Detection and Classification Technologies for Large Area Clearance
Abstract
In response to ONR BAA Announcement #N00014-22-S-B001, specific research effort on the Large Area Clearance Experiment (LACEx), we p,eiver coils and integrate it on a unmanned ground vehicle (UGV); 2) to adapt advanced EMI models for processing dynamic data sets p,rovided by the time domain EMI sensors mounted on the UGV; 3) to develop a unified, fast, robust and physics based forward and inver,se Magnetic and EMI models to process EMI/Magnetometer data jointly; 4) to demonstrate the completely integrated (hardware-software,) systems detection and classification performance at test sites to fully characterize subsurface targets from the time domain EMI, and magnetic sensors data sets collected on the UGV in dynamic mode. Airfields are the most immediate and lucrative targets for an, the infrastructure that supports the aircraft, basically means destroying the entire air operations. Recent wars in in Iraq, Afghan,istan and other locations have demonstrated that one of most challenges problems to resume airfields operations after attacks is sub,surface unexploded munitions (UM). During attacks, although majority of munitions explode and produce shrapnel on ground surface and, subsurface, some munitions penetrate in soil up to 20 feet without detonation. Since, most if not all UM are whole metallic or cont,aining substantial amounts of metals, they are easily detectable with metal detectors. However, it is difficult to distinguish UM, p,articularly deep buried, from the widespread and closely spaced shrapnel. Current deep subsurface UM detection, locating and identif,ication techniques rely on magnetometer whose sensing capabilities are limited due to undesirable signals produced by widespread met,allic clutter. New sensors, techniques, and procedures are required to detect, locate, and identify subsurface UM with high confiden,ce and to separate signals from the deep subsurface UM and from shrapnel (surface and intermediate subsurface).The LACEx programs o,bjective is to develop fast, robust, and cost-effective large area clearance technologies for mapping and identifying explosive haza,rds at airfields for all the categories of mission scenarios. Over past 21 years our team has designed, built, and demonstrated adva,nced time domain electromagnetic induction systems for subsurface metallic anomalies detection and classification, among them handhe,ld ultralightweight,tor sensor. These sensors, when combined with our in house developed advanced forward and inverse EMI models, such as the orthonorma,lized volume magnetic source, joint diagonalization, and differential evolution, have demonstrated excellent classification performa,nce in DoD-administered blind classification tests conducted at increasingly challenging UXO sites. Recent live site dynamic (contin,uously moving sensors) classification data analysis from various UXO sites showed that these advanced models provide the same or eve,n better subsurface targets detection and classification capabilities, than the standard, three step: 1. Dynamic data collection and, targets selection; 2. Cued (static) data collection and data processing; 3. Classification feature parameter selection and target i,dentification, approach.At the end of the project, we aim to deliver advanced EMI-magnetic signal processing algorithms and a lightw,eight EMI sensor array, with four transmitter and four vector receiver coils based on time domain electromagnetic induction, mountab,le on a UGV and interfacing with robot control software, as well EMI and magnetic data processing algorithms and modules.Approved fo,r Public Release
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 16, 2022
- Source ID
- N000142212396
Entities
People
- Fridon Shubitidze
Organizations
- Board of Trustees of Dartmouth College
- Office of Naval Research
- United States Navy