Controlling the Seas: Introducing a New Concept in Autonomous Surface/Underwater Vehicles

The boundary between the sea and sky is an important place to be. It’s the critical connecting layer for commercial and military information exchange between the undersea world to aerial, space and shore. Being present at this boundary between sea and sky, with cost-effective endurance in challenging conditions, requires the use of autonomous surface vehicles.

Designing long duration autonomous surface vehicles requires access to inexpensive, low-power computing; a persistent source of power; and durable mechanical engineering. The volume markets created by cellular telephones and video gaming has totally transformed the economics of computing. Coupled with creative mechanical designs that harness energy from the ocean for vehicle propulsion, and new generations of solar cells for electricity, the pervasive application of Autonomous Surface Vehicles (ASVs) is more practical. New advancements in communications, and sensor technologies are also enabling developments in ASV and Autonomous Underwater Vehicles (AUVs) while paving the way for increased coverage, rapid information delivery, increased safety and, lower cost.

Tying these all together is what Liquid Robotics has done with their Wave Glider®. They’ve developed the world’s first wave and solar powered autonomous surface vehicle that provides sustainable ocean operations and makes it possible for real time data collection and information for commercial missions such as conducting seismic surveys, environmental and water quality monitoring for oil & gas companies; measuring weather conditions and climate change; and tracking great white sharks. Leveraging these commercial technologies, Wave Gliders are used in defense missions for Anti-Submarine Warfare (ASW); Intelligence, Surveillance and Reconnaissance (ISR); and security of national resources in Marine Protected Areas (MPAs), marine sanctuaries, and Exclusive Economic Zones (EEZ). Each one of these operations requires a persistent, 24×7, monitoring and surveillance presence that is not economically or operationally feasible with manned assets.

Designing an autonomous vehicle like the Wave Glider, poses a mixture of complex technological challenges such as persistence, scale, reliability, and cost, to name a few. Add to this the challenge of operating a floating computer center with sophisticated communications and sensors in salt water, during hurricanes, and at sea for a year at a time.

So how do you provide seafloor to space surveillance across the vast, hazardous oceans? What challenges and technological advancements make the deployment of fleets of networked ASVs/AUVs, interoperating with manned systems, a reality in the maritime theater?

Creative Mechanical Design

Persistence at sea requires solving the energy re-supply challenge. A technique to harvest energy from wave motion was the key insight that made long duration missions possible. This basically gives the vehicle unending thrust for free. The vehicle can endure severe conditions through a combination of mechanical design and sophisticated materials. This gives it the ability to maintain presence for many months at a time, where most other autonomous vehicles are limited to hours or days. The Wave Glider is the only surface vessel that does not retreat when a hurricane approaches.

The Wave Glider generates thrust by being built in two parts: one part floats on the surface and the second part is below the surface where it is calm. It looks roughly like a surfboard that is covered in solar cells and antennas. It is connected by an umbilical cord to a submersed component that is a rack of wings and a rudder. As the float moves up and down in the surface waves, the wing rack, which is down where the sea is calm, moves up and down too. But its wings are mounted on hinges in such a way that the vertical oscillation is converted into forward thrust. While the Wave Glider picks up free forward thrust from the waves, it gets free electrical energy from solar panels stored in batteries and distributed through a sophisticated power subsystem. There are no fuels of any kind on the vessel. No emissions are produced. (To see a video explaining how the Wave Glider works, go to www.techbriefs.com/tv/Wave-Glider)

How a Wave Glider’s unique propulsion system works.
The selection of materials for a vessel that has to survive salt water and hurricanes for extended periods required significant engineering work. The hull is primarily made of composites. Titanium is used for many components, as is carefully selected grades of stainless steel. All of the external electrical connections are designed to be wet-mateable. The umbilical between the floating and submersed halves of the vehicle is particularly sophisticated – not only are there strength members that have to survive significant shock loads, the electrical wires that are embedded in the umbilical need to maintain continuous connectivity through arbitrary flexing and shocks. Even the paint is involved in durability through the reduction in bio-fouling.

A testament to this innovative engineering design is proven through the success and experience with long distance missions such as the journey of multiple Wave Gliders across the Pacific ocean from San Francisco, CA to Bundaberg, Australia. This scientific initiative, named PacX (Pacific Crossing), spanned approximately 400 days while traveling through a Category 5 Typhoon and overcoming the East Australian Current before arriving in Australia. This achievement was awarded the Guinness World Record for “the longest distance traveled by an unmanned, autonomous surface vehicle”.

