Reducing the Threat to Public Safety: Improved metallic object characterisation, location and detection

Lead Research Organisation: Swansea University
Department Name: College of Engineering

Abstract

The United Nations (UN) have announced that "With more than 30,000 foreign terrorist fighters from some 100 countries around the world, terrorism is a global threat requiring a comprehensive and unified response." This statement followed a spate of recent terrorist attacks, including those in France and Germany (July 2016), and a growing sense of global uncertainty in the western world. Promoting improvements to the identification and location of metallic threat items is an important aspect of the unified response and is an area where engineering and science can make a significant impact. Improvements in metal detection (MD) technology also provides wider benefits to the humanitarian cause of clearing landmines in developing countries. Wider benefits exist for the technology being transferred to MD companies developing devices for the non-destructive testing (NDT) of materials for safe structures, ensuring food safety, improved scrap metal sorting, as well as in medical imaging and archaeological searches.

Of course current metal detectors do find highly conductive objects and their simple design (and portability) has made them a highly cost effective modality for safety and security applications. Unfortunately, current technology is not able (or has limited capability) to distinguish between objects of different shape and materials of objects and can only detect objects within a small stand-off distance (or buried depth).

This proposal is aimed at overcoming these drawbacks through an interdisciplinary approach to improving MD technology, combing engineering, mathematics and scientific computation. Our hypothesis is that the response of metallic items in low frequency electromagnetic fields can be accurately described using a tensor based approximation. To test this hypothesis, we will develop a complete laboratory demonstration of our MD approach. This includes the following novel aspects: an efficient and adaptable software that can compute tensor coefficients for in-homogeneous objects, an algorithm for identifying different targets from field measurements with embedded uncertainty quantification as well as enhancing MD measurements by building new coil arrays based on optimised coil design. The goal is that our complete software and measurement package will lead to a step change in MD.

Planned Impact

The potential benefit of this research can be grouped as leading to societal, knowledge, people and economic Impacts and are closely interwoven with EPSRC's ambitions of achieving productive and resilient nations leading to UK prosperity. Societal and economic benefits stem from the improved identification of threat objects through the improvements to metal detection that we will develop. There has been a spate of recent terrorist attacks in France and in Germany and a result the UK threat level has remained as severe, implying that a terrorist attack is likely. Early identification of threat objects may help prevent future attacks.

The eradication of landmines in the developing countries also presents a huge challenge and the Find a Better Way (FABW) charity supported by charitable and government donations was established to aid the development of new technology to assist the clearance of the 1M active mines have clear societal benefits in the developing world as well as in areas of former conflict such as the Falkland Islands.

The UK generates 177M tonnes of waste each year much of which could potentially be recycled. The UK target is to recycle 76% of steel and 52% of aluminum in 2016 with targets set to grow in coming years. There are considerable economic benefits to be had from developing better scrap sorting processes based on improved metal characterisation that will result from our research and this can help reduce wasted resources (Typical recycled prices: aluminum £0.60/kg, stainless steel £0.90-0.75/kg, copper £3.10-3.60/kg).

In 2015, 1,514 food contamination incidents were reported by the Food Standards Authority in the UK. Many of these were associated with contamination resulting from bolts, nuts and wires displaced while food passing through the production line. Improvements in metal detection in the growing climate of mechanized food production will lead to both UK societal benefits (improved quality of life) and reduced potential legal action against food production companies (economic impact).

In this proposal we will work closely with our project partners (DSTL, Rapiscan and Safeline) and representatives from the Clearance of Landmines and Explosives (CIRCLE) and FABW to ensure rapid impact in all these areas.

Publications

10 25 50
 
Description Note that EP/R002134/1 and EP/R002134/2 relate to the same project, which was moved from Swansea University to Keele University in September 2020. Through this project there has been significant progress in the theory of metallic object characterisation as well as algorithms for locating and identifying hidden objects, which includes:

1) The development of an explicit formula for the characterisation of an isolated conducting permeable objects in terms of a rank 2 magnetic polarizability tensor description. The link to the multipole expansion used by electrical engineers has been explained and published in the electrical engineering community.

2) The development of a new complete asymptotic formula that characterises conducting permeable objects in terms of a new set of higher order generalised magnetic polarizability tensors, which has been published in the applied mathematics community.

3) The development of new asymptotic expansions that provide the leading order term for the perturbation magnetic field caused by the presence of inhomogeneous conducting permeable objects and multiple conducting permeable objects, extending the previous rank 2 magnetic polarizability tensor description to these cases. This has been published in the applied mathematics community.

4) A thorough investigation of how the magnetic polarizability tensor behaves with frequency (ie its spectral behaviour) has been undertaken and published in the applied mathematics community.

5) A MUSIC type algorithm for the location of multiple objects has been proposed and probabilistic machine learning dictionary based classification algorithm has been development for object identification.

6) Software has been released on GitHub for the computation of magnetic polarizability tensors for different objects, which has been accelerated by a reduced order model. This software has demonstrated the characterisation of realistic threat and non-threat objects such as guns, knives, coins and keys.

Together with EP/R002134/2 (which relates to the work on the same project after it was moved to Keele University), overall the project has led to 15 outputs including 8 journal publications. These developments have applications to improving metal detection for security screening, landmine searches and production line applications.
Exploitation Route We have invited our project partners (Rapiscan, Safeline and DSTL) as well as representatives from Sensor Technologies for Security Group (STSG) at The University of Manchester and the Find a Better Way (FABW) Charity to our meetings and we have shared our project deliverables. Knowledge impact is already been generated in the form of research reports, software, data sets and presentations. We have invited our project partners to engage in each of our workpackages (WP)s by asking for feedback on our developments, which will use ensure the practical applicability of our results. We look forward to testing our research developments against threat objects (or suitable surrogates or model pieces) that our project partners have agreed to provide. We also look forward to providing training to our project partners on the software that is being developed, which will lead to the development of new skills and people impact.

