1.

The environment for processing the data from the simulator

The environment for data processing, statistics calculation and model building has been designed and set up in collaboration with the co-funding company. In addition, several methods for data verification, correction and processing have been developed. The environment for data processing is already actively used in practice by the co-funding company.

F.15 Development of a new information system/databases

2.

MOHDS

An algorithm for autonomous vehicle driving has been developed. The algorithm optimizes the traveling time and the fuel consumption, and works in real time. In addition, it takes into account other vehicles on the road. Moreover, the algorithm also includes drivers' behavior models, which enables to obtain drivings similar to human drivings. The algorithm is also connected with the professional driving simulator. Important achievements of the algorithm development within this project are the connection with the industry through the involvement of experts from the co-funding company in the algorithm design, and the transfer and future use of the algorithm in practice. The algorithm is publicly available at https://dis.ijs.si/erik/postdoc.html

F.08 Development and manufacture of a prototype

COBISS.SI-ID: 31187751

3.

Discovering human-like driving strategies with a multiobjective optimization algorithm

This report describes the implementation of the Multiobjective Optimization algorithm for discovering Human-like Driving Strategies (MOHDS). More precisely, it describes the source code of the algorithm, which consists of two main algorithms: the MOHDS algorithm and the SCANeR Autonomous Driving algorithm. The MOHDS algorithm searches for driving strategies, while the SCANeR Autonomous Driving algorithm integrates these strategies in the SCANeR Simulation Environment.

F.08 Development and manufacture of a prototype

COBISS.SI-ID: 31186727

4.

Multiobjective discovery of driving strategies for autonomous vehicles

The yearly survey of activities of the Faculty of Computer and Information Science at the University of Ljubljana includes, besides the description of regular work, also the description of the selected projects. For the year 2017, this project has been selected and described. The project description includes the project goals, the algorithm development, the main results and the funders.

F.35 Other

COBISS.SI-ID: 293534464