Agent-based Computing for Intelligent Transport Systems

(See the Agents4ITS Demo Server for a live demonstration of our technology.)

Join us -- we have Ph.D. positions and student projects / thesis topics available!

Motivation

Intelligent transport and mobility have become the most important areas for the application of multiagent systems techniques. Variety of agent-based algorithms and techniques will be essential for delivering the vision of seamlessly integrated intelligent transport systems that will be capable of autonomously managing and optimizing their operation so as to fulfil diverse needs of many relevant stakeholders in convenient, economical and eco-friendly ways.

Research Areas

We explore the application of several agent-based computing and artificial intelligence techniques to mobility and transport problems.

Fully multimodal resource-aware journey planning

Fully multimodal journey plannerWe research and develop next-generation journey planning algorithms supporting the full spectrum of modern mobility services, combining individual and collective, fixed-schedule as well on-demand means of transport while taking into account individual user preferences and availability of transport services and resources. See the Journey Planner live demo.


Agent-based microsimulation modelling of multimodal transport systems

Running AgentPolis-based simulation modelWe study how behaviour of complex large-scale transport systems can be simulated bottom-up by modelling the behaviour and interactions of millions of individual entities—people and vehicles—in the system. The high-level of detail provided by data-driven agent-based models enables representing non-linear patterns and phenomena beyond traditional approaches, and allows the resulting transport models to answer a wider range of what-if questions, including the impact of infrastructure developments, adoption of new mobility and transport policies or changes in mobility services available. The simulation technology can also be used to explore future scenarios concerning next generation transport technolgoies, such as mobility-on-demand systems, electric mobility or autonomous cars. See AgentPolis for more details.

Multiagent mechanisms for next-generation flexible mobility services

Simulation testbed for flexible mobility servicesWe explore how cooperative as well as market-based mechanisms can be used to better coordinate the use of capacity-limited transport services and resources. Specifically, we explore negotiation and planning techniques for real-time ride sharing and auction mechanisms for the dynamic pricing and allocation of transport services. We also develop an open-source AgentPolis-based simulation testbed that facilitates analysis and evaluation of multi-agent coordination mechanisms for next-generation flexible, on-demand mobility services.


High-fidelity transport network accessibility analysis

Transport network access analyzerWe develop data- and computation-intensive methods for analysing how different urban/region areas are served by public transport, taking into account factors such as travel times, number of interchanges and service frequencies. See the Transport Network Analyser live demo.




Multicriteria urban cycleplanning

Cycloplanner screenshotWe develop multicriteria urban cycleplanners capable of taking into account a wide range of cycling-specific factors, including altitude profiles, road surface, turn frequency and quiteness. See the cycleplanner live demo.




Game-theoretic optimization of transport security measures

Fare inspection modelling and optimizationWe explore how game theory-inspired techniques can be used for more efficient protection of vulnerable transport systems and their users, including optimizing fare inspection on public transport networks.




Awards

  • Best exhibit in Cluster 2 "Smart and sustainable cities for 2020+" at the ICT2013 conference - awarded to SUPERHUB (including our journey planner) in November 2013 
  • Otakar Motejl fund award (2nd prize) for best applications using open data - awarded to our Transport Analyser in November 2013

Funding

Our research on agent-based computing in transport systems is undertaken as part of several large collaborative research projects, including the European Union FP7 projects SUPERHUB (grant agreement no. 289067) and MyWay (grant agreement no. 609023), and the Technology Agency of the Czech Republic RODOS competence center.

SUPERHUB project logoRODOS - Rozvoj Dopravnich Systemu LogoMyWay project logo

Team

Michal Jakob (group leader, contact person), Jan Hrnčíř, Michal Čertický, Qing Song, Malcolm Egan, Marek Cuchý, Jan Nykl and Pavol Žilecký.

Contact us if you would like to work with us -- we have Ph.D. positions and student projects / thesis topics available. 

