‘Robots can make engineering activities safer, quicker and cheaper’
Image credit: Nick Smith
BT’s Jon Wakeling discusses how robotics innovation at its telecoms research laboratories in Suffolk is taking the heavy lifting out of ‘digging holes and climbing poles’.
“At the moment, the devices we are developing still have an operator in the loop,” says Jon Wakeling. “But if you were to talk to me in five-to-ten years’ time, we’d be discussing device automation and autonomy.” Wakeling, who is acting director at Openreach Research (part of British Telecom’s Applied Research division), is discussing robots currently under development in BT’s latest research facility at Adastral Park, where he leads “a small team performing technology and capability discovery, working with universities to take concepts through to early-stage prototypes”.
In a utopian future, according to Wakeling, we will have robots at our disposal that dig under roads and across gardens to deploy cables, freeing up human engineering capacity. That is the long-term objective, he says, “but there are a host of challenges” that could obstruct this pathway. While many such obstacles can be circumvented through technology development, there remain others connected to regulation and public trust, especially: “how will governments and individuals react to autonomous devices tunnelling underneath their driveways?” Robotic devices, he says, are the route to keeping people connected to the phone network, “but we’ve got a long way to go”.
Wakeling sees his role as collaborating on moving these devices from low technology readiness levels (TRL) “through to round about TRL7” (which is system prototype demonstration level). He observes that lots of these ideas fall by the wayside, “but some will make their way through the product development hopper to the point where they can be industrialised and commercialised. At that point you can start to bring in manufacturing partners, so that you can look at the research and prototypes to understand what can genuinely be manufactured.” What Openreach’s research is all about, he says, is to accelerate from the lower to higher TRLs.
Openreach is the wholly owned BT plc company that maintains telephone cables, ducts, cabinets and exchanges connecting virtually every home and business in the UK to the national broadband and telephone network. Wakeling says BT’s interest in robotics is driven by its target of delivering optical fibre connection to 25 million UK premises by the end of 2026. It’s a programme that requires “an awful lot of very basic civil engineering in terms of digging holes and climbing poles”.
And there are plenty of reasons to apply robotics to this fieldwork, “which hasn’t changed much in the past hundred years”. The British engineer reels them off: “It’s expensive, time consuming and can be dangerous. Also, a lot of the processes we have in the field are quite slow. We think the technology has reached the level of maturity where we can ask if robotics can help to dig in new infrastructure, retain pre-existing infrastructure and make the engineering activities safer, quicker and – let’s be honest – cheaper.”
At the back end of 2021, BT confirmed its commitment to developing its fibre network with the opening of the UK’s first dedicated telecoms civil engineering test centre at its Adastral Park laboratories in Martlesham Heath, Suffolk. The 5,000 sq. ft. (500m2)research facility has the stated aim of playing “a key role in developing innovative robotics solutions to speed the deployment of essential infrastructure, working with the university sector and other utilities to trial a new range of UK-developed robotics which are applicable to telecoms and utility sector civil engineering challenges worldwide”.
BT has said that it will use the opportunity to engage with the university sector and robotics start-ups, “as they develop robotic solutions to challenging civil engineering tasks”, including addressing problems such as how to clear blocked ducts, mend collapsed ducts and install new fibre network infrastructure “without incurring the cost and delays that come with digging up roads and pavements.”
Wakelin describes the research facility as centring on simulations of three environments we find in the real world. First is “emulating the underground environment which we can fill with any soil type found in the UK. We can control this for moisture and aggregate content, and we can also lay different surfaces on top, such as paving slabs or tarmac. If we want to test robotic devices for digging in the ground, then we want a controlled environment where we can look at how a robot behaves in a given environment.” New biomimetic robotic locomotion and excavation techniques inspired by digging and burrowing mammals and insects, coupled with the latest technologies developed for space exploration, aerospace and medical applications, “are showing promise for delivering ‘trenchless’ infrastructure deployment”.
The second environment under research at Adastral Park is ducting. “We are working to recreate existing problems that occur in this environment. I usually refer to our duct network as being the second artificial water drain for the UK,” says Wakeling, “because when it rains, it fills with water and all the associated silt and detritus (such as leaves) gets washed in. As our ducts follow the contours of the land, they have natural sumps, and these fill up with material and get blocked to the point where you cannot push another cable through. If you try to push a fibre cable through to replace a copper cable, then you have problems. If we want to develop robots that can survey this sort of environment, we need to be able to see what’s going on inside the duct, so we use transparent ducts.”
Meanwhile in the overhead environment, magnetic, climbing and cable-traversing robotic techniques are “also maturing, enabling proof-of-concept trials on wireless tower and overhead cable poles”. Wakeling says one of the strengths of the facility is that it has on-site a full-height telegraph pole with platform access to allow different pole-top fixtures to be fitted to test robots designed to lift tools, equipment, or cable to the top of a pole. Cables can be run to another pole to provide a single 35m span for development of cable-car-like devices for pulling in new cables or dealing with tree canopies. A series of shorter posts provides a ‘pole-transit’ testbed for developing other cable-travelling devices.
