“We’re entering a future where these drones will fly all over the countryside,” McKenna said. “But the long-term future of this software is that it lets people fly around.”
With the UK’s National Grid, which operates the country’s energy supply, the relationship became more concrete after the group pledged funding to accelerate the development of Sees.ai’s technology. The partnership’s first goal is to demonstrate that the system can be used to better maintain the grid’s 21,900 steel towers.
Networks need to be constantly adjusted to remain reliable, and regular checks are important. National Grid has 99.99% reliability: it hopes to improve this by locating critical issues long before outages occur. In the UK’s humid climate, the risk of corrosion is high and once it starts, it can be difficult to stop. When rust affects its structural integrity, the tower needs to be replaced, so early detection can save costs in the long run.
National Grid spends around £16m a year to paint its towers and expects to spend £35m over the next five years to replace corroded steel. Given the high cost of research and development, Sees.ai’s drone system is not necessarily cheaper than other inspection methods, but National Grid expects it will enable more frequent and timely data capture, which in turn will save costs through more targeted asset replacements. If the trial is successful, National Grid expects to save UK consumers more than £1 million by 2031.
But until the mass deployment of cost-effective drones, the only option is to use helicopters. A helicopter can inspect 16 pylons an hour at a cost of £2,000 an hour, but flying a VLOS drone isn’t much better, as the pilot below is laborious and slow. When the weather is nice, the VLOS drone team can inspect up to 10 pylons. “It’s the human factor that’s causing the problem,” said National Grid’s condition monitoring manager Mark Simmons.
Sees.ai isn’t the only company tackling this problem, many other companies rely on systems that use GPS and a compass for positioning. The problem is that these technologies are prone to failure, especially near steel or strong electromagnetic fields around high-voltage power lines. Relying on pre-existing data can also be unstable because the world is constantly changing.
GPS technology is also not always accurate, especially when used to measure altitude or in rural areas with poor satellite coverage, according to David Benowitz, head of research at research platform Drone Analyst. He said the risk of a collision in busy airspace was higher because there would always be a “bubble of suspicion”. The greater the vulnerability, the greater the risk.
Therefore, the only way to roll out these technologies is to limit risk in other ways, such as making simpler flights farther away from potential collisions. But with each limit imposed, “the solution becomes less applicable and less scalable,” Benowitz said. If we were to replace manned helicopters, we would need to develop a solution “without these limitations” that would allow for a safe overview and detailed inspection of most assets in the grid, not just remote areas.
For this, more reliable and robust technologies are required: every operating system needs to have multiple layers of security. “In order for us to be able to fly close enough to the tower to get the best data, we need more intelligence than GPS,” Hjamlmarsson said. But regulators like the FAA and CAA also need to make changes in order for these more advanced System development and proper testing create room to prove they are safe. “It’s a chicken-and-egg scenario,” Benowitz said. “These systems are not cutting edge, so there is no problem rolling them out at scale and cost, but regulations need to evolve with the times.”