Publications

Papers

  • Michal Certicky, Michal Jakob, Radek Pibil, Zbynek Moler: Agent-based Simulation Testbed for On-demand Transport Services (Demonstration). Proceedings of the 13th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2014). 2014.
    BiBTeX | PDF (432)
  • Michal Certicky, Michal Jakob, Radek Pibil, Zbynek Moler: Agent-based Simulation Testbed for On-demand Mobility Services. In Proceedings of the 3rd International Workshop on Agent-based Mobility, Traffic and Transportation Models, Methodologies and Applications (ABMTRANS). 2014.
    BiBTeX | PDF (437)
  • Qing Song, Pavol Zilecky, Michal Jakob, and Jan Hrncir: Exploring Pareto Routes in Multi-Criteria Urban Bicycle Routing. In IEEE Intelligent Transportation Systems Conference. 2014.
    BiBTeX | PDF (442)
  • Jan Hrncir and Qing Song and Pavol Zilecky and Marcel Nemet and Michal Jakob: Bicycle Route Planning with Route Choice Preferences. In Prestigious Applications of Artificial Intelligence (PAIS). 2014.
    BiBTeX | PDF (439)
  • Michal Jakob, Jan Hrncir, Luis Oliva, Francesco Ronzano, Pavol Zilecky, and Jason Finnegan: Personalized Fully Multimodal Journey Planner. In Prestigious Applications of Intelligent Systems (PAIS). 2014, p. 1225 - 1226.
    BiBTeX | PDF (440)
  • Jan Nykl, Michal Jakob and Jan Hrncir: Advanced Public Transport Network Analyser. In Prestigious Applications of Intelligent Systems (PAIS). 2014, p. 1229 - 1230.
    BiBTeX | PDF (441)
  • Malcolm Egan and Michal Jakob: A Profit-Aware Negotiation Mechanism for On-Demand Transport Services. In European Conference on Artificial Intelligence (ECAI). 2014, p. 273 - 278.
    BiBTeX | PDF (444)
  • Michal Jakob and Zbynek Moler: Modular Framework for Simulation Modelling of Interaction-Rich Transport Systems. In Proceedings of the 16th IEEE Intelligent Transportation Systems Conference (ITSC 2013). 2013.
    BiBTeX | PDF (427)
  • Hrncir, Jan and Jakob, Michal: Generalised Time-Dependent Graphs for Fully Multimodal Journey Planning. In Proceedings of IEEE Intelligent Transportation Systems Conference (ITSC). 2013, p. 2138--2145.
    BiBTeX | PDF (426)
  • Hrncir, Jan and Rovatsos, Michael and Jakob, Michal: Ridesharing on Timetabled Transport Services: A Multiagent Planning Approach (preprint). Special issue of Journal of Intelligent Transportation Systems: Technology, Planning, and Operations. 2013.
    BiBTeX | PDF (428)
  • Hrncir, Jan and Rovatsos, Michael: Applying Strategic Multiagent Planning to Real-World Travel Sharing Problems. In 7th International Workshop on Agents in Traffic and Transportation, AAMAS. 2012.
    BiBTeX | PDF (381)
  • Michal Jakob, Zbynek Moler, Antonin Komenda, Zhengyu Yin, Albert Xin Jiang, Matthew P. Johnson, Michal Pechoucek and Milind Tambe: AgentPolis: Towards a Platform for Fully Agent-based Modeling of Multi-Modal Transportation (Demonstration). In 12th International Conference on Autonomous Agents and Multiagent Systems. 2012.
    BiBTeX | PDF (343)
  • Petr Kalina and Jiri Vokrinek: Improved Agent Based Algorithm for Vehicle Routing Problem with Time Windows using Efficient Search Diversification and Pruning Strategy. In Proceedings of the Third International Workshop on Artifitial Intelligence and Logistics. Lyon: CNRS-ENS, 2012, p. 13-18.
    BiBTeX | PDF (377)
  • Petr Kalina and Jiri Vokrinek: Parallel Solver for Vehicle Routing and Pickup and Delivery Problems with Time Windows Based on Agent Negotiation. In Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, SMC 2012. New York: IEEE - Systems, Man, and Cybernetics Society, 2012, p. 1-6.
    BiBTeX | PDF (378)
  • Petr Kalina and Jiri Vokrinek: Algorithm for Vehicle Routing Problem with Time Windows Based on Agent Negotiation. In Proceedings of the Seventh International Workshop on Agents in Traffic and Transportation (ATT) at AAMAS 2012. Geneve: European Office of Aerospace Research and Development, 2012, p. 1-10.
    BiBTeX | PDF (376)
  • Jiri Vokrinek and Antonin Komenda and Michal Pechoucek: Agents Towards Vehicle Routing Problems. In AAMAS 2010: Proceedings of the Ninth International Conference on Autonomous Agents and Multi-Agent Systems. Toronto, Canada: IFAAMAS: Internatioal Foundation for Autonomous Agents and Multiagent Systems, 2010, p. 773-780. ISBN 0-98265-710-0/978-0-9826571-1-9.
    BiBTeX | PDF (46)
Top