Simula Research Laboratory recently contributed to IEEE IV 2025 with two papers presented during the AI4CCAM-sponsored workshop entitled “Advancing AD in Highly Interactive Scenarios through Behaviour Prediction, Trustworthy AI, and Remote Operations“, which were also accepted for presentation as posters in the main Conference. These papers both relate to scene understanding using qualitative spatio-temporal representations:
Automatic Cause Determination in Road Scene Understanding Using Qualitative Reasoning and Four-Valued Logic
Road scene understanding in automated driving (AD) aims to build a comprehensive analysis of video sequences taken on the road by embedded or fixed cameras (e.g., mounted on vertical road signals). One goal is to identify the relevant actors in the scene and another goal is to determine the causes that have triggered a specific action of the ego car (i.e., stop, slow down, turn left, etc.). In a complex urban environment, these causes can be multiple, confusing, possibly contradictory to other causes and not easily expressible using simplistic reasoning. Still, providing accurate automatic cause determination supports a) user acceptance by providing appropriate explanations to the car passengers and road users; b) increased road safety by providing detailed road scene understanding to traffic. In this paper, AI4CCAM proposes using spatiotemporal reasoning and Belnap’s four-valued logic to formulate complex causes of AD action in a road scene.
Explainable Scene Understanding with Qualitative Representations and Graph Neural Networks
This paper investigates the integration of graph neural networks (GNNs) with Qualitative Explainable Graphs (QXGs) for scene understanding in automated driving. Scene understanding is the basis for any further reactive or proactive decision-making. Scene understanding and related reasoning is inherently an explanation task: why is another traffic participant doing something, what or who caused their actions? While previous work demonstrated QXGs’ effectiveness using shallow machine learning models, these approaches were limited to analysing single relation chains between object pairs, disregarding the broader scene context. AI4CCAM proposes a novel GNN architecture that processes entire graph structures to identify relevant objects in traffic scenes.
AI4CCAM held its 5th General Assembly in Barcelona, Spain, 8 and 9 July, hosted by the project partner
Barcelona Supercomputing Center (BSC)
Two intense days with a packed agenda! But AI4CCAM is coming to an end, and it’s important to take stock and prepare for the final event.
The meeting was opened by Simula Research Laboratory, the project coordinator, and IRTSX, technical coordinator, with a project roadmap showing milestones and results achieved so far and the next steps.
So plenty of space to Uses Cases and technical updates, as well as to communication and financial aspects.
The partners had the opportunity to discuss in depth the organization of the project’s final event, which will be held on November 17th and 18th in Brussels.
A visit to MareNostrum 5 was also organized by BSC: a pre-exascale EuroHPC supercomputer hosted at BSC-CNS. The system is supplied by Bull SAS combining Bull Sequana XH3000 and Lenovo ThinkSystem architectures and it has a total peak computational power of 314PFlops. The system will provide 4 partitions with different technical characteristics that jointly can fulfil the requirements of any HPC user.
News and updates from the AI4CCAM Participatory Space!
The AI4CCAM Participatory Space is working on the creation of a glossary of capital and common terms used in the project.
This participatory process is aimed to create a glossary of terms with the involvement of different CCAM stakeholders. The process consists of proposals of terms to be included along their definitions. Later on, discussions about the correctness of these definitions take place and a final survey to decide the best option is performed. Results are monitored and agreed definitions are incorporated to a document that represents the glossary.
A new set of definitions for the glossary regarding Ethics and Governance was released and participants will be asked to share feedback on the correctness and understandability of the proposed definitions.
A second update is about the publication of a user acceptance survey!
Do you work in transportation, cybersecurity, AI, automotive, or are just concerned about how current innovation shapes future forms of mobility?
We’d be glad to read you out, if you would like to take 8 minutes to participate to our quick survey about road user acceptance of automated mobility.
Your answers will help AI4CCAM better understand different views and perception of this topic, and you might learn something interesting along the way…
The IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV 2025) is the premier annual forum organized by the IEEE Intelligent Transportation Systems Society (ITSS). IV 2025 took place in Cluj-Napoca, Romania, 22-25 June, and AI4CCAM was there!
