Autonomous systems designed to enable Europe’s ExoMars rover to navigate on the Red Planet and collect scientific data are finding their way into applications that would make oil extraction more efficient.
According to Walt Aldred, former research director and scientific adviser for drilling at Schlumberger Gould Research, one of the world’s leading providers of drilling systems, the two environments – the distant Mars planet and Earth's terrestrial underground oil reservoirs – present many similar challenges.
“We are operating in harsh environments, in subsurface environments, with high temperature and high pressure, and we were looking for technologies that would help us do this,” Aldred said.
“Similarly to a rover operating on Mars, we have to deal with high latency communication. We have constraints when it comes to power and we have uncertainties in the environment. For instance, we can’t make a detailed map beforehand. We are in a GPS-denied environment so the navigation is difficult.”
The oilfield services company has therefore collaborated with bespoke software firm SciSys, which has been involved in several UK space missions, including the upcoming ExoMars. Together, they looked at some of the cutting edge technologies developed for space projects and combined them with the requirements of hydrocarbon exploration with the aim to develop an autonomous drilling system that would be able to interact and cooperate with a human team.
“We were able to develop a proof of concept system that can actually drill autonomously or with a limited amount of autonomy,” Aldred said. “We have tested this concept in Mexico and Iraq.”
Both sides are benefitting from the cooperation, Aldred says, as the space industry in return gets an opportunity to test its intricate systems in extreme terrestrial conditions before sending the probe on a costly space mission.
“We can test the technology on Earth in quite difficult environments, which would then hopefully make the process easier for projects like the ExoMars rover because you would see if the technology works,” Aldred explained.
““For example, when we have requirements, particularly in dealing with degrees of uncertainty, it might be higher than what they have set. Therefore, in developing a solution with us, they will have a solution that is more robust for the problem that they have.”
According to Aldred, this type of technology research cooperation is not yet common.
“It’s quite difficult to just come to them and say – this technology is applicable. You have to understand what the technology does, it has to be in a state where you can extract a component and plug it into another system. In other words, to take it out from a space system and put it into a drilling system, which is not a trivial exercise necessarily.”
Schlumberger and SciSys are part of the Autonomous and Intelligent System Partnership co-funded by the Engineering and Physical Sciences Research Council. The partnership, which includes representatives of various industries, including nuclear, defence and transport, looks at possible common problems with the aim to developed shared solutions.
“If you can commercialise a technology in multiple markets, it’s much more attractive than commercialising it in a very specialised market like ours,” said Aldred.
SciSys found that the number of Earth-based applications that could take advantage of systems developed for Mars exploration is higher than expected.
“Things like autonomous navigation could be used in driverless cars and robots,” said Mark Woods, Head of Autonomy and Robotics at SciSys.
“Another technique that we have been developing quite heavily is what we call science autonomy. The problem with the rover on Mars is that it can send back only a limited amount of data. So you want to have some artificial intelligence algorithms in place that would allow the rover to figure out which things are really of interest.”
In the next step, the robot combines all relevant visual data into a 3D model with an overlay of information from other sensors.
This autonomous ability to explore the environment, map it and find objects of interest could be of use, for example, to operators of railway and underground tunnels, which need to be regularly inspected for cracks and faults.
“At the moment, they are sending real people to sort these things out,” said Woods. “It’s a classical problem of the dull, dirty and dangerous. A lot of people think that in 2016, we shouldn’t be sending people to underground tunnels where there are serious safety risks.”
Instead of surveyors taking pictures and writing reports, an ExoMars-like rover would scan the tunnel, measure cracks, and deliver all the data in a form of a 3D model into an operator’s iPad, computer or virtual reality headset.
“These tunnels can be very long,” Woods said. “By using multiple robots, you can potentially do everything and also at the same time build much more useful 3D models. It’s a very different capability than what we currently have. These tunnels may change year on year and with different people carrying out the assessment every year, the results can also be quite different.”
Similarly, Mars-inspired robots could work in contaminated environments in the nuclear industry or inspect oil and gas pipe lines in the desert and other inaccessible environments.