Among the several public deliverables recently released by AI4CCAM, D4.1 “AI4CCAM Validation Handbook” is included too.
AI4CCAM includes three use cases (UC):
- UC1: Validating AI-enhanced ADAS
- UC2: Enhancing VRU trajectory
- UC3: Validating User Acceptance of CAVs
The objective of this document is to define how AI4CCAM is going to validate these three use cases. To do so, AI4CCAM starts from the digital framework (developed in WP1) and the trustworthy framework (developed in WP3) and concretize these general guidelines into specific evaluations.
This handbook is precisely an attempt at translating ethical requirements all the way down to
algorithms. And the reason it is needed is to try to mitigate the potential negative impacts of the exponential AI advancement.
Visit our Library and download the document!
Among the several public deliverables recently released by AI4CCAM, “AI4CCAM Trustworthy AI Documentation Framework – Initial Version” is included too.
The document provides a documentation framework of how to address trust and adoption when it
comes to AI (Artificial Intelligence) in the scope of Connected, Cooperative and Automated Mobility
(CCAM). More specifically, guidelines are provided addressing the following topics as they relate to
trustworthy AI in the scope of the CCAM ecosystem:
– fairness & diversity,
– transparency-led user agency, communication, and acceptance
– governance techniques for accountability in AI-based CCAM operations with ethical dilemma
identification.
To address the challenges involved by the implementation of AI in a CCAM environment, the
guidelines suggest several steps to be taken and points to take under consideration, including one
central action that is overarching as a common thread across topics, which is the inclusion of citizens
in the discussions and debates revolving around AI in CCAM.
To improve the implementation of responsible practices, a common ground must be found in terms of
the definitions of responsibility, but also of fairness, diversity, user-agency, transparency and
accountability. To achieve this goal, AI4CCAM proposes to use the Joint Research Center, Confiance AI, and Connected Automated Driving glossary documentation as a base, not only for this specific document, but for the overall AI4CCAM project and to take into consideration the AI Act that was adopted on May 21st, 2024 (European Parliament, 2024).
To feed the guidelines and techniques outlined in this document, desk research was conducted on
ethics implications when it comes to AI in the mobility space, reflecting the ethical principles
highlighted by the High-Level Expert Group on AI, as well as on trends on CAV adoption worldwide.
Following this desk research, an iteration workshop was organised among all AI4CCAM consortium
members, ahead of a second public stakeholder forum held in Karlsruhe during the 2024 IT-Trans fair.
Finally qualitative face-to-face research was conducted among car drivers in Paris, Berlin, and Warsaw,
to explore levers and barriers to CAV adoption.
Visit our Library and download the document!
AI4CCAM has recently released several public deliverables related to project activities, progresses and results. Among these, the “Participatory AI4CCAM Space – Initial version”.
One of the goals of AI4CCAM (specifically under Work Package 3) is to create a space for cooperative learning, reflection and discussion using participatory methods that involve CCAM stakeholders and end-users. To achieve this, an online participatory space to create a community of CCAM stakeholders was created.
The purpose is to raise awareness and literacy on the ethical, legal, social, economic and cultural implications of using AI for CCAM and to collect feedback from the community regarding a set of topics of interest identified in the project.
The space is an open-source web-based platform, its the design and definition of its contents are strongly related to the research carried out by the projects partners and external CCAM stakeholders in different events.
Visit our Library and download the document!
AI4CCAM partners met on 20 and 21 June, in Mladá Boleslav for the fourth General Assembly of the project, hosted by Skoda.
A crowdy room to discuss project achievements and upcoming goals for each Work Package of AI4CCAM, especially considering that AI4CCAM is going to held soon its mid-term review, foreseen in September 2024.
The main focus of the meeting was on the project demonstration and scientific impact; the data management during and beyond project, both analysed and discussed thanks to specific breakout sessions.
A visit at Skoda premises was the best way to end an intense and productive General Assembly!
The AI4CCAM second newsletter is out!
Such a special issue as AI4CCAM recently turned one! Watch the video telling so many months of work in just 120 seconds.
The newsletter is opened by message from our project coordinator, Arnaud Gotlieb, Simula Research Laboratory, underlining the strong results the project has started to deliver. In particular, the first publications of the project have demonstrated the high level of Research and Innovation activities. The design of autonomous driving (AD) scenarios in urban environment presenting ethical risks, the analysis of barriers and levers for user acceptance of AD in Europe, the preparation of three complementary and ambitious use cases as well as the finalisation of a participatory space dedicated to EU-citizen demonstrate high engagement and commitment of all the partners in the consortium.
Also, watch the interviews to our Ethical and Scientific Advisory Board, and read the latest news and upcoming events.
The newsletter is the best way not to loose any news and update on AI4CCAM.
On May 15, 2024, the first AI4CCAM Stakeholders Forum Workshop was held as a side event at the IT-TRANS Conference in Karlsruhe, Germany. This workshop gathered experts from various organizations to delve into the ethical, perceptual, and cybersecurity dimensions of Trustworthy AI in Connected Cooperative and Automated Mobility (CCAM). The discussions focused on identifying key levers and barriers to CCAM adoption and provided valuable insights to guide the future of AI in automated vehicles. The event facilitated meaningful discussions and provided useful insights to guide the future of AI in automated vehicles. The outcomes of these discussions will contribute to the development of AI4CCAM guidelines for AI that emphasize diversity, inclusion, and cybersecurity.
