Aims of the workshop
The domain of traffic and transportation is geographically and functionally distributed; its subsystems have a high degree of autonomy, yet are highly dependent on each other. Typically we need to deal with settings characterised by a variety of dynamics including real-time constraints and conflicting goals. More and more intelligent algorithms are used and still needed for coping with the complexity of participating in traffic, managing transportation, etc. On the other hand, there is already a wealth of relevant data about different aspects of traffic and transportation systems available that support innovative intelligent approaches. Consequently, many applications in this domain can be adequately modelled using intelligent, autonomous agents and multi-agent systems. The growing interest of the Artificial Intelligence community in the traffic and transportation domain meets the growing need of the traffic community for more secure, efficient, sustainable (i.e. resource-saving and ecological) transportation solutions.
From a scientific perspective, considering the recent trends and achieved results of deep learning approaches, it is important to investigate to which extent these approaches can bring significant contributions to solving open issues in the application area, so contributions on machine learning, deep learning, and data-driven approaches in general to problems in the traffic and transportation sector are particularly welcome.
From the application perspective, original and novel challenges to researchers in the area are formulated at an ever growing rate: autonomous vehicles, human driver assistance systems (which often call for a proper modeling of drivers’ and pedestrians’ behaviours for appropriately testing), innovative cooperative systems employing V2V communication in connected vehicles or with smart city infrastructures, represent additional topics of interest for the workshop in addition to more traditional topics, as for example related to agent-based traffic simulation or traffic control, etc.
In short, the purpose of this workshop is to bring researchers and practitioners together in order to share their views on how large-scale complex transportation systems can be modelled, simulated, controlled and managed - both at micro and macro level - employing and combining AI-based techniques, models, approaches, under the umbrella of autonomous agents and multiagent systems.
Topics of interest
- Applications of AI technology in traffic, transportation, and transport logistics
- Intelligent Optimization (e.g., traffic assignment, routing, route choice)
- Autonomic and autonomous transportation systems
- Coordination in intelligent transportation systems
- Intelligent, adaptive traffic control
- Distributed decision making in traffic, transportation and transport logistics
- Intelligent vehicles, intelligent assistance systems and human involvement
- Mobile devices in smart transportation systems
- Intelligent monitoring of transportation systems, data collection, filtering, prediction and distribution of traffic information and transportation data
- Autonomous and connected vehicles, collaborative driving
- Self-* properties and theory of intelligent traffic and transportation systems
- Agent-based simulation of traffic and transportation systems and their behavior, also based on agent-based and/or cognitive approaches to modelling traffic participants
- Agent-based pedestrian and crowd simulation
- Data driven approaches in the domain of traffic and transportation.
- Deep learning architectures for temporal and spatial traffic data
- Verification and validation of intelligent transportation systems
- Future technologies, smart transportation