Ensuring the safety, resilience and security of waterborne digital systems

(HORIZON-CL5-2024-D6-01-10)
https://ec.europa.eu/info/funding-tenders/opportunities/portal/screen/opportunities/topic-details/horizon-cl5-2024-d6-01-
10?tenders=false&openForSubmission=false&programmePeriod=2021%20-
%202027&frameworkProgramme=43108390&pageNumber=10

Call:
Safe, Resilient Transport and Smart Mobility services for
passengers and goods (HORIZON-CL5-2024-D6-01)
Type of action: HORIZON-RIA HORIZON Research and Innovation Actions
Type of MGA: HORIZON Action Grant Budget-Based [HORIZON-AG]

Planned opening date: 07 May 2024

Deadline date: 05 September 2024 17:00:00 Brussels time

Sanja Bauk, Leonidas Tsiopoulos

10/01/2024
Expected Outcome – Project Outputs:
• Increased safety and resilience of waterborne digital
systems, including system of systems and their functions
and considering both malicious intervention and system
failure with regard to the application of artificial intelligence
methodologies, networks of sensors and the onshore on-
board communications.
• Improved system design addressing human factors issues in
the changing levels of human/automated system
interactions
• Assurance of the resilience, safety and security of
waterborne digital and connected systems is undertaken
on the basis of robust methodologies to a similar standard to
that applied within other sectors (aerospace, nuclear,
medical, automotive etc.) which apply safety critical digital
technology and their application in safety critical conditions
including the safety of navigation and its systems.
Expected Outcome – Project Outputs:

• Robust by design waterborne digital and connected

systems for safety and resilience, incl. reliability regimes
such as fail safe, fail secure, fail to operation etc., Hazard
Operability (HAZOP) system of systems, security,
hardware and equipment data, etc.
• Methodologies to enable effective HAZOP analysis and
validation of waterborne digital systems are developed and
disseminated, increasing the use of common approaches,
also when using artificial intelligence applications.
• Increased software safety, incl. functional analysis and
reliability assessment).
• Increased cyber security for operation and maintenance,
incl. software maintenance.
Expected Outcome – Project Outputs:

• … a HAZOP methodology for whole system assessment

of highly digitized, connected complex vessels …
• On-board systems and functions integration by design,
for safe and secure operation should be used to test and
demonstrate the safety and security of the applications.
• The developed methodology will be applied to a
representative complex highly digitized vessel, safety
critical systems and functions will be identified, and
appropriate reliability regimes and mitigation measures will
be established with consideration of both malicious
intervention and system failure.
Expected Outcome – Project Outputs:

• Comprehensive HAZOP (Hazard Operability) studies are

essential for such vessels, yet the methodologies are poorly
established...
• This can be difficult for waterborne transport due to the
variability of ship designs, complexity and lack of relevant
data concerning the integrated components.
• HAZOP, yet such trials focus on hardware or subsystems
such as rudder control rather than addressing the entire
integration…
• For safety critical systems, reliability regimes need to be
established to identify the safe default state in case of
system failure or the identification of malicious
intervention. In this respect the best system state could be:
“fail operational”, “fail soft”, “fail safe”, “fail secure”, “fail
passive”, “be fault tolerant”…
Project Output:
Integrated safety & security system, which leads to HAZOP

Human factor

Organizational & legal environment Integration with environment

(IMO flag states, classification (COLREG, weather…)
societies …)

Shore-based
Control System/
Engine (AS) Bridge (AS) Port
Autonomous
Engine
Monitoring &
Controlling
systems

Engine Data Smart Blockchain

Logger containers technology

Adopted from: G. Kavallieratos, S. Katsikas, V. Gkioulos, “Cyber-Attacks Against the Autonomous Ship,” In: Katsikas, S., et al. Computer Security.
SECPRE CyberICPS 2018. Lecture Notes in Computer Science, Vol. 11387. Springer, Cham. https://doi.org/10.1007/978-3-030-12786-2
 System Theoretic Process Analysis

Autonomous vessel's system safety control structure

Overall Expected Outcome - 1

• The call asks for TRL 5-6 at the end which is rather high
going much beyond simulations.
• This requires connection to a ship builder (DeltaMarin and its
subcontractors?) and their latest vessel designs.
• We need (from the Application side):
1. Vessel system components functional and „extra“-functional
(performance, resiliency, security) requirements.
2. Realistic Data through a Digital Twin.
3. Other ?
• We provide:
1. A complete ”reliability” framework integrating state-of-the-art
Safety Analysis/Threat Modelling/Risk Assessment methods
with tool-chain support for Validation (Modeling, Simulation,
Verification) and Execution (Online
Testing/Monitoring/Detection) and Mitigation.
1. With AI extensions where needed.
Overall Expected Outcome - 2
• As in Avionics and Automotive domain, we will deal with Mixed-
Critical Systems.
• The Safety-Critical subsystems must always be isolated from the
non-safety-critical subsystems.
• Existing ”Isolation” techniques should be revisited and possibly
improved.
• Due to the safety-criticality levels, any AI method used/developed
should be ”Explainable” by construction, to be able to reason about
decisions and also allow ”tracing”.
• Advancing methods for (AI or non-AI-based) ”Prediction” of critical
states will be crucial.
• Works exist for isolated Cyber Security Threat Modeling/Risk
Assessment or isolated Safety Analasis/Management but not much
exists in integrating these (this is driving also the Call description).
• Also, all these (24 papers) works do not have much connection to AI.
• Also, all are validated through simulations only – Critical corner cases
cannot be covered.