The field of robotics has always been one of fascination right from the day of Terminator. Even though we still do not have robots that can actually replicate human action and intelligence, progress is being made in the right direction. Robotic applications range from defense to civilian, in public safety and fire fighting. With the increase in urban-warfare robot tracking inside buildings and in cities form a very important application. The numerous applications range from munitions tracking to replacing soldiers for reconnaissance information. Fire fighters use robots for survey of the affected area. Tracking robots has been limited to the local area under consideration. Decision making is inhibited due to limited local knowledge and approximations have to be made. An effective decision making would involve tracking the robot in earth co-ordinates such as latitude and longitude. GPS signal provides us sufficient and reliable data for such decision making. The main drawback of using GPS is that it is unavailable indoors and also there is signal attenuation outdoors. Indoor geolocation forms the basis of tracking robots inside buildings and other places where GPS signals are unavailable. Indoor geolocation has traditionally been the field of wireless networks using techniques such as low frequency RF signals and ultra-wideband antennas. In this thesis we propose a novel method for achieving geolocation and enable tracking. Geolocation and tracking are achieved by a combination of Gyroscope and encoders together referred to as the Inertial Navigation System (INS). Gyroscopes have been widely used in aerospace applications for stabilizing aircrafts. In our case we use gyroscope as means of determining the heading of the robot. Further, commands can be sent to the robot when it is off balance or off-track. Sensors are inherently error prone; hence the process of geolocation is complicated and limited by the imperfect mathematical modeling of input noise. We make use of Kalman Filter for processing erroneous sensor data, as it provides us a robust and stable algorithm. The error characteristics of the sensors are input to the Kalman Filter and filtered data is obtained. We have performed a large set of experiments, both indoors and outdoors to test the reliability of the system. In outdoors we have used the GPS signal to aid the INS measurements. When indoors we utilize the last known position and extrapolate to obtain the GPS co-ordinates.
| Identifer | oai:union.ndltd.org:ucf.edu/oai:stars.library.ucf.edu:etd-1377 |
| Date | 01 January 2005 |
| Creators | Ramamurthy, Mahesh |
| Publisher | STARS |
| Source Sets | University of Central Florida |
| Language | English |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Electronic Theses and Dissertations |
Page generated in 0.0027 seconds