Researchers, academics, and practitioners from universities, industry, and government agencies shared their latest research papers, simulation challenges, and applications on Intelligent Vehicles and Intelligent Vehicle Infrastructures. The symposium featured Plenary Talks, Technical Sessions, Poster Sessions, Tutorials, Workshops, Exhibitions, and Industrial Demonstrations and Challenges.
Simula Research Laboratory, the AI4CCAM coordinator, contributed to IEEE IV 2025 with two papers presented during the AI4CCAM-sponsored workshop entitled “Advancing AD in Highly Interactive Scenarios through Behaviour Prediction, Trustworthy AI, and Remote Operations”, which were also accepted for presentation as posters in the main Conference. These papers both relate to scene understanding using qualitative spatio-temporal representations.
From 29th June to 3rd July, Athens hosts the International Symposium on “Navigating the Future of Traffic Management”, organized locally by the Institute of Communication and Computer Systems (ICCS) of the National Technical University of Athens (NTUA) in collaboration with leading global and European organisations at the forefront of traffic management research and operations.
The Symposium aims to explore and highlight innovative solutions for dynamic traffic management, strategies to improve traffic flow with a focus on safety, sustainable and environmentally conscious practices, as well as future research directions and needs in the field.
AI4CCAM will be there!
INLECOM will participate in the session “Navigating the Challenges of Data Management and TMS Assets” with Konstantinos Loupos, sharing valuable insights on “Artificial Intelligence and Digital Solutions for Transport Infrastructure Resilience” and highlighting the methodologies developed within AI4CCAM.
AI4CCAM will also be at the Mini Exhibition taking place: Tuesday, 1 July & Wednesday, 2 July 2025. Visit us to find out more about the project!
Bringing together more than 100 international speakers and representatives from leading institutions in transport, technology, and public policy, the symposium will cover five key tracks shaping modern transportation management. The Symposium will spotlight innovative solutions to some of the most persistent urban challenges — including traffic congestion, road safety, environmental sustainability, and the integration of smart infrastructure. Topics will include intelligent traffic signal systems, dynamic toll pricing, real-time traffic monitoring, and integrated traffic control platforms.
For more information on the event, click here
36th IEEE INTELLIGENT VEHICLES SYMPOSIUM will be held June 22-25, Cluj-Napoca, Romania, featuring Plenary Talks, Technical Sessions, Poster Sessions, Tutorials, Workshops, Exhibitions, and Industrial Demonstrations and Challenges.
AI4CCAM is the sponsor and one of the organizers of the Workshop “Advancing Automated Driving in Highly Interactive Scenarios through Behavior Prediction, Trustworthy AI, and Remote Operations” on 22 June.
Automated driving is being developed and deployed in complex road traffic environments with dense interactions. Ensuring safe and comfortable operation requires adaptable solutions — from reliance solely on onboard sensors and computing to integration with intelligent infrastructure, connectivity, and remote operation. As vehicles become increasingly intelligent, maintaining their efficiency and safety throughout their operational lifetime is essential for preventing accidents and fostering public trust.
This workshop explores three critical topics:
- Multi-road user interactions and behavior prediction/planning
- AI trustworthiness and CCAM user perception
- Remote operation of intelligent connected and automated vehicles
This full-day workshop (8.30 am to 18.15 pm) includes three sessions focused on these topics, complemented by two interactive poster sessions. Each session will open with a keynote speaker, followed by workshop paper presentations. The workshop then concludes with a round table discussion featuring experts from diverse backgrounds, ensuring comprehensive coverage of all main topics.
Firstly, technological and human factor challenges on remote operation covering three aspects—monitoring, assistance, and driving—will be discussed.
Then the discussion will be focused on aligning AI-based systems’ trustworthiness across technical, ethical, and social dimensions. It will explore how to develop CCAM systems that are not only technically robust and reliable but also perceived as trustworthy by users. This includes addressing challenges in human-machine interaction, transparency, and ethical considerations.
The last session presents work examining the balance between safety and efficient behavior in automated vehicles, discussing approaches to predict and plan comfortable trajectories that meet human expectations while maintaining safety. The workshop will focus on multi-modal approaches and their evaluation, considering the diversity of valid solutions of a traffic situation and facing unknown intentions of others.
