Subsea saviours

Your boat has capsized. Seconds before you were tossed into the angry spume you sent a Mayday signal, but as conditions worsen the chances of rescue seem slim.

Off in the distance you spy a rescue vessel fruitlessly searching for you in the foam and, amid the crashing waves, you are taunted by the hum of a spotter plane, straining for a glimpse through the driving rain.

Just when you think things can get no worse, you catch a glimpse of a terrifying dark shape slicing through the water beneath your legs.

Resigned to your fate as shark-fodder all hope seems lost. But then, your subsea stalker reveals its true identity. It is a robot. And it has come to rescue you.

Greeting you with an assurance that help is on its way, your submersible saviour transmits your precise location to the master rescue vessel and inflates a life raft for you to cling to.

This, according to underwater vehicle specialists, is the future of marine search-and-rescue.

Increasingly used in a variety of offshore and military subsea applications, the autonomous underwater vehicle (AUV) is, like its aerial cousin the UAV, coming of age. Growing affordability, improvements in range and speed and increasingly advanced subsea imaging techniques are combining to make AUVs attractive for a range of new applications, including search-and-rescue (SAR). Murphy claims that AUVs, dropped into the water from an aircraft or launched, torpedo-like, from a lifeboat, could be used to rapidly survey huge areas of the surface of the water, enhancing the capabilities of marine search-and-rescue services and making the process considerably less dangerous for the rescuers.

With marine incidents typically occurring in adverse weather conditions, conventional approaches to marine SAR are, he claimed, at an immediate disadvantage, with the aircraft and boat crews frequently struggling to locate ‘men overboard’ in the chaos of a storm. ‘The underwater vehicle is immune to the harsh weather conditions on the surface,’ he claimed. ‘Helicopters and boats get confused very quickly by adverse sea conditions, but if you go under water the effects of waves are negated.’

Murphy’s idea is to replace the height of vision provided from the deck of a boat or helicopter with what he terms ‘depth of water’, where upward-pointing sensors mounted on the rescue AUV would monitor the surface of the water from below, scanning a wider search space than is possible with existing techniques.

Murphy is not the first person to try to take the human element out of marine SAR. Two years ago, Canadian AUV pioneer International Submarine Engineering(ISE) worked with the Canadian airforce on the development of an unmanned rescue vehicle designed to operate on the surface of the water.

The Search and Rescue Portable, Air-Launchable (SARPAL) GPS-enabled remote controlled dinghy was designed to be dropped from an aircraft. Upon hitting the surface of the water, it automatically inflated and onboard cameras enabled operators on the plane to direct the craft towards survivors.

James McFarlane, ISE president, said the system was designed to overcome a fundamental shortcoming of existing SAR techniques. ‘Often, if you throw a floatation package down, the wind blows it away and you watch the person drown.’

ISE successfully demonstrated two prototypes before the funding ran out three years ago. But McFarlane believes the time could now be right to relaunch the concept.

Murphy agrees, and suggested the technology could receive a boost from the military’s enhanced focus on port security and the development of autonomous military surface vessels such as the BAE protector.

Back beneath the waves the remotely operated submersible (ROV) the AUV’s more primitive tethered forbear, has a pedigree in investigating underwater accidents, including the sinking of Russian nuclear submarine Kursk in 2000. ROVs developed by Hampshire firm SeaEye were rushed to the Barents Sea to probe the wreckage. Everyone aboard Kursk lost their lives.

SeaEye’s technology was used again to carry out video surveillance of another stricken Russian sub, Pritz, which became trapped on the seabed in 2005. This time, the information gathered by the system was a vital part of a rescue mission that brought the entire crew safely back to the surface.

The robot in question, the 500kg Panther Plus, is equipped with sonar and an acoustic tracking system for locating a distressed submarine and a series of manipulators and cutters for clearing debris and inserting emergency supplies to the trapped submariners.

The Russian navy was so impressed that it is putting SeaEye’s ROV technology at the heart of a new rapid response submarine rescue system under development.

SeaEye’s owner, Saab, was recently asked to supply an AUV for a Swedish murder investigation.

Carl-Marcus Remen, Saab Underwater Systems spokesman, said police used the company’s long-range AUV62, typically used for mine reconnaissance, to search for the body of a woman at the bottom of a lake near Motala.

Equipped with a high-definition sonar able to probe the 40cm of silt on the lake bottom, the craft took just two days to survey in detail what would have taken police divers weeks.

