Guidance

Innovative Research Call 2023: Clarification Questions

Updated 16 August 2023

1. Buildings and Areas

Q: Would there be interest in a system-of-systems approach for both military and domestic requirements?

A: We would be interested in any approach capable of detecting our target threats, so long as it is technically feasible and viable to deploy. There are a range of end user groups who have differing operational requirements and limitations (sometimes associated with available budgets and the technical skills of the operators) so the viability of deploying and maintaining any system would be key to its success. It is important to identify who the potential end users would be (scientist, engineer, police officer, security guard, etc.) and to ensure the proposed solution would be suitable for use by that group (or groups).

Q: Many things appear to be in scope, but are there any pressing priority gaps?

A: A difficult question to answer as every search operation is slightly different in regards to the size/scale, venue layout and the complexity of the area to be searched, etc. However, priority is given to threats that can cause the greatest harm in a particular scenario. In some building search scenarios, the time available to complete the search is limited. This means that there is often a trade-off between speed and assurance. False indications which lead to a lengthy alarm resolution step often have the biggest impact on the time available, so selectivity/accuracy of any technique used is important in a building search context. Also, if it is not possible to confirm that a venue is clear from threats, this may lead to cancellation or rescheduling of an event, which is obviously not a desired outcome.

Q: In respect to explosives trace detection, is it appropriate to utilise multiple parallel sample collections back to a single high assurance analysis instrument to increase capacity AND what are the challenges of that?

A: Yes. This is a methodology that is already used in some cases. One of the main difficulties is ensuring all samples are collected in the same reproducible way and that the inevitable random error associated with sampling is reduced as far as possible. At present, sample collectors undertake training and familiarisation, but building further consistency into the system would be beneficial.

Q: Are building searches affected by the local level of existing security AND does the layout of the building affect the procedure of the search?

A: Yes. Although the overarching concepts and doctrine remain the same, the tactical delivery of each operation is slightly different, depending on the specific variables at that location. One of the variables is whether there is an existing security presence and protocol at that specific venue. Where this exists, it may be possible to use the existing security presence to prevent members of the public entering specific areas of a building, although a high turnover of casual security staff may pose a challenge in itself.

Q: Which stakeholders would be looking to use the proposed innovations (Police, military, ambulance, etc.) and should a solution seek to meet everyone’s requirements?

A: In some cases, a range of stakeholders may wish to use the same solution in exactly the same way. However, in other cases, the stakeholders’ concepts of operations may differ greatly. Although it may be attractive for a supplier to develop a ‘one-size-fits-all’ solution, care should be taken to ensure this does not dilute its effectiveness. It may be better for a solution to focus on being excellent in one use case, rather than average in several.

Q: If samples are collected from defined areas repeatedly, can that data be collated to understand trends or to monitor baseline levels?

A: Yes, in theory this would be beneficial, however, depending on the technique being used this can be quite difficult to achieve. For instance, if explosives trace detection were being used, it would rely on obtaining truly quantitative results. Although several quantitative analytical techniques exist, there are significant random errors associated with sample collection processes, which leads to the overall results being qualitative/indicative only. For example, the background level of trace explosives in the Brighton Grand Hotel is known to still be high, following the 1984 bombing of the hotel, however, it is not possible to accurately define a quantitative level of contamination.

Q: Does the scope include underwater detection techniques?

A: Yes, providing it has some form of target discrimination/identification capability. We currently have underwater anomaly detection capabilities, such as sonar, so any proposed solution should provide enhanced performance over these. There is currently a growing military interest in this area.

2. Goods

Q: What level or size of threat are you trying to detect?

A: This is difficult to answer because the “goods” category can encompass a small parcel, palletised goods, or a 40ft container, therefore the actual threat quantities within those respective contexts will widely vary. There is a great deal of sensitivity regarding specific quantities we are looking to detect. We acknowledge that quantities below 1 kg, especially surrounded by clutter are very challenging, but would provide us with a beneficial capability. To give some steer, the Explosives Regulations 2014 states that, with some exceptions, for an individual with an “Acquire and Keep” licence, manufacture of explosives in excess of 100g is illegal, so for us, the ability to detect any explosives in excess of this amount is desirable.

Q: In terms of Raman spectroscopy that is used for materials identification, how do we deal with black powders (and other threats that are hazardous with Raman)?

A: Raman is successfully used against precursors, hazardous chemicals and illicit drugs. It best works on white and light-coloured materials. Dark materials can be problematic. Training is given which mitigates some of the risks (small sample size, lower laser power, delay in beginning analysis etc). Raman is just one technology in use. It is worth keeping in mind that not all techniques are suitable for use against all threats.

