Research of methods and technologies for determining the position of the mobile object in space
The object of research is the process of tracking the position of a mobile object in space. One of the weakest points in tracking systems for the position of a mobile object in space is the problem of eliminating the ambiguity of determining key points when scanning the environment. This problem is...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PC Technology Center
2018-05-01
|
| Series: | Tehnologìčnij Audit ta Rezervi Virobnictva |
| Subjects: | |
| Online Access: | http://journals.uran.ua/tarp/article/view/147861 |
| Summary: | The object of research is the process of tracking the position of a mobile object in space. One of the weakest points in tracking systems for the position of a mobile object in space is the problem of eliminating the ambiguity of determining key points when scanning the environment. This problem is especially important when several methods (or technologies) of position tracking are applied simultaneously. There is a need for additional calibration and adjustment.
The study used the results of the analysis of methods and technologies for automatically determining the position and orientation of three-dimensional objects using technical vision systems. Analysis of the considered popular systems and methods for measuring the spatial position of objects, as well as algorithms and navigation technologies of a mobile robot, has shown that each of the considered systems has its advantages and disadvantages. And it is used depending on the objectives of this system.
A comparative analysis of the main types of algorithms of the SLAM method has been carried out. The perspectives of this method – the use of artificial intelligence methods and an extended Kalman filter – improve the speed of the SLAM method. Proof of this is the huge number of open projects to create this type of navigation in various competitions:
• VSLAM – implementation of the SLAM method based on computer vision methods;
• RGBDSLAM – package for registering a cloud of points with RGBD sensors, such as Kinect or stereo cameras;
• Hector_mapping – SLAM for platforms without odometer – only based on data from LIDAR, etc.
Since most modern technologies are increasingly using standardized formats of Wi-Fi, Bluetooth, GPS signals, it can be argued that using and analyzing information from a large number of sensors will increase the accuracy of determining the coordinates of an object several times. Creating the necessary information field of navigation and routing will allow to map and localize a mobile object on the ground with great accuracy |
|---|---|
| ISSN: | 2226-3780 2312-8372 |