Developement of a Path Planning Algorithm for Robotic Car

OData support
Csorvási Gábor
Department of Automation and Applied Informatics

The prerequisite for the effective usage of the autonomous vehicles in the future is the ability to navigate independently from any human control in a structured environment. For an autonomous vehicle is a key issue to find an optimal path from the start tot he goal, which can be travelled as fast as possible.

During my work, I learned the basic concepts of robot control, which are essential for understanding motion planning methods. After introducing the principles, I describe some path planning algorithms, which are widely used in the industry nowadays. These methods are the Reed-Shepp, Hybrid-State A*m RTR, RRT and C*CS algorithms, that are described in the first part of my dissertation. After having a comprehensive picture of the state-of-the-art path planning algorithms, such a method is presented in more depth

The OSEHS (Orientation-Aware Space Exploration Guided Heuristic Search) planner can be used with high performance for planning tasks in narrow spaces, such as a parking scenario. I studied the operation of the algorithm to gain a deeper understanding of the advantages and disadvantages of this method. The essence of this method is to gain space and orientation information from free space before starting the depth-first search in the workspace after the possible states.

My most important task was to implement the planner to a C++ software library, which was developed in the university. To be convinced of the correct functioning and effectiveness of the program, as a final step, I tested my code in different environments. The test results can be found at the end of my thesis.


Please sign in to download the files of this thesis.