Lidar-based object classification and tracking with TurtleBot3

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

According to the 4th Industrial Revolution, unmanned cars are gradually used by many people in many fields. Autonomous car technology requires several sensors, such as LIDAR and cameras, and actuators which must work harmoniously. To achieve higher accuracy of processing with high speed, we need to focus not only on developing accurate technology, but also pay attention to the execution speed of every processes.

In this thesis, I will implement simulation with Turtlebot 3 (“burger” edition) in a virtual environment in “seek-and-hide” mode. In the given field, there are several obstacles located randomly and a moving target to track. The mobile platform and its environment are designed and modelled in Matlab environment which also serves to execute and test signal processing and control algorithm.

Without the use of machine learning techniques, we aim to extract a feature efficiently and create an algorithm accordingly. The method is tested in a simple environment where features of objects are easy to detect using a LIDAR sensor.

Downloads

Please sign in to download the files of this thesis.