The challenge

Context of the challenge

Bomb disposal is dangerous work undertaken by military and policing agencies among other government organisations. A key part of the process to safely disable a suspected explosive device while limiting physical contact, is using x-ray images to identify the electronic components that make up the switching mechanism designed to detonate a suspected explosive device.

This task can be further complicated by the presence of tripwires, anti-tamper mechanisms, and other countermeasures that may trigger the device if disturbed

The gap

Analysing x-ray images of explosive devices is a highly specialised skill. Because specialists are not usually deployed in the field, operators on the ground must capture images and send them to remote experts for analysis and advice.

This process can be slowed down by poor-quality images taken under operational pressure, the need for additional imaging, and the time it takes to send large files over limited communications links.

Advances in artificial intelligence and machine learning, used to identify key electronic components in complicated circuits, could bring some of this analysis closer to the point of need. This could help operators identify and classify electronic components more quickly in the field, while ensuring human operators remain responsible for all operational decisions.

Example use case

Sergeant Heyland is a member of a specialised Explosive Ordnance Disposal (EOD) team deployed overseas.

While on patrol, he finds a suspected Improvised Explosive Device (IED). It looks complex, so the analysis is likely to be difficult. After securing the area, he takes x-ray images of the device from multiple angles to gather as much information as possible, then withdraws to a safe distance.

Using a newly deployed analysis tool, Sergeant Heyland processes the images himself, onsite. The software identifies and labels electronic components within the images, flags anything it is uncertain about and asks for further images to improve confidence in the assessment.

Once Sergeant Heyland provides these further images, the system identifies a range of electronic components and provides assesses their likely characteristics and functions and highlights any that may be critical to how the device works.

As this is safety-critical work, Sergeant Heyland still consults UK-based specialists. However, he can now send smaller, more focused information, so his questions are answered more quickly.

This means Sergeant Heyland can complete his task more efficiently, while specialists continue to provide expert oversight.

Project scope

This challenge is about developing machine learning algorithms that can identify and label electronic components from x-ray images while assessing their likely function within a circuit.

Applicants should aim to deliver a demonstrator at a minimum of Technology Readiness Level (TRL) 5 (technology validated in a relevant environment) within the 12-week project period.

Essential requirements:

• Deliver a demonstration to the sponsors, along with the software, source code and a report detailing how it functions.
• Use machine learning techniques to identify and analyse electronic components
• Ensure the software can be used by trained operators who are not electronic component specialists
• Provide traceability and explainability for outputs.
• Show confidence scores for identified components and extracted information.
• Run on commercially available hardware.
• Support multiple data types, particularly TIFF and JPEG files.
• Demonstrate identification and analysis of through-hole components, such as capacitors and resistors.

Desirable requirements:

• Support offline or disconnected operation.
• Support retraining or refinement using additional labelled data.
• Demonstrate identification and analysis of surface mount components, such as microcontrollers, capacitors and resistors.
• Generate structured reports suitable for technical assessment activities.

Constraints:

• Training data will be made available to the winning solution provider.
• Solutions should avoid reliance on proprietary datasets where possible.
• The solution must use x-ray images as its primary source of analysis.
• Solutions should provide evidence-backed outputs and clearly explain any uncertainty.

Not required:

• A horizon scan.

Eligibility

This challenge is open to sole innovators, industry, academic and research organisations of all types and sizes. There is no requirement for security clearances.

Solution providers or direct collaboration from countries listed by the UK government under trade sanctions and/or arms embargoes, are not eligible for HMGCC Co-Creation challenges.

Invitation to present

Successful applicants will be invited to a pitch day, giving them a chance to meet the HMGCC Co-Creation team and pitch the proposal during a 20-minute presentation, followed by questions.

After the pitch day, a final funding decision will be made. For unsuccessful applicants, feedback will be given in a timely manner.

Clarifying questions

Clarifying questions or general requests for assistance can be submitted directly to  [email protected] before the deadline with the challenge title as the subject. These clarifying questions may be technical, procedural, or commercial in subject, or anything else where assistance is required. Please note that answered questions will be published to facilitate a fair and open competition.

How to apply

Please submit your application on the HMGCC Co-Creation website. Any queries please email [email protected] and [email protected].

All information you provide to us as part of your application will be handled in confidence.