Periodic Reporting for period 1 - SELFY (SELF assessment, protection & healing tools for a trustworthY and resilient CCAM)
Período documentado: 2022-06-01 hasta 2023-11-30
SELFY is a European funded project under the Horizon Europe programme, HORI-ZON-CL5-2021-D6-01-04. SELFY will increase CCAM ecosystem’s safety, security, robustness, and resilience by researching and developing a toolbox made of collabo-rative tools of which the main pillars are:
i. situational awareness, through collaborative perception and monitoring
ii. data sharing, advanced processing for detection of malicious events and deci-sion-making
iii. resilience, increased ability to adapt and respond to cyber-threats and cyber-attacks
iv. trust, including guarantees regarding privacy, confidentiality, integrity and im-mutability of data in a collaborative CCAM environment
SELFY tools can operate individually and/or cooperatively with other tools, generating a distributed global solution, where SELF-protection, SELF-response and SELF-re-covery decisions will be managed locally or globally as appropriate for any given cyber-attack, malicious activity or hazard, extending the Operational Design Domain (ODD). In Europe, 50 million connected cars are expected in circulation by 2026, and EU reg-ulations require for cybersecurity certificates for vehicles. In this scenario, SELFY long term impact is grounded on a value proposition that is OEM and/or vendor agnostic, thus facilitating adoption at all stages of the value chain. Additionally, the consortium involves all necessary stakeholders for co-creating a sound and comprehensive CCAM solution: including OEMs, road operators, traffic and infrastructure manage-ment, researchers, and policy makers.
SELFY expected outcomes include 5 traffic and infrastructure management organiza-tions adopting the SELFY toolbox, which will benefit of a >90% effectiveness rate in detection of vulnerable vehicles and security breaches and an increase of >75% in the mitigation rate at demonstrated in TRL6 by the end of the project.
i. situational awareness, through collaborative perception and monitoring
ii. data sharing, advanced processing for detection of malicious events and deci-sion-making
iii. resilience, increased ability to adapt and respond to cyber-threats and cyber-attacks
iv. trust, including guarantees regarding privacy, confidentiality, integrity and im-mutability of data in a collaborative CCAM environment
SELFY tools can operate individually and/or cooperatively with other tools, generating a distributed global solution, where SELF-protection, SELF-response and SELF-re-covery decisions will be managed locally or globally as appropriate for any given cyber-attack, malicious activity or hazard, extending the Operational Design Domain (ODD). In Europe, 50 million connected cars are expected in circulation by 2026, and EU reg-ulations require for cybersecurity certificates for vehicles. In this scenario, SELFY long term impact is grounded on a value proposition that is OEM and/or vendor agnostic, thus facilitating adoption at all stages of the value chain. Additionally, the consortium involves all necessary stakeholders for co-creating a sound and comprehensive CCAM solution: including OEMs, road operators, traffic and infrastructure manage-ment, researchers, and policy makers.
SELFY expected outcomes include 5 traffic and infrastructure management organiza-tions adopting the SELFY toolbox, which will benefit of a >90% effectiveness rate in detection of vulnerable vehicles and security breaches and an increase of >75% in the mitigation rate at demonstrated in TRL6 by the end of the project.
During this first period, all the WP have started.
The WP1 has been implementing the actions to verify the overall progress of the project both in the scientific activities and financial, legal and administrative fields as well as ethics and data management.
WP2 has put the design basis for the technical WPs. Scenarios, use cases, requirements and System Architectures have been defined, as well as interfaces between tools. Its outputs have been collected by the first technical task of the three technical WPs. There-fore, T3.1 T4.1 and T5.1 have been the link to the SACP (WP3), CRHS (WP4) and TDMS (WP5) architectures. WP3, WP4 and WP5 have developed their relative tools with a 40-60% implementation state and validate some partial results.
WP6 has started defining a validation plan and methodologies. In this WP the integration task of the tools to validate the use cases has also started. The validation activities of simulation and real world validation have also started and will be more active during the second period of SELFY project.
WP7 has defined the communication and dissemination strategy, and a set of promo-tional materials. Dissemination has been done with SELFY first results. In task T7.2 a plan was set up for a proper management of the exploitable results, consisting of five phases. T7.3 related to standardization and social sciences also started with the prioriti-sation of standards selected in WP2 and the engagement of the AB and external stake-holders in three workshops. A first accessibility analysis of the web page was carried out, as well as a methodology to gather recommendations and guidelines to get to know the implications that cybersecurity within the field of autonomous mobility has on the groups of women, elderly and disabled.
Overall, in the project there are 11 main primary datasets for validation, from which some are private, others are under discussion to be left open in public repositories. In the second period, the decision will be made, and the open datasets will be uploaded to SYGMA.
The WP1 has been implementing the actions to verify the overall progress of the project both in the scientific activities and financial, legal and administrative fields as well as ethics and data management.
WP2 has put the design basis for the technical WPs. Scenarios, use cases, requirements and System Architectures have been defined, as well as interfaces between tools. Its outputs have been collected by the first technical task of the three technical WPs. There-fore, T3.1 T4.1 and T5.1 have been the link to the SACP (WP3), CRHS (WP4) and TDMS (WP5) architectures. WP3, WP4 and WP5 have developed their relative tools with a 40-60% implementation state and validate some partial results.
