security attacks: Bad actions done by hackers trying to break into computer systems or machines.
- Simulation environment: A pretend place where scientists test things instead of doing it in real life.
- Robustly detected and mitigated: Strongly found and fixed problems caused by something bad happening.
Anomaly Detection from Cyber Threats via Infrastructure to Automated Vehicle
The advancement of autonomous vehicles has enabled the development of safer and more efficient transportation systems. However, one of the major challenges associated with this technology is cyber security. To address this challenge, researchers from the University of Twente have proposed a method to increase the vehicle's self-awareness in intersections by introducing an anomaly detection algorithm. This paper titled "Anomaly Detection from Cyber Threats via Infrastructure to Automated Vehicle" explores how Infrastructure-to-Vehicle (I2V) information can be utilized in autonomous driving scenarios with limited visibility and high-density traffic.
Background Information
Autonomous vehicles rely heavily on external sensors such as cameras, LiDAR, radar and GPS for navigation purposes. In addition to these sensors, I2V communication provides additional data about surrounding objects that may not be visible or accessible to onboard sensors due to their limited range or blocked view. This data can be used by automated vehicles for object tracking in intersections and other areas where visibility is reduced.
However, I2V communication also introduces new cyber security risks that must be addressed before it can be safely implemented in autonomous driving applications. To mitigate these risks, the authors propose an anomaly detection algorithm that assesses the health of I2V communication against various cyber security attacks. The analysis is conducted in a simulation environment using cyber-attack scenarios from the Secredas Project which focuses on Cyber Security for Cross Domain Reliable Dependable Automated Systems (CREDAS).
Methodology
The proposed anomaly detection algorithm consists of two main components: feature extraction and classification models. For feature extraction, statistical features are extracted from raw sensor data collected during normal operation as well as during simulated attack scenarios provided by Secredas Project dataset . These features are then used as input into a supervised learning model which classifies each sample into either “normal” or “anomalous” categories based on its similarity with previously seen samples . The model was trained using a Support Vector Machine (SVM) classifier with radial basis function kernel .
Results & Discussion
The results demonstrate that anomalies injected during various types and amplitudes of I2V cyber attacks can be robustly detected and mitigated by the autonomous vehicle without compromising accuracy or safety performance . This capability allows for safer and more comfortable driving experiences while maintaining accurate object tracking in intersections even when sensor redundancy is lost due to malicious interference .
Conclusion
This research paper provides valuable insights into how Infrastructure-to-Vehicle (I2V) information can enhance sensing capabilities while mitigating potential cyber threats faced by automated vehicles operating under limited visibility conditions . By introducing an anomaly detection algorithm , it becomes possible for automated vehicles to detect malicious interference quickly , allowing them to maintain accurate object tracking even when sensor redundancy is lost due to malicious interference . The results demonstrate that this approach could provide significant benefits for safe navigation under challenging conditions while ensuring secure transmission between infrastructure elements and automated vehicles .