Though the body was not found, the system confirmed its absence and, according to Remen, the Swedish police are now talking to Saab about a lease contract.

According to SeaByte, a UK-based developer of AUV software tools, there could be a strong case for a similar arrangement in the UK. Company spokesman Dr Ioseba Tena told The Engineer he sees an opportunity for an enterprising AUV developer to devise a low-cost system for SAR that could be leased out when required.

He added that as the AUV market grows and the cost of the technology falls, it will become more attractive to search-and-rescue authorities. ‘We think they’re a pretty good platform for search-and-rescue,’ he said. ‘While the AUV market has taken longer to take off than initially predicted, as cost comes down SAR will begin to make more and more sense.’

While the use of AUVs to search lakes and oceans moves the technology into a new domain, such applications are some way from Murphy’s more dynamic vision of life-saving rescue robots. For this to become a reality, some fundamental technological breakthroughs are required.

One reason that tethered ROVs are attractive for search-and-rescue is that they are typically equipped with robotic arms, or manipulators, that can be operated by technicians on the surface. Murphy believes AUVs will require the ability to carry out what he calls ‘intervention activities’ to fulfil their potential. ‘They can’t touch and feel and sample solid matter — for example a ROV has manipulators and grabs, and you can do all sorts of things because there’s a human in the loop.’

The development of an autonomous manipulation system presents significant challenges, requiring advanced computer processing, sophisticated machine vision techniques, and dexterous robot manipulators.

As in many other areas of technology, it seems that the space industry, skilled in developing robotic manipulators for lunar and Martian landers, may provide some of the answers.

In one promising project, engineers at the University of Maryland’s Space Systems Lab have adapted robotic technology developed for spacecraft assembly to autonomously collect samples from hydrothermal vents. According to the team, the SAMURAI (Sub-polar Ice Advanced Manipulator for Universal Sampling and Autonomous Intervention) has an autonomous vision system and a six-degree-of-freedom dexterous manipulator that can be fitted with a range of end-effectors.

In a separate development, researchers at MIT’s AUV laboratory are developing a mechanical arm for the recently-launched Odyssey IV AUV. Prof Chryssostomos Chryssostomidis, the group’s leader, says the arm will enable the underwater vessel to perform manipulations such as opening or closing a valve, picking up objects and even carrying our repairs.

An even more important capability for the rescue AUV is the ability to locate and identify humans in the water. Murphy pointed to numerous sensing and sonar systems that could be adapted for this purpose.

For example, sonar-based ‘fish finders’, used by fishermen, are relatively inexpensive and could potentially provide the area coverage and resolution required for human search-and-rescue. Such systems have been used to detect resting marine animals floating on the surface and could, said Murphy, ‘equally be applied to the location of survivors of a marine incident’.

He said a rescue sub might also benefit from an adapted version of the Recco Avalanche System, which uses harmonic radar to pinpoint the location of trapped individuals by picking up signals from reflectors attached to their clothing. ‘Such a system could be applied to marine outer and safety clothing, and the detector located upon the autonomous search vehicle could facilitate faster and more efficient location of victims of marine accidents,’ he said.

Another promising technology is high-resolution synthetic aperture sonar, a sophisticated form of sonar that combines a number of acoustic pings to form an image with much higher resolution than conventional sonars.

Norway’s Kongsberg Maritime, one of the world’s largest manufacturers of AUVs, is excited about the application of this technique.

Bjorn Jalving, AUV chief, said while the technology was developed for naval mine countermeasures, it has many other potential applications.

‘Due to unprecedented resolution and area coverage rate, one possible application of AUVs equipped with synthetic aperture sonar can be very efficient search-and-salvage operations,’ he said. Though current search applications are limited to looking for debris on the seabed, the system could also be used to look up at the surface of the water. Jalving added that when travelling at four knots, the company’s HISAS1030 system is able to scan 400m swathes of seabed down to a resolution of 5cm².

Murphy is keeping a close eye on such developments and continues to sketch out his plan and build on his dialogue with the search-and-rescue services who are, he claimed, excited about the technology.

‘We’ve talked to the RNLI and the US coast guard about having these things semi-autonomously following a lifeboat and there’s a definite interest. The problem the rescue services have is limited assets and limited coverage capability — and that’s where this sort of thing comes in.’

For the immediate future it will be business as usual for lifeboat crews. But there is evidence that as AUV technology matures, underwater robots could one day become an invaluable help for rescue operations.