Simultaneous localization and mapping robot building with Android platform integration

OData support
Supervisor:
Dr. Csorba Kristóf
Department of Automation and Applied Informatics

Mobile robotics is one of the most interesting research fields of technology in the world of science today. It is fascinated by scientists, mathematicians, and engineers alike. And due to its seemingly unlimited applications and extreme challenges, mobile robotics is a major focus of many research institutes. With the help of advanced technology, scientists have successfully overcome many challenges related to mobile robotics. On the other hand,

there are still a lot of problems remain unsolved and are waiting for adequate solutions.

Navigation is a notable branch of mobile robotics research, and it is considered as one of the most natural applications due to mobile robots’ particular characteristics. Indeed, there have been many papers on robots’ navigation and a large portion of them are dedicated for Simultaneous Localization and Mapping (SLAM) technique. Originally developed by Hugh Durrant-Whyte and John J. Leonard, SLAM is considered the most suitable solution for autonomous robots to map an unknown environment while at the same time keeping track of their current location. Even though SLAM itself has been a subject for intensive research, only recently  from 2005  have SLAM algorithms based on cameras (or Visual SLAM) become prominent due to the development of mobile cameras. Because of the richness of data and the limit of bearing-only information it provides, Visual SLAM became an interesting and challenging new direction in robots’ navigation.

The main purpose of the thesis is to build a robot that is capable of implementing a suitable SLAM algorithm to make a map of a previously unknown environment. In order to implement Visual SLAM successfully, we equip our robot with adequate proprioceptive and exteroceptive sensors. Furthermore, we also apply computer vision’s techniques for processing images, extracting images’ features and executing complicated operations such as features’ detection and loop-closure for mapping. Finally, in order to increase its mobility and to explore new possibilities, we integrate some functions for improving the robotás operation using an Android smartphone.

The following tasks will be considered:

• Building an autonomous land robot using suitable microcontrollers and sensors.

• Establishing a wireless communication between the robot and a computer.

• Building a map of an unknown environment by applying Visual SLAM technique.

• Exploring the possibilities of an Android smartphone for the project.

Downloads

Please sign in to download the files of this thesis.