Autonomous Driving
Different levels of autonomy are the goal of autonomous vehicle research. The development deals with the programming of basic ADAS systems up to subsystems and units of control algorithms of autonomous vehicles. An integral part of control algorithms is AI programming and it has offshoots such as DNN, which are needed in various research areas of autonomous driving such as detection of objects, lanes, signs, intersections, trajectory planning, etc. In order to test control algorithms, computational vehicle models, which are then connected in a loop within different types of testing such as MIL - Model in the Loop, SIL - Software in the Loop, HIL - Hardware in the Loop, VIL Vehicle in the Loop or SCIL - Scenario-In-The -Loop. The research is also focused on the area of the virtual world, which is closely connected with testing, so that it is possible to simulate an autonomous vehicle with a detailed virtual world of transport infrastructure replacing real parts of cities, villages, highway sections in the Czech Republic, either from map data or from actual measurements. Scenario creation and V2X testing are also related to this. In order to be able to deploy the developed systems, which are computationally demanding from the point of view of using DNN, into experimental vehicles, special ECU hardware designed for autonomous driving is used.
Research results from this area are continuously used to modernize study plans in bachelor's, master's and doctoral studies.