We've now written on the underground utilities problem from two perspectives: why the records themselves keep falling out of date, and how AR scanning can verify placement before a trench gets backfilled. RodRadar's Live Dig Radar addresses this same issue from a third perspective: not fixing the record, and not verifying after the fact, but intervening at the precise moment a bucket is about to strike the utility.
What Live Dig Radar Actually Does
LDR Excavate places a ground-penetrating radar sensor in the bucket of an excavator. Every time the bucket bites into the ground, the system takes a new radar scan of what's immediately in front, diagnoses what it sees using onboard AI hardware, and gives the driver an immediate visual and audible warning – along with an approximation of the depth – before the next scoop. Network Plus, one of the largest utility and infrastructure contractors in the UK, working with Severn Trent Water, Cadent Gas, National Grid, and UK Power Networks, recently announced a multi-year rollout of this technology across its entire fleet of over 200 excavators. That's a level that indicates utility strike prevention is no longer a niche safety feature; it's a serious commercial operational focus.
The Innovation Is "In the Ground, In the Moment"
Traditional ground-penetrating radar and electromagnetic locators are used prior to a dig, and generally need somebody trained in their use to read the trace. That's a fundamentally different moment in the workflow than what LDR provides: detection built into the tool doing the digging; no one needs to read an external device. Every time you move the bucket you get a new scan and that's super important because even with a pre-dig survey you can get the ground or utility location wrong-and a live, always-updating layer of detection picks up what a one-off pre-dig scan might have missed or read wrong.
Introducing the New 30cm Bucket: Designed for the Toughest Environment
Network Plus is the first operator to use RodRadar's new Type 0 LDR Excavate - a 30cm-wide (roughly 12in) digging bucket (even narrower than the Type 4 and Type 5 buckets used in 2025 on major sites with larger 20-40 tonne excavators). And there's a reason for that: in congested urban streetworks, where tight clusters of shallow electricity, gas, water and telecoms are typically all close by, a narrow bucket is the only sensible way to go about that kind of excavation. Reducing the detection gear to fit that narrow bucket delivers real-time radar right where the greatest density of shallow, hazardous utilities lies.
Why the £2.4 Billion Figure Matters
Utility strikes aren't just a safety metric – they represent a direct, tangible cost to the UK's infrastructure industry that they've been keeping an eye on for some time, and the £2.4bn a year figure is a large enough number that avoiding strikes has gone from being a compliance tick in a box to a real commercial imperative. For every strike, there's an unplanned repair to make, a delay to the programme, a disruption to homes and businesses connected to that utility, and in a few cases a safety incident affecting the team doing the excavating. A big, company-wide multi-year programme from a contractor of Network Plus's size – serving four major national asset owners – indicates that this isn't a pilot experiment; this is now business-as-usual equipment going forward.
How This Fits Alongside Documentation and Verification
Detection during the dig, verification before backfilling, and well-verified records after the fact are three different moments in the same lifecycle, and each one addresses a different failure mode. Accurate, up-to-date utility records are still worthless if a bucket hits a line that is just a little bit wrong compared to what the record indicated. AR verification before backfill is still worthless if the line was hit earlier in the same dig. Real-time radar while excavating is the layer that detects that third failure mode at the source – it's a complement to good records and good documentation, not a substitute for either.
What the AI Classification Layer Actually Adds
Up to now, GPR raw data has only been reliably interpreted by a trained technician - nobody wants to have to differentiate between a utility pipe and a rock, tree roots, or a density change in the soil, which is a lot to ask of any straightforward reading (which is part of why traditional GPR surveys have been performed as a dedicated, non-continuous activity rather than a continuous operation). Incorporating AI classification into the processing by the bucket itself removes that limitation: instead of giving the operator just the raw trace and expecting them to figure out what it is, the system delivers a direct alert rather than a wall of data. That change – from "here's a raw signal, figure it out yourself" to "here's what we detected and how deep it is" – is perhaps just as important as the real-time operation itself, since it's what lets a machine operator operate instead of requiring an expert on the site.
Frequently Asked Questions
Does this make the pre-dig utility survey or accurate GIS records redundant?
No. Pre-dig surveys and good record-keeping mean fewer surprises for the excavator crew at the site. Live Dig Radar is the safety layer that fits over whatever you missed, misjudged, or slightly misplaced.
Does the operator require specialist training to use the radar feed?
This is intentionally avoided - the system makes its own classifications and delivers an immediate alert instead of a raw trace that needs a specialist to interpret it, like many traditional GPR survey devices.
Is this only intended for larger utility contractors?
The new 30cm bucket size was created for a very specific reason: larger bucket sizes that were already available simply weren't a good fit for the small, tightly congested urban areas where many utility strikes are actually happening - indicating that this technology is being purposefully scaled down, toward less massive excavation jobs.
Implications for Companies Operating Near Buried Infrastructure
The wider story across all three of these underground-utility technologies - record-keeping, AR verification, and now real-time detection - is that the industry is finally trying to hit this problem at every weak point, rather than using a single layer of protection to catch all errors. If you're doing GIS-based work excavating around existing infrastructure, the best question to ask isn't "do we have good records?" but instead: "are we protecting every point along the line at which a mistake could happen?" Because the UK alone loses £2.4 billion a year to utility strikes, and that sure sounds like it's not one thing alone getting us there.
