Asus Xtion Pro Camera Performance in Constructing a 2D Map Using Hector SLAM Method
IRAQI JOURNAL OF COMPUTERS, COMMUNICATIONS, CONTROL AND SYSTEMS ENGINEERING,
2021, Volume 21, Issue 3, Pages 1-11
AbstractSimultaneously and location mapping (SLAM) is an important technique for achieving a full autonomous navigation system by constructing a 2D map of the surrounding environment. A performance of two popular distance sensor (Asus Xtion Pro Camera and 2D LiDAR) in building a map of the indoor environment using Hector SLAM is presented in this paper. Navigation system using 2D LiDAR can only detect object on a certain level of plane. This leads to miss the obstacles that are below and/or above the level of laser scan. So the generated map will be inaccurate that causes collision during autonomous navigation. Asus Xtion Pro sensor can be a low cost alternative for a laser distance sensor in addition to its ability to provide 3D data. Using data of depth image, the entire obstacle will be detected to prevent collision. Many experiments in real time scenarios in indoor environment have been conducted to evaluate the performance of the RGB-D sensor vs. 2D LiDAR in constructing a 2D map. Furthermore, the results also indicate that some modifications on the parameters of Hector SLAM method are able to enhance the accuracy of map which is constructed by Asus Xtion Pro camera. Therefore, Asus Xtion Pro offers a good alternative to build a 2D map using Hector SLAM. This work is implemented in ROS on Raspberry Pi 3 B+.
- Article View: 75
- PDF Download: 51