Development of vehicle control algorithms for safe maneuvering

OData support
Dr. Kiss Bálint
Department of Control Engineering and Information Technology

„The future of driving” [1]. This sentence gives the main idea of the EU project called HAVEit (Highly Automated Vehicles for Intelligent Transport), which is carried out by a consortium of multinational companies and research institutes. The aim of this consortium formed in 2008 is to secure public transportation by intelligent vehicles. The long-term vision is to develop fully automated vehicles for commercial purposes. The BME Department of Automobiles is concerned in this project too. Our task is to investigate the safety of the used algorithms. Although the orientation of the research topics running in HAVEit are known, we have to re-develop most of the tested algorithms. In my thesis, I designed modules which could be the basic of these research works. My aim was to develop algorithms which produce the trajectories to an overtaking maneuver and control the vehicle along the planned path. All of the codes are implemented in MATLAB – Simulink environment. The used control problems are well-known in the literature but them utilization in the studied applications needs unique solutions. The task is a complex problem. To manage it I had to partition the project. Firstly, I represent a method for the path planning algorithm, then I design control systems for the separated vehicle models. I used state-space theory (LQG) for the control of the lateral dynamics. The model of the longitudinal dynamics is simpler, so the velocity controller was designed in the frequency domain (PI). The presented components are not solving the general problems. These modules are specific solutions to the simplified tasks. The results of the simulations confirmed the theories but the elimination of the remained subtasks needs further special knowledge in the field of control theories and research works.


Please sign in to download the files of this thesis.