WP6 has started defining a validation plan and methodologies. In this WP the integration task of the tools to validate the use cases has also started. The validation activities of simulation and real world validation have also started and will be more active during the second period of SELFY project.
WP7 has defined the communication and dissemination strategy, and a set of promo-tional materials. Dissemination has been done with SELFY first results. In task T7.2 a plan was set up for a proper management of the exploitable results, consisting of five phases. T7.3 related to standardization and social sciences also started with the prioriti-sation of standards selected in WP2 and the engagement of the AB and external stake-holders in three workshops. A first accessibility analysis of the web page was carried out, as well as a methodology to gather recommendations and guidelines to get to know the implications that cybersecurity within the field of autonomous mobility has on the groups of women, elderly and disabled.
Overall, in the project there are 11 main primary datasets for validation, from which some are private, others are under discussion to be left open in public repositories. In the second period, the decision will be made, and the open datasets will be uploaded to SYGMA.
During this first period SELFY has progressed beyond the state of the art in these items already identified in the proposal phase. The tools are at 40-60% of implementation and not yet validated within the selected scenarios and use cases.
1. SACP tools: RSU and Smart Vehicle with Camera and Sensor Integration: De-veloping a robust on-road cooperative perception system using multiples sensors and extending the capabilities of the on-board perception system exchanging and fusing data with other CCAM’s agents. SELFY aggregates shared sensor data which deal with spatial and temporal inaccuracies and differentiates attacks and classical sensor failures. Fusing the objects extracted by video analysis tools it’s able to detect anomalies and anomalous behaviour
2. CRHS tools: CRHS tools provide continuous status of the security of their ele-ments, detecting anomalies and auditing in a circular and continuous way all the fleet of vehicles. VSOC gets a complete status of the system and can set trust levels to the different CCAM’s agents. The Validation and Verification and the run-time execution techniques based on modelling enhance the robustness of designs in CCAM systems. New algorithms are developed to robustify CCAM’s situations like Platooning, maneuvers… A supervisor system is capable of con-tinuous self-assessment of the vehicle and its surroundings and can participate in the decision-making strategies. Safe operational modes and fallback strategies allows vehicles to remain operative in case of relevant performance failures. Inmune-based system using AI can detect and learn from compromised situa-tions. Algorithms based on AI are applied for the global processing of alerts and incidents in the VSOC, which is a first approach to a holistic monitoring and con-trol of the vehicle fleets in CCAM system helping OEMs to meet the current UNECE WP.29 regulation.
3. TDMS tools: TDMS tools provide reinforcement in security communications with remote attestation techniques which can be combined with Secure Update Mech-anisms. Binary Scanning tools also can be used to validate a new image. Post Quantum Cryptography approaches also protect CCAM vehicles against brute force attacks in a long-term scenario. TDMS tools also focus on privacy and anonymization, with AI algorithms which detect faces and license plates and per-forms distortion of these regions in the images and also provide strategies to change identifiers related to a vehicle in V2X services in order to preserve end user privacy.
1. SACP tools: RSU and Smart Vehicle with Camera and Sensor Integration: De-veloping a robust on-road cooperative perception system using multiples sensors and extending the capabilities of the on-board perception system exchanging and fusing data with other CCAM’s agents. SELFY aggregates shared sensor data which deal with spatial and temporal inaccuracies and differentiates attacks and classical sensor failures. Fusing the objects extracted by video analysis tools it’s able to detect anomalies and anomalous behaviour
2. CRHS tools: CRHS tools provide continuous status of the security of their ele-ments, detecting anomalies and auditing in a circular and continuous way all the fleet of vehicles. VSOC gets a complete status of the system and can set trust levels to the different CCAM’s agents. The Validation and Verification and the run-time execution techniques based on modelling enhance the robustness of designs in CCAM systems. New algorithms are developed to robustify CCAM’s situations like Platooning, maneuvers… A supervisor system is capable of con-tinuous self-assessment of the vehicle and its surroundings and can participate in the decision-making strategies. Safe operational modes and fallback strategies allows vehicles to remain operative in case of relevant performance failures. Inmune-based system using AI can detect and learn from compromised situa-tions. Algorithms based on AI are applied for the global processing of alerts and incidents in the VSOC, which is a first approach to a holistic monitoring and con-trol of the vehicle fleets in CCAM system helping OEMs to meet the current UNECE WP.29 regulation.
3. TDMS tools: TDMS tools provide reinforcement in security communications with remote attestation techniques which can be combined with Secure Update Mech-anisms. Binary Scanning tools also can be used to validate a new image. Post Quantum Cryptography approaches also protect CCAM vehicles against brute force attacks in a long-term scenario. TDMS tools also focus on privacy and anonymization, with AI algorithms which detect faces and license plates and per-forms distortion of these regions in the images and also provide strategies to change identifiers related to a vehicle in V2X services in order to preserve end user privacy.