First, AI4CCAM presented the European Trustworthy AI Framework, which is central to the efforts on Trustworthy AI for CCAM. This framework aims to ensure an adequate ethical and regulatory environment. The four principles for delivering Trustworthy AI are:
• Human autonomy: ensuring users can make informed and autonomous decisions.
• Prevention of harm: protecting against vulnerabilities and ensuring safe AI systems.
• Fairness: avoiding bias and discrimination.
• Explicability: making AI processes and decisions transparent and understandable.
All these principles are crucial for delivering Trustworthy AI in CCAM.
The framework also includes seven ethical key requirements that are needed to implement the principles. These requirements are:
• Human agency and oversight
• Technical robustness and safety
• Data governance and privacy
• Transparency
• Diversity, non-discrimination and fairness
• Societal and environmental well-being
• Accountability
The workshop consisted of roundtable discussions that explored various aspects of delivering Trustworthy AI in CCAM:
• Fairness, Diversity, and Inclusion: Participants explored the requirements for ensuring diversity, non-discrimination, and fairness in the context of connected and automated vehicles (CAVs). They examined implementation strategies focusing on system design and validation and analyzed trade-offs with other priorities like privacy and safety.
• Sharing Road Space with other types of vehicles, including non-CAVs: Participants discussed the integration of CAVs into public transport environments, addressing issues such as space sharing, potential increases in private CAV usage, and safety concerns related to passengers and driverless vehicles.
• Fostering Cybersecurity: Participants examined the significance of cybersecurity and privacy for CCAM, focusing on balancing these aspects from the users’ perspectives.
• Ensuring Transparency in Industry Communication: This group discussed governance techniques to ensure transparent and accurate communication about the capabilities of autonomous vehicles, aiming to avoid over-claiming and ensure clear definitions of roles and accountability.
The workshop saw active participation from all attendees, fostering collaborative discussions that deepened the understanding of the challenges and opportunities in CCAM. The format proved useful for leveraging the expertise of the CCAM community, enriching AI4CCAM’s efforts to develop guidelines for Trustworthy AI.
The workshop concluded with the presentation of the AI4CCAM Participatory Space, an online platform that aims to connect stakeholders from the CCAM environment. Informative content along with interactive components such as discussions or surveys will be used to engage participants to the adoption of Trustworthy AI literacy applied to CCAM.
AI4CCAM will actively participate in the ECCOMAS Congress 2024, happening 3-7 June in Lisbon, Portugal.
The 9th European Congress on Computational Methods in Applied Sciences and Engineering is a scientific organization founded in 1992, grouping together European associations with interests in the development and applications of computational methods in applied sciences and technology. The mission of ECCOMAS is to promote joint efforts of European universities, research institutes and industries which are active in the broader field of numerical methods and computer simulation in Engineering and Applied Sciences, to address critical societal and technological issues with particular emphasis on multidisciplinary applications and disseminate innovative research in the fields of ECCOMAS.
The main event organized by ECCOMAS is its biennial European Congress, providing a rendezvous for scientists and engineers from within Europe and around the globe. The main objective of these congresses is to provide a forum for presentation and discussion of state-of-the-art developments in scientific computing applied to engineering sciences. Equal emphasis is given to basic methodologies, scientific development and industrial applications. The ECCOMAS Congress includes invited lectures, invited Special Technological Sessions (STS), contributed papers from Academy and Industry and organized Minisymposia.
AI4CCAM will be presenting the paper “AI4CCAM: Trustworthy AI for Connected, Cooperative Automated Driving” within the session “Contributions of EU-funded projects managed by CINEA towards greener and digital transport”, by Arnaud Gotlieb, Simula Research Laboratory, the project coordinator.
For more details on the Session, click here
To complement the AI4CCAM consortium with a wider view and receive feedback and support on the project activities and developments, a long-term Ethical and Scientific Advisory Board (ESAB) has been created, including CCAM experts and operators, public administrations and authorities, policy-makers and regulators, big industrial players, and associations.
The ESAB is led by Prof. Jean-Gabriel Ganascia, Professor of Computer Science at Sorbonne University. It also involves: Margriet Van Schijndel-de Nooij, Program director Responsible Mobility bij Technische Universiteit Eindhoven; and Jerome Perrin, Recherche scientifique et développement industriel dans les domaines énergie, environnement, mobilité – Associations caritatives – Ethique et Théologie.
The scope of the Advisory Board is to maintain an accurate body of knowledge on AI in CCAM operations, enabling the continuous consideration and analysis of ethical aspects and also facilitating interactions with relevant associations and policy-makers.
AI4CCAM interviewed Jean-Gabriel Ganascia. Watch the video!
To complement the AI4CCAM consortium with a wider view and receive feedback and support on the project activities and developments, a long-term Ethical and Scientific Advisory Board (ESAB) has been created, including CCAM experts and operators, public administrations and authorities, policy-makers and regulators, big industrial players, and associations.
The ESAB is led by Prof. Jean-Gabriel Ganascia, Professor of Computer Science at Sorbonne University. It also involves: Margriet Van Schijndel-de Nooij, Program director Responsible Mobility bij Technische Universiteit Eindhoven; and Jerome Perrin, Recherche scientifique et développement industriel dans les domaines énergie, environnement, mobilité – Associations caritatives – Ethique et Théologie.
The scope of the Advisory Board is to maintain an accurate body of knowledge on AI in CCAM operations, enabling the continuous consideration and analysis of ethical aspects and also facilitating interactions with relevant associations and policy-makers.
AI4CCAM interviewed Margriet van Schijndel. Watch the video!
