Artificial Intelligence based cybersecurity for connected and automated vehicles

Projektinformationen

CARAMEL

ID Finanzhilfevereinbarung: 833611

DOI

10.3030/833611

Projekt abgeschlossen

EK-Unterschriftsdatum 27 September 2019

Startdatum 1 Oktober 2019

Enddatum 30 Juni 2022

Finanziert unter

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Gesamtkosten

€ 7 033 655,10

EU-Beitrag

€ 4 998 503,51

Koordiniert durch

FUNDACIO PRIVADA I2CAT, INTERNET I INNOVACIO DIGITAL A CATALUNYA
Spain

Periodic Reporting for period 2 - CARAMEL (Artificial Intelligence based cybersecurity for connected and automated vehicles)

Berichtszeitraum: 2021-01-01 bis 2022-06-30

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

From padded dashboards to seat belts and from rear-view cameras to active safety measures. Nowadays, cars becoming safer, smarter, and “greener” through connectivity, Artificial Intelligence (AI), and Machine Learning (ML), while cybersecurity aims for more sustainable safer roads with zero fatality. The CARAMEL project developed cybersecurity solutions for the new generation of cars: i) autonomous cars, ii) 5G connected vehicles iii) electromobility and iv) Remote control vehicles. CARAMEL applied a proactive method based on AI and ML techniques to mitigate cybersecurity-originated safety risks on roads. Considering the entire supply chain, CARAMEL developed innovative anti-hacking intrusion detection/prevention systems for the European automotive industry. So far, CARAMEL analyzed the security and privacy requirements of future mobility, identified the attack surface and modeled the potential threats, designed the overall architecture and defined the system specifications, elaborated use cases, and identified cyber threat detection and response techniques for all four project's pillars. Moreover, CARAMEL worked on enabling technologies and ML/AI techniques to detect and mitigate cyber threats to future mobility, in particular, innovative distributed PKI infrastructure, AI-based context-rich and context-aware solutions, holistic ML-based solutions based on the fusion of multiple data sources, and the CARAMEL backend to help mobility actors like road owners, traffic managers, etc. In addition, CARAMEL's design of an innovative anti-hacking device is a key solution to protect future roads and vehicles.
The damaging effects of cyberattacks on Cooperative Connected and Automated Mobility (CCAM) can be tremendous. Such as the damage to the reputation of vehicle manufacturers, the increased denial of customers to adopt CCAM, the loss of working hours, material damages, increased environmental pollution due e.g. to traffic jams or malicious modifications in sensors’ firmware, and the danger for human lives. CARAMEL proactively addresses modern vehicle cybersecurity challenges by using advanced AI and ML techniques while seeking methods to mitigate associated safety risks. Addressing cybersecurity considerations for current CCAM vehicles, well-established ICT sector methodologies will be adopted, allowing vulnerabilities and potential cyberattack impacts assessment. Past automotive industry initiatives and cybersecurity projects have reached security assurance frameworks for networked vehicles, but several newly introduced technological dimensions like 5G, autopilots, and smart charging of Electric Vehicles (EVs) introduce cybersecurity gaps, not been fully addressed yet.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

The CARAMEL project applies cybersecurity approaches to detect and mitigate cybersecurity threats in the automotive domain. The project focuses on four selected pillars: In pillar 1 develop innovative technologies to detect attacks against the vehicle sensors, e.g. disturbs the car object detection algorithms. To this end, a novel intrusion detection system – the anti-hacking device – was integrated directly into the vehicle. The anti-hacking device uses ML technology to detect attacks. Due to its unique design, the detection algorithms can easily, safely, and rapidly be updated regularly to counter novel attacks in a short time. CARAMEL secured autonomous or semi-autonomous driving against advanced attack scenarios and will drive the acceptance and adoption of these innovations by the European general public.
In pillar 2 CARAMEL developed advanced attack detection technologies for the CCAM vehicle, mainly the V2X protocols between cars and roadside units. Additionally, partners implemented innovative threat detection mechanisms for direct integration into onboard units. Additionally, CARAMEL developed an innovative bridge technology to make competing 802.11p and C-V2X networking technologies compatible using MEC devices or roadside units. In pillar 3 partners implemented security mechanisms to protect against European EV charging infrastructure wide-scale attacks: Charging stations are part of the European critical infrastructure, and attacks that cripple this infrastructure would affect the flow of passengers and goods in Europe, CARAMEL implemented ML models to detect attacks on the whole eCharging backend infrastructure to facilitate early mitigations.
In pillar 4 the Korean partners, who still have one year of project based on Korean funding, continued working on remote control vehicle based on mmWAVE (23GHz) use case, building a data processing architecture and developing a Malicious Traffic Detection Solution LSTM (Long-Short Term Memory)-based cyber-attack anomaly prediction/detection.
All these activities made the evolution of Automotive technology to a more connected and software-driven future more secure, therefore driving public acceptance of these technologies in Europe as well as giving European companies a competitive advantage.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

The CARAMEL project applies cybersecurity approaches to detect and mitigate cybersecurity threats in the automotive domain. CARAMEL's Innovation action focuses on four selected pillars: In pillar 1 develop innovative technologies to detect attacks against the vehicle sensors, e.g. disturbs the car object detection algorithms. To this end, a novel intrusion detection system – the anti-hacking device – was integrated directly into the vehicle. The anti-hacking device uses ML technology to detect attacks. Due to its unique design, the detection algorithms can easily, safely, and rapidly be updated regularly to counter novel attacks in a short time. CARAMEL secured autonomous or semi-autonomous driving against advanced attack scenarios and will drive the acceptance and adoption of these innovations by the European general public.
In pillar 2 CARAMEL developed advanced attack detection technologies for the CCAM vehicle, mainly the V2X protocols between cars and roadside units. Additionally, partners implemented innovative threat detection mechanisms for direct integration into onboard units. Additionally, CARAMEL developed an innovative bridge technology to make competing 802.11p and C-V2X networking technologies compatible using MEC devices or roadside units. In pillar 3 partners implemented security mechanisms to protect against European EV charging infrastructure wide-scale attacks: Charging stations are part of the European critical infrastructure, and attacks that cripple this infrastructure would affect the flow of passengers and goods in Europe, CARAMEL implemented ML models to detect attacks on the whole eCharging backend infrastructure to facilitate early mitigations.
In pillar 4 the Korean partners, who still have one year of project based on Korean funding, continued working on remote control vehicle based on mmWAVE (23GHz) use case, building a data processing architecture and developing a Malicious Traffic Detection Solution LSTM (Long-Short Term Memory)-based cyber-attack anomaly prediction/detection.
All these CARAMEL activities made the evolution of Automotive technology to a more connected and software-driven future more secure, therefore driving public acceptance of these technologies in Europe as well as giving European companies a competitive advantage.

imatge.png