MULTI-SENSOR UNDERGROUND
UTILITY LOCATION
Research project by
Lashminarayan Venkatesan, Graduate Student
(lvenkat@ece.ncsu.edu)
Dr. L.E. Bernold, Associate Professor, Dept. of Civil Engg.
Dr. G. Byrd, Assistant Professor, Dept. of Electrical and Computer Engg.
|
Overview/Background :
|
||
|
The excavation of earth, using heavy
equipment, is a necessary element of most construction related operations.
Although most excavations are completed relatively safely, there
are intrinsic dangers involved. One of the primary causes of these
dangers is the striking of underground utility lines during excavations
using heavy machinery. Deaths are only one of the results of striking
utility lines; power outages, potentially dangerous gas leaks, and
disruptions of communications (fiber optic cables, telephone cables)
are among the other problems. |
||
Problem Statement :
Defining the accurate location, direction, the type of
material, its size, hollowness, depth and other such characteristics of
the buried utility is a non-trivial problem because the complexity and
non-deterministic nature of the soil characteristics and the properties
of the buried utility lines. Human interface has to be simple and user-friendly
and rugged to survive the harsh environment of the construction site.
| Metal Antenna Detection System : | ||||
|
|
Buried Utility Detection System (BUDS) is a real-time that provides output to the machine operator as he/she digs. It is active detection system that generates and transmits its own magnetic fields and detect the coupling effect with any buried utility line within its detection range. A magnetic field is generated through the transmitter module and metal detector coil. Its impact on any metal object in its detection range will be coupled and picked up by the receiver module of the detector. The control unit processes the signals from the search coil. The data acquisition board is connected to an analog output port of the control unit via a simple cable. |
|||
Research Objective :
Creating a reliable and comprehensive database of sensor contours and their mappings to physical buried utilities with respect to various parameter such as depth of utility in the soil, distance of antenna from the soil, diameter of pipe, type of material and hollowness of pipe etc.
Developing a sensor fusion model to combine the sensor information of
the Metal Antenna sensor and the Ground Penetrating Radar to perform probability
estimation and combining sensor information.
Research the idea of sharing of property information between sensors to increase the probability of being correct about the conclusions of the buried utility.
Develop a model of a parallel processing machine and associated parallel
algorithms for an effective accurate on-the-fly utility detection system.
Research Methods :
Perform extensive experimentation for
developing a knowledge base for feature extraction from the radar pictures
and sensor contours obtained from the two sensor systems.
Use this database to develop a knowledge rule set that can estimate the
probability of the characteristics of the buried utilities.
Implement the above Artificial Intelligence Knowledge Base Model using
Parallel Processing and associated algorithms.
Integrate the concept of sensor networks in this model.