LGP Discrimination and Residual Risk Analysis on Standardized Test Sites - Camp Sibert

Abstract

The demonstration described in this report was conducted at the Former Camp Sibert, Alabama, under project ESTCP MM-0811 LGP Discrimination and Residual Risk Analysis at Camp Sibert. It was performed under the umbrella of the ESTCP Discrimination Study Pilot Program. The MM-0811 project demonstrates the application of the LGP Discrimination Process to the problem of UXO discrimination. At the Camp Sibert site the objective was to discriminate potentially hazardous 4.2 mortars from non-hazardous shrapnel, range and cultural debris. Digital Geophysical Mapping ( DGM ) was acquired by the ESTCP Program Office from a variety of sensor arrays. The LGP Discrimination Process begins with the DGM from a site suspected of containing UXO. It then extracts attributes from anomalous regions (targets) in the DGM, uses Linear Genetic Programming ( LGP ) and the extracted attributes to rank the targets in their order of likelihood of being UXO, and finally, applies statistical residual risk analysis to determine which of the ranked targets may be safely left in the ground as not- UXO. In this report, we describe the performance of the LGP Discrimination Process in three separate tracks. The tracks were different in the sensor data combinations and the techniques used for attribute extraction. The three tracks may be described as follows: 1. EM-only-track. The sensor set used was the MTADS EM61 Array ( EM61MTADS ). The attributes extracted were statistical attributes drawn from the DGM of that sensor; 2. Combined-track. The sensor sets used were the EM61MTADS and the MTADS Magnetometer Array ( MAGMTADS ). The attributes extracted were statistical attributes drawn from the DGM of both of these sensors; 3. Inversion-track. Under MM-1505, SAIC had previously extracted phenomenological features from the EM61MTADS and MAGMTADS sensors in the Camp Sibert DGM. The attributes used were those phenomenological features.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2010
Accession Number
ADA603044

Entities

People

  • Dean A. Keiswetter
  • Frank D. Francone
  • Larry M. Deschaine

Tags

Communities of Interest

  • Cyber
  • Human Systems
  • Materials and Manufacturing Processes
  • Sensors
  • Weapons Technologies

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Computer Programming
  • Data Acquisition
  • Data Analysis
  • Detection
  • Detectors
  • Global Positioning Systems
  • Inertial Measurement Units
  • Information Science
  • Jet Propulsion
  • Machine Learning
  • Munitions
  • Risk
  • Risk Analysis
  • Statistical Analysis
  • Unexploded Ammunition
  • Uxo Detection

Readers

  • Military/Explosive Ordnance Disposal (EOD) Technology
  • Regression Analysis.

Technology Areas

  • Biotechnology
  • Biotechnology - Bioremediation