Publication date: 
2018/12/21
The international team from the Faculty of Electrical Engineering at the Czech Technical University in Prague and Delft University of Technology won in the competition announced by Taxify and Robotex International 2018 in spring of this year. The task was to optimize the operation of the fleet of passenger service vehicles. The competition teams had to design a dispatching algorithm for a simulated fleet of autonomous taxi vehicles in the Estonian capital of Tallinn. A successful algorithm had to maximize reward for transported passengers, while respecting various constraints that simulate actual operating conditions. The winning solution was then awarded a prize of € 4,000.

Taxify's goal is to provide the public with efficient transportation across the city and ensure that the car is delivered to any place where Taxify operates within few minutes. In the future, it plans to use autonomous cars. The contestants had the task assuming that in 2022 this model of transportation would work in Tallinn for which they had to program algorithms for the efficient management of such a fleet.

The Artificial Intelligence Center (AIC) from the Faculty of Electrical Engineering of the Czech Technical University in Prague and the Autonomous Multi-Robots Lab (AMR) of Delft University of Technology have long been working on the analysis and optimization of transport systems. As scientist Martin Schaefer explains, "Participating in the competition allowed us to apply our expertise and experience using artificial intelligence to solve complex transport problems," had been the motivation for participation in the Taxified Self-Driving Fleet Optimization Challenge AIC.

"The assignment was really complicated, we had to work with different data sources, the demand data had to be linked to the map data from the road network and finally we had to deal with some difficult problems with the algorithm," adds Olga Kholkovskaia, a student of open informatics at the Faculty Electrical Engineering, CTU in Prague.

Successful solutions were based on effective assignment of passengers to vehicle plans. The winning team used techniques of mathematical optimization and artificial intelligence to find optimal plans for individual vehicles so as to serve as many passengers as possible with a minimum number of vehicles.
 
Contact person: 
Name: 
Ing. Libuše Petržílková
E-mail: 
libuse.petrzilkova@fel.cvut.cz
Phone: 
+420 731 077 387