Q: Is there previous data collected from searches that could be shared with industry (to aide AI development for given scenarios)?

A: Unfortunately, there are no large datasets in a sharable format that are available. If a proposal required Government Furnished Assets (GFA) it would be closely looked at, but there should always be a contingency plan to cover non-availability of the GFA.

Q: What percentage of items are screened by X-ray and is automated threat detection deployed?

A: In the postal-type environment, any consignments due to be loaded as air freight will be screened for aviation security threats, but these are for export. For consignments received at an international hub there is no automated screening on arrival currently, but high-risk consignments will be separated, subjected to risk-analysis and potentially physical inspection. The speed that the commercial flows move at on belts make it difficult to implement technology that has a high degree of confidence. In addition the operator would not be supportive of technology that would adversely impact their business model and cause them to fall short of their service level agreements.

Q: Is anything being done in the UK/Europe to enable (X-ray) exposure to drivers (to speed up interrogation of goods on lorries)?

A: Many of the existing large x-ray scanners have a capability that allows screening with x-rays to commence once it recognises the drivers cab has cleared. It is understood that a few places in Europe have permitted this, but it is not a procedure that is currently possible in the UK, although there is interest among the stakeholders to explore the benefits.

Q: Are (goods in) electric vehicles in scope and what adaptions would have to be made?

A: The market for electric vehicles will continue to grow and they are considered in scope for this call under the vehicles challenge area, as are hybrid and hydrogen vehicles. The configuration of these vehicles will tempt some criminals to try and exploit it. Screening the load-carrying area is the focus under this particular challenge.

3. People and/or possessions

Q: Is there scope to modify ConOps to enable different outcomes and different technical solutions?

A: We are interested in any innovations which deliver benefit to the end user. This may include changes to Concepts of Operation as well as development of hardware, software or new/modified processes. The bid will need to fully explain what the innovative approach is and how the benefit of this approach can be quantified.

Q: How should industry tailor its project proposal, should they focus on smaller components or try to deliver a complete solution?

A: We are interested in any innovations which deliver benefit to the end user. This could be a full system or a component of the system. If the bid only addresses a partial solution to an issue and requires integration with existing solutions, then we would expect to see any risks to exploitation outlined in the bid as well as the end user.

4. Vehicles

Q: Should we consider temporary or permanent solutions AND should it be threat detection or threat identification?

A: We are interested in both temporary and permanent solutions for vehicle screening. Given the complex detection nature of threats in vehicles and the large amount of effort to resolve a false/nuisance alarm, solutions that provide the detection AND classification of threats i.e. explosives, firearms, gas canisters, are in scope but those delivering only anomaly detection are NOT in scope. We are interested in threat identification, if possible, but realise this is technically challenging. A specific niche application to consider is the ability to provide assurance that there is no threat left in a vehicle.

Q: What is in place currently, for both, in motion and stationary screening of vehicles (and can that data be made available for AI development)?

A: Currently, there are limited affordable options available for screening vehicles that are stationary or in motion that do not use either X-ray technologies or manual search tools and techniques.

Some fleets of vehicles are regularly screened so there may be baseline data available, but it is unlikely to be packaged in a way that makes it usefully shareable. A greater challenge would be acquiring data for unknown vehicles due to the level of modification expected.

Q: Can we reduce the number of required searches based on obvious vehicle modifications (for concealed hides)?

A: Due to the number of variations possible in vehicles, for example, addition of a roof rack or spoiler, we see only limited utility in a solution that is restricted to obvious modifications. Furthermore, significant numbers of concealed hides are designed into the fabric of the vehicle so as not to be an obvious modification but are still required to be investigated / analysed for threat detection.

Q: Are there any examples that could be shared of threats in vehicles (to help understand the scope)?

A: Different scenarios will involve different potential threats. We expect terrorists to innovate so whilst there are previous examples (Glasgow Airport in 2007, Tiger Tiger nightclub in 2007) new methods of attack and therefore new screening challenges are likely. If the solution is concentrating on a Vehicle Borne Improvised Explosive Device (VBIED) threat, for example in a moving vehicle, the mass range is 10’s – 100’s kg in smaller vehicles and 100’s – 1,000’s kg in larger vehicles. For searches of stationary vehicles, we are interested in location of components, materials of concern or complete threat devices and weapons.

Q: What are the threat levels/quantities that should be considered for vehicles?

A: The threat level or quantity will depend on the proposed CONOPS, but for VBIEDs can be considered to have the mass range of 10’s – 100’s kg in smaller vehicles and 100’s – 1,000’s kg in larger vehicles.