As the development of autonomous watercraft progresses, engineers and designers are busy making sure these devices can communicate effectively.

In time, these vessels will become as ubiquitous a part of the seascape as those piloted by humans. Swarms of them will patrol harbors, support military missions, conduct search-and-rescue operations, transport goods, and nearly every other maritime operation for which their presence would prove practicable. 

Before this will happen, however, they will have to be able to talk to each other and their human operators. The mission requirements they are given before deployment, and the data they collect, must be shared effectively and securely. In and of itself, such a task would pose a minimal challenge. There is no shortage of communications packages available to do all that. 

Factor in the environment in which autonomous watercraft would operate, however, and the task becomes considerably more challenging. In the most simplistic terms, saline and fresh bodies of water are inherently laden with obstacles to effective communication. Differences in salinity, density, temperature, turbidity (simplistically defined as current laden with sediment, causing it to move quickly) pose only a partial list of impingements to effective communications. 

As with other aspects of robotic operation, academia and industry are teaming to address the problem, with considerable support from government.

Underwater GPS
 
At Florida Atlantic University in Boca Raton, Florida, George Sklivanitis heads a team that is working on development of better sensing and embedded network systems. Their activity centers upon propagation of signal processing for wireless networks that must function in unconventional environments — underwater, over the air, in airports, under the ground in subterranean networks. 

Sklivanitis outlined his work during a poster session at this year's AUVSI Xponential trade show in Chicago.

“My goal is wireless communications and networks,” says Sklivanitis, a research assistant professor at FAU’s Department of Computer and Electrical Engineering and Computer Science. 

To accomplish what they are trying to do, Sklivanitis says his team must overcome a series of particular problems. 

First, the solution requires what he calls a localization of assets. 

“You might have some hazards where you’re deployed, where you don’t have access to GPS [global positioning systems]. Underwater, for example,” Sklivanitis says. “This is of significant interest to military applications.” 

His team has been focusing research on how to transmit data from one area to another via common links, which sense interference and jamming. Once it is clear where neighbor vessels and interference are located, the next step is to establish a network of assets and route data from one access point to another. 

“Those are the fundamental problems we think need to be [resolved to] achieve the dream in the marketing domain of autonomous systems,” Sklivanitis says. 

Much of the team’s recent focus has centered upon localization of underwater vehicles. Under a grant provided six years ago by the National Science Foundation, they have been attempting to develop a simple method of localizing unmanned underwater vehicles, using the least amount of known location. 

“It’s an ‘underwater GPS,’ if you may,” Sklivanitis says. “To do that, we found out in theory that we will need to [have] two acoustic, ultrasonic beacons, placed in specific locations either above the surface or underwater.” 

Once the two beacons are established, the team then can program them to transmit coded signals that manipulate the physical waveforms emanating from them. 

In civil applications, such an arrangement may be further enhanced to include a third beacon, allowing for triangulated transmissions and communications. 

“We are able to really manipulate the resolution” of vehicles’ specific locations, Sklivanitis says, “and transmit jointly in space and time.”

By using coded signals with fixed lengths, and increasing the lengths of the codes they use, Sklivanitis says they are able to increase the resolution of the localization that occurs on board each unmanned underwater vehicle.  

The arrangement entails inherent tradeoffs, Sklivanitis says. If the mission requires accuracy down to the millimeter, payload limitations would cut into a vehicle’s computation capacity. Mission demands would justify which capability gets the most attention. 

“The nice thing about the technology is it’s not software-defined. We can reprogram them on the fly while the vehicle is moving through the water,” Sklivanitis says. 

During their research, Sklivanitis’ team had to come to terms with a host of sources of interference — some natural, some man-made. Marine life such as snapping shrimp interfered with the signals picked up by the watercrafts’ microphones. To resolve the problem, the team came up with what Sklivanitis calls a “normal signal-processing tool” that is capable of analyzing the data and identifying both useful and non-useful signals. 

“We were actually watching for angles of arrival of the signal onboard the UUV, trying to robustly estimate direction from first and second signals in order to localize two known locations,” Sklivanitis says.

The process is similar to how GPS systems work with three satellites. Doing so robotically, while avoiding corrupted data inside the signals picked up by receivers, requires the use of a signal matrix that ensures identification and detection of corrupted data. 

Other unwanted signals, emanating from reflective areas in the operational environment, may give a false impression of the location from which they emanate. The signal tool Sklivanitis and his team are working on can identify these occurrences as well, he says. 

The autonomous communications project is taking place along with other side activities that address the issue of internalizing information and building robot communications links and networks.

       One such project, under the sponsorship of the Air Force Research Laboratory in Rome, New York, involves development of the next generation of unmanned airborne swarms that are always connected and can avoid interference at any point, be it physical or coming from the network. The same system would have applications underwater, Sklivanitis says. 

“They are similar in that every system has to be adoptive and autonomous, and sense and react to things they see in a wireless environment. They have to be proactive … to avoid any type of interference,” Sklivanitis says. “The difference, really, is the environment.” 

For example, because acoustics and ultrasound have very limited bandwidth underwater, the better communications medium is radio signal. The lesson here is it is best to be flexible and determine spectrum resources function best in which environments, Sklivanitis says. 

The team’s work will continue at the university’s ocean mechanical engineering center at Dania Beach, Florida, and at five satellite campus locations. 

AUV communications
 
At McLean, Virginia-based Saltenna LLC, technologists believe they have made a significant breakthrough in underwater communications that would dramatically facilitate communications among autonomous vessels in nearly any condition. 

In collaboration with the University of Maryland electrical and computer engineering department and with the support of DARPA, Saltenna also is addressing issues such as limited bandwidth and slow propagation speeds, which can be the source of communications errors. 

“Conventional radio frequency signals have potential for high-bandwidth communications but are severely limited due to rapid attenuation in seawater,” says Igor Smolyaninov, Saltenna cofounder and chief technology officer, who holds a doctorate in physics and a master’s degree in electrical engineering.

Smolyaninov also outlined his work during a session at this year's Xponential conference.

       Directional optical links can provide better bandwidth over 200 megabytes per second, Smolyaninov says. Applying directional optical telecommunications links could provide improved bandwidth. 

But until Smolyaninov and his team embarked on their current project, directional optical links had not been successfully integrated in unmanned underwater vehicles, because technology did not yet allow for sophisticated pointing, acquisition and tracking. 

“In addition, optical links are strongly affected by water turbidity,” Smolyaninov says. “Our technology, based on surface electromagnetic waves, has much higher bandwidth compared to acoustic communication.” 

As such, the Saltenna system’s technology is not affected by such factors as water turbidity, Smolyaninov says. 

The team tested their system in December 2017 in seawater at the U.S. Naval Surface Weapons Center in Panama City, Florida. The tests were conducted in separate environments — near the sea floor and in proximity to water basin borders. They used separate transmitting and receiving antennas and radio systems that were enclosed in watertight containers, with divers verifying the depths at which they were deployed. The results matched their expectations.

Besides DARPA, the Office of Naval Research is interested in using Saltenna’s radios for autonomous watercraft operation, Smolyaninov says. 

Above: Florida Atlantic University students are working toward wirelessly controlled underwater vehicles and rapidly deployable underwater inflatable ultrasonic arrays for infrastructure-less underwater localization. Photo: FAU Below: Saltenna tests its underwater antennas at Naval Surface Warfare Center Panama City Division, Florida. Photo: Saltenna