Gas source localization via behaviour based robot and weighted arithmetic mean
This research is concerned with the localization of gas source in a dynamic indoor environment problem using a single mobile robot system. Since the environment is unknown to the robot, an intelligent algorithm is required to enable the robot to traverse through the environment without any interrup...
Saved in:
| Format: | Thesis |
|---|---|
| Language: | English |
| Subjects: | |
| Online Access: | http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/78216/1/Page%201-24.pdf http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/78216/2/Full%20text.pdf http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/78216/4/Ahmad%20Shakaff.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This research is concerned with the localization of gas source in a dynamic indoor environment problem using a single mobile robot system. Since the environment is unknown to the robot, an intelligent algorithm is required to enable the robot to traverse
through the environment without any interruption from obstacles. All experiments were done on an experimental testbed consisting of a large gas sensor array (LGSA) to monitor real-time gas concentration within the testbed. The measurements from the LGSA were taken as the ground truth and were useful as it can be compared to the measurements taken
from the gas sensors on the mobile robot. A pattern tracking system was also utilized to
record the robot's odometry. Initially, two preliminary experiments were conducted to
better understand the conditions within the experimental testbed and the gas sensor's
performance within the environment when the robot is moving at different speeds. From the
preliminary experiments, we can confirm that the conditions within the testbed are indeed
dynamic and gas sensor's performance differs when the mobile robot is moving at different
speeds. We then proceed to implement two algorithms (i.e Zig-Zag and Braitenberg) to test
the robot's performance in traversing through the experimental testbed while taking gas
sensor measurements. The Braitenberg algorithm was separated into two variants (i.e Repel
and Attract) which then were implemented with the mobile robot. |
|---|