By combining these topics, the workshop provides a comprehensive view of automated driving challenges, emphasising the crucial roles of remote operation supported by humans, and AI in developing trustworthy, efficient, and safe automated mobility solutions.
AI4CCAM will especially be represented in Session 2 on “Aspects of trustworthy AI for CCAM” and Session 3 on “Multi-stakeholder road interactions“.
AI4CCAM is reaching its end and a final event is planned on 17 and 18 November, in Brussels, Belgium.
Through collaborative research, AI4CCAM contributed to the CCAM ecosystem by addressing a very specific issue related to trustworthy AI and automated driving.
The event in Brussels will be the right place to be to discover the demonstrations developed within the project that highlight key innovations in the use of artificial intelligence (AI) for safer, more ethical automated mobility, and also reflecting the project’s multidisciplinary approach and commitment to open-source tools, scalable validation environments, and human-machine interaction research.
Stay tuned, more information will follow soon!
On 28 May, the ULTIMO project is organising the 5th edition of its webinars series.
Entitled “Advancing Safety in Automated Mobility”, the webinar (14.00 to 15.30 CET) aims at delving into the latest advancements and challenges in ensuring the safety of automated vehicles, and AI4CCAM will be represented by the project coordinator, Arnaud Gotlieb, Simula Research, discussing “Improving safety with trustworthy Artificial intelligence for CCAM“.
Along with AI4CCAM, leading experts from around the globe will delve into the latest advancements and challenges in ensuring the safety of automated vehicles. A range of critical topics will be covered, including:
- Perception Systems for Vulnerable Road Users (VRUs): Exploring how advanced perception systems detect VRUs’ attention, intention, and mobility aids, and their integration with external Human-Machine Interface (HMI) systems for effective communication.
- Vehicle-to-Everything (V2X) Communication: Discussing the role of V2X technologies in enhancing road safety through real-time data sharing between vehicles and infrastructure.
- Safety of the Intended Functionality (SOTIF): Evaluating the safety of automated driving systems and leveraging naturalistic driving data for comprehensive safety assessments.
- Regulatory and Policy Frameworks: Reviewing international standards for automated and connected vehicles and the importance of collaboration between industry, academia, and policymakers.
- Artificial Intelligence in Automated Driving: Highlighting the role of AI in enhancing vehicle perception and decision-making, and the use of machine learning techniques to improve ADS functionality.
Moderated by Jorge Pinazo Donoso (Automotive Technical Innovation & Development Manager, Capgemini), the webinar will provide a comprehensive overview of the current state and future prospects of automated mobility safety.
ADAS & Autonomous Vehicle Technology Expo and Conference returns to Stuttgart, Germany, Europe’s most influential hub for automotive innovation, research and development on May 20, 21 & 22, 2025, and AI4CCAM will be there!
Exhibitors will showcase the latest technologies to enable and accelerate end-to-end autonomous and ADAS applications, including testing tools, simulation, software, sensing and AI.
You can also benefit by networking with international delegates and speakers from the adjoining Automotive Testing Expo Europe.
Elem Güzel, Virtual Vehicle Research, on 22 May (13.10-13.35 – Room A W2/3) will be holding a speech on “Integrating VR and CARLA for realistic vulnerable road user simulations“:
Driving simulators often struggle to accurately simulate vulnerable road users (VRUs) in safety-critical scenarios. We present a framework integrating immersive virtual reality and human motion capture with the CARLA driving simulator. This setup allows simulations where real human agents act as VRUs interacting with an AI-equipped ego car featuring advanced ADAS functions. It enables testing and validation of new ADAS features with real pedestrian behaviors, providing valuable data on VRU reactions in dynamic traffic environments. This approach enhances the realism of VRU simulations and supports the development and evaluation of safer, more effective ADAS technologies for real-world applications. The speech will be sharing with participants:
- An overview of SOTA driving simulators and their limitations, covering the behavior of vulnerable road users in dynamic traffic environments
- The capabilities of the framework that integrates the CARLA simulator with the VR Meta Quest Pro headset, involving real human agents as pedestrians
- The application domains and the use cases of the proposed VR+CARLA simulation framework.
For more information on the event, click here