In the medium to long term, and with reference to EPSRC's productive and resilient nations, we would like to work with Rapiscan to embed our developments in to the design of its next generation MD technology for detecting threats public safety (R3: Develop better solutions to acute threats) as informed by DSTL. Furthermore, by working in partnership with STSG and FABW to develop deployable mechanism for accelerating landmine clearance (P2: Ensure affordable solutions for national needs). These developments will be realised through industrial secondments and applications for EPSRC impact acceleration funding. Our developments will have wider impact including contributing to a sustainable society, with a focus on the circular technology, and managing resources effectively and sustainably through improvements to metallic waste sorting applications with Safeline (R4: Manage resources efficiently and sustainably).
Sectors Aerospace, Defence and Marine,Manufacturing, including Industrial Biotechology,Security and Diplomacy,Transport

URL http://magpoltensor.wordpress.com
 
Description The agreement between the numerically computed polarization/polarizability tensors for multiple inhomogeneous conducting permeable objects and those experimentally obtained from field measurements by the group of Professor Peyton, The University of Manchester has been excellent. We have obtained new mathematical results that provide an explicit expression for the rank 2 magnetic polarizability tensor for multiple inhomogeneous objects and provided algorithms for the location and identification of conducting, permeable objects in metal detection based on MPTs. We have also obtained new mathematical results that allow conducting permeable objects to be more accurately characterised by higher order generalised magnetic polarizability tensors (GMPTs). We have obtained new mathematical results provide a detailed understanding of the spectral behaviour of MPTs. We have released software for the computation of MPTs. We have explained our findings to the engineering community through presentations at engineering conferences and publications of articles in engineering journals. Through our progress meetings we have engaged with our industrial collaborators to communicate our latest developments in improving metal detection. An impact event was held in 2021, which attracted 80 industrial and academic attendees.
First Year Of Impact 2018
Sector Aerospace, Defence and Marine,Manufacturing, including Industrial Biotechology,Security and Diplomacy,Transport
Impact Types Societal,Economic

 
Description Generalised Magnetic Polarizability Tensors: Invariants and Symmetry Groups
Amount £42,253 (GBP)
Funding ID EP/V049453/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 03/2021 
End 03/2022
 
Title Software - MPT Calculator 
Description MPT-Calculator is a series of python scripts which calls the NGSolve high order finite element method (FEM) library https://ngsolve.org for computing the magnetic polarizability tensor (MPT) for object characterisation in metal detection. In the case of frequency sweeps, this is accelerated by the Proper Orthogonal Decomposition (POD) technique. It is designed as an educational and research tool for engineers, mathematicians and physicists working both academia and industry and it is hoped those interested in characterising conducting permeable objects will find it useful. The MPT characterises the shape, conductivity, permeability of conducting permeable object, is frequency dependent and is independent of the object's position. The rank 2 MPT is symmetric and has at most 6 independent complex coefficients. However, for objects with mirror or rotational symmetries the number of independent coefficients is smaller. MPT-Calculator computes the MPT using a range of different numerical schemes 1) A hp FEM discretisation of the transmission problems using NGSolve to compute MPT for a single frequency. 2) A hp FEM discretisation of the transmission problems using NGSolve for performing the computation of the MPT over a range of frequencies. 3) A Proper Orthogonal Decomposition (POD) reduced order model, which greatly accelerates the computation of the full order model in 2. for computing the MPT over a range of frequencies. The technical details of the implementation have been described in paper submitted for publication. Plots of the computed tensor coefficients as a function of frequency are created and the output data and plots are automatically stored so that they can be recreated, if desired. A series of example geometries are included as is a detailed tutorial. 
Type Of Technology Software 
Year Produced 2020 
Open Source License? Yes  
Impact A paper has been submitted. 
URL https://arxiv.org/abs/2001.07629
 
Description Mathematics for Detecting, Locating and Characterising Metal Objects Impact Event 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Industry/Business
Results and Impact An impact event was organised to promote the research activities of the EPSRC funded project Reducing the threat to public safety: Improved metallic object characterisation, location and detection. It attracted around 50-60 participants from industry, academia and other organisations. The event sparked interest in the work done as part of the project and awareness was raised of the research activity and its outputs. The event was held online given the on-going COVID-19 pandemic in 2021.
Year(s) Of Engagement Activity 2021
URL https://www.icms.org.uk/events/2021/mathematics-detecting-locating-and-characterising-metal-objects
 
Description Research Seminar ( ETH Zurich, Switzerland) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Around 20-30 postgraduate students, researchers and academic staff attended this talk I gave at the Seminar for Applied Mathematics, ETH Zurich, which sparked questions and discussions afterwards.
Year(s) Of Engagement Activity 2019
 
Description Research Seminar (Cardiff University) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact Around 10-20 postgraduate students, researchers and academic staff attended this talk I gave at the School of Mathematics, Cardiff University, which sparked questions and discussions afterwards.
Year(s) Of Engagement Activity 2020
 
Description Research Seminar (Keele University) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Postgraduate students
Results and Impact An online talk given to around 10-20 postgraduate students, researchers and academic staff at the School of Computing & Mathematics, Keele University, which sparked questions and discussions afterwards.
Year(s) Of Engagement Activity 2020
 
Description Research Seminar (Sheffield) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact Around 20-30 postgraduate students, researchers and academic staff attended this talk I gave at the School of Engineering, The University of Sheffield, which sparked questions and discussions afterwards.
Year(s) Of Engagement Activity 2019