EvoIR: Evolutionary Algorithms for Intelligent Route Generation for Emergency Vehicles in Urban Traffic (2016-2017)
This project deals with research in Evolutionary Computing applied to Control Systems Engineering, by using evolutionary algorithms to generate routes for emergency vehicles. The main challenge is providing responses in a continuously evolving environment within a prescribed time frame, while using limited resources and information that is often incomplete or uncertain.
When it comes to saving lives, short reaction time is crucial. Am emergency vehicle must arrive to its destination in as very little time as possible. Traffic in a busy city can almost double the transit time for a passanger car and cause significant delays for emergency services.
EvoIR aims at reducing the time spent in traffic by the emergency vehicle by providing its driver with a route planning system which is can be implemented in real-time and is optimized using intelligent evolutionary algorithms.
In order to achieve its objective, our project focuses on :
- development of cost functions for multi-criteria objectives
- development of specific route encoding systems as possible solutions of the problem, using heterogeneity based techniques within the chromosomes.
- design of evolution mechanisms specific to this encoding (selection, recombination, mutation)
- test and validation of the algorithms on urban traffic models.
This project has ended.
Project leader Monica Patrascu
Core Team Vlad Constantinescu, Andreea Ion, Răzvan Nitu
Associates & students Iuliana Bereș, Eugen Petre
This work has been funded by University Politehnica of Bucharest, through the “Excellence Research Grants” Program, UPB – GEX. Identifier: UPB–EXCELENȚĂ–2016 Evolutionary Algorithms for Intelligent Route Generation for Emergency Vehicles in Urban Traffic, 05/26.09.2016 (EvoIR).
- Constantinescu V., Patrascu M. 2017 – Route Encoding in Evolutionary Control Systems for Emergency Vehicles, 15th International Conference on Intelligent Transportation Systems Telecommunications ITST-2017, Warsaw, Poland, 10.1109/ITST.2017.7972216
- Patrascu M., Patrascu A., Beres I. 2017 – An Immune Evolution Mechanism for the Study of Stress Factors in Supervised and Controlled Systems, IEEE Conference on Evolving and Adaptive Intelligent Systems, EAIS 2017, Ljubljana, Slovenia, 10.1109/EAIS.2017.7954823
- Patrascu M., Ion A. 2017 – Self-adaptation in Genetic Algorithms for Control Engineering: the Case of Time Delay Systems, 21th International Conference on Control Systems and Computer Science, CSCS21, Bucharest, Romania, 10.1109/CSCS.2017.10
- Patrascu M., Constantinescu V., Ion A. 2016 – Controlling Emergency Vehicles in Urban Traffic with Genetic Algorithms, 9th Eurosim Congress on Modelling and Simulation, EUROSIM 2016, 12 – 16 September 2016, Oulu Finland