Motion planning using the Safety Velocity Obstacles method

OData support
Supervisor:
Gincsainé Dr. Szádeczky-Kardoss Emese
Department of Control Engineering and Information Technology

The main task of a mobile robot is to reach the goal from a start position. As a main criterion, we always need to guarantee a collision-free motion among the obstacles, while the robot is moving from the start to the target position. This way, we can ensure the safety of the robot and the environment. We can use the motion planning algorithms to plan this movement.

The main purpose of my work was to develop and implement a motion planning algorithm that can ensure the avoidance of static and moving obstacles in the workspace of the robot. I used the Velocity Obstacles method, which is suitable for motion planning of mobile robots in a dynamic environment.

With this method, I planned evasive maneuvers by using information about actual velocity and actual position of the robot and obstacles which information are supposed to be known. To choose the velocity of the robot, in the most former publications, the authors always tried to find the velocity, that is the fastest solution to reach the target position. While following such a path, the robot usually goes very close to the obstacles which can result a collision-risk, if only inaccurate size, position and velocity information are available about the obstacles.

In my work, next to the fastest target reaching method, I introduced a new method, that can ensure the safest motion planning. I named this method Safety Velocity Obstacles (SVO) method. Next to SVO, I introduced a motion planning method using rules and I implemented the Artificial Potential Field method, also have compared this with the Safety Velocity Obstacles (SVO) method.

I always tested the result of the robot’s motion planning by simulations and I made videos of the motion of the robot and the obstacles.

This motion planning algorithm is suitable for motion planning of autonomous vehicles, so in the future there are a lot of opportunities for further development.

Downloads

Please sign in to download the files of this thesis.