Wakeling says maintaining current telecoms infrastructure is vital because the cost of replacing it is simply so high. “Physically replacing a 100-metre duct run is hugely expensive. Depending on the surface, it could cost anything up to £175 per metre to dig in that infrastructure.” While cost avoidance is one consideration, perhaps more significant is the inconvenience of replacing infrastructure. When carrying out public works, “if you want to either partially or fully close a footway or carriageway, you must apply for permission which can take months to be granted”.
It’s not unheard of for delays caused by bureaucracy to slow down maintenance projects by as much as two years. “You can’t build a plan around that,” says Wakeling, who thinks that robotics technology is a solution to getting around such delays. “But there are also good secondary benefits to consider: every time someone digs up a road, tailbacks are created and there is a domino effect into the economy in terms of lost work hours and environmental impact. So there are plenty of reasons why we would like to remedy existing problems rather than lay new infrastructure.”
Wakeling explains how, when BT started to develop the idea of creating a robotics research facility two-and-a-half years ago, “we had absolutely no expertise or experience in robotics. We thought that if we were going to create this ourselves from the ground up it would probably take four or five years to deliver anything useful. So, we decided to try to intercept all the good work that was going on in universities” that was being funded by UK Research and Innovation (UKRI) in the form of four robotics hubs, bringing together academia and industry to work in the seemingly unrelated nuclear, offshore energy and robotics for space sectors.
“You can draw parallels with what these technologies are being developed to do in these environments and what we are doing. So, we have intercepted some of these ideas and worked to repurpose them for our use cases. In doing that, we’ve probably halved the time it would have taken to get us to the preliminary prototypes that we have today: what we call our proof-of-concept devices. We’ve been working with half a dozen universities on these devices, and we will continue to do so until we get to a level of maturity in the technology where we can field-trial them.”
Wakeling says BT is experimenting with multiple technologies in a variety of form factors to assist the organisation’s fibre deployment programme. When it comes to congested duct exploration, he cites the example of a device comprising three inflatable chambers that imitate a worm’s peristaltic movement. “We’re using this technology because it naturally has a way of finding the path of least resistance. We want to see how small we can get this technology; if we can get cameras on it so that it can characterise blockages and so allow us to get a better understanding of what’s going on.”
As the Worm - which is being developed by the University of Bristol - has a soft and flexible exterior, it is not necessarily suited to environments containing sharp media such as stones and other obstructions. This means the laboratory is simultaneously developing alternative innovations, such as the Cobra in conjunction with the University of Nottingham, which “piggy-backs onto work that they’ve been doing with Rolls-Royce”.
The Cobra has its origins in turbine blade inspection, “where it could weave its way around – and we wanted something similar for duct inspection”. This is needed because “strange geometries” occur in ducts, created by a phenomenon known as ‘cable twist’, in which copper wires coming off a drum retain some of the wound form as a corkscrew. “They don’t lie flat at the bottom of the duct. Having several of these cables corkscrewing at different rates makes it difficult to thread a new one into the space, as it can cul-de-sac because of the way that other cables are configured. But if you could steer the tip, you could push past that. The University of Nottingham has put an adjustable head on the end of one of our rods, and you can steer it to pull cables past a blockage.”
A third in-duct technology under development is the Bristlebot, a 3D-printed tube developed by the University of Leeds that, as its name suggests, is a robot with broom bristles stuck into its exterior. Inside the tube is a vibration motor, and the angle of the bristles transmits the vibration into forward motion. One application for Bristlebot is in new housing. “Housing estate developers are supposed to put ducts underground with drawstrings in them so that we can come along and put the cable in.” Nine times out of ten, says Wakeling, these drawstrings are conspicuous by their absence, which means the engineers on site must call in a rodding team to put in the drawstrings. The Bristlebot, which Wakeling hopes will one day form a part of the site engineer’s everyday kit, “will pull the drawstring through for you. It’s part of our acceleration approach to taking away all these irritating little problems that we come up against in the fibre deployment programme.”
At this point, Wakeling explains that his association with BT goes back to the 1980s when he read communications engineering at the University of Plymouth as part of an apprenticeship programme. “I did various placements around BT as part of my university course, and arrived at BT’s research labs, as they were called back then, in 1989. At the time I was working on satellite communications,” as part of a tour of duty around the telecoms giant that also included a stint in the CTO’s office. This was followed by “a return to research, where I am able to exploit my experience in field engineering and group strategy, in which I was involved in the planning for the first superfast broadband rollout back in 2010”.
Wakeling likes that his current incarnation at BT is a return to his roots, or, perhaps more accurately, his high-vis, hard hat and muddy boots. “My first placement as an apprentice was in the field with the gangs that were digging holes and putting up poles, clearing ducts and putting in cables. I would argue that my career has almost come full circle in that I’m now looking to see if I can fix those issues and challenges I encountered back then. I have seen problems from the field engineering perspective. I have seen problems from the planning perspective. I’ve seen the whole thing, which is why it’s doubly exciting to be back in the research lab trying to solve these problems with the latest technology.”
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.