Compute Capability vs. Power Consumption

This is an issue that is being helped by the cellular phone and tablet industry. ARM-based multicore CPU chips from vendors such as NVIDIA and Qualcomm can scale available compute resources based on workload, thus reducing power consumption to minimal levels. This type of dynamic CPU technology can be utilized to run basic vehicle navigation on minimal CPU power, but allow the CPU to scale up when data needs to be transformed into information to reduce communications overhead or on-demand onboard computational analysis. The concept of having the ASV’s control application adjust the number of online CPU cores and maximum CPU frequency to limit power consumption is a viable solution today.

Open Software Operating Environment

The ASV and/or AUV’s operating environment is becoming more sophisticated in order to satisfy increasing mission complexities. In the past, one might elect to use a real-time operating system (or develop one from scratch) and custom application. However, today there are more options including Linux and Java, both of which offer a rich set of capabilities, security, and reliability. The use of readily available open software platforms, tools, and languages aid the development of applications and sensor integration. Additionally, utilizing Linux and Java eases the task of finding qualified engineering resources, as the talent pool is larger and more current.

A Wave Glider Autonomous Surface Vehicle (ASV) awaiting deployment.
The ability for the autonomous vehicles to easily adapt to specific mission requirements drives the need for flexibility in the operating environment that can be custom tailored without major modification. A pluggable architecture allows new network interfaces, sensors, and navigation methods to be developed while leaving the core of the operating environment intact. Pluggable components tend to be smaller in size, which makes on-mission modifications possible without the need to recover and service the ASV/AUV. One of the special features of the plugin facilities in Java is that it allows for dynamically loading new software on-the-fly in the midst of a mission, while protecting the integrity of the core system software from bugs in plugins. Liquid Robotics’ Regulus, the control software on the Wave Glider, makes extensive use of these facilities.

These dynamic & flexible software environments aren’t just for adapting to mission changes. When an autonomous vehicle is on a mission lasting months, far away from human assistance, software updates to fix recently discovered bugs or adapt to failures can help dramatically. For example, on a recent mission a Wave Glider’s compass failed because it got too close to the North Magnetic Pole. Specialized software was written to “fake” the existence of a compass using the onboard GPS, and this was uploaded over a satellite link, saving the mission.

Autonomy

Today's missions require autonomy, not only at the single vehicle level, but also vehicle-to-vehicle. Imagine 100 to 500 vehicles working together as a set to collect data and/or to cover large swaths of the ocean. With good onboard autonomy, humans can concentrate on the strategic mission of the fleet, rather than the moment-by-moment tactics of each individual vessel.

Behaviors can be much more complex, fusing many kinds of sensors. For example, chemical sensors and cameras can be integrated to follow the edge of an oil slick.

Perhaps even more complex than autonomous navigation is autonomous vehicle health management. A persistent autonomous vehicle has to take care of itself. It has to detect failures, report them, fail-over when redundant systems are available, and sometimes, even attempt to recover the function of the device. It’s remarkable how often power cycling revives an ailing sensor or sweeping a rudder back-and-forth can clean obstructions from bio-fouling. Handling these issues is a significant part of the Regulus operating software.

Interoperability

As unmanned systems become more pervasive, interoperability between manned and unmanned systems is a critical capability. A fleet of AUVs does not necessarily have to be homogeneous. It should be an interconnected collection of ASVs, UAVs, and manned surface vehicles using the best of breed in each area. The utilization of standardized software interfaces make communications and integration of heterogeneous systems much more cost-effective and offer reduced power consumption, increased reliability, and mission agility.

Imagine a fleet of ASVs monitoring an area of ocean, searching for targets of interest and then sending target location, target type, pictures, and video of the target back to shore for analysis as they are acquired. Now add the ability for ASVs to coordinate with unmanned and manned assets to collect, analyze, and report a more complete, real time situational awareness to command headquarters.

The SHARC (Sensor Hosting Autonomous Remote Craft) is a special version of the Wave Glider designed for the defense industry.

As the design and use of ASVs/AUVs advances, the industry must drive forward with open standards to reduce the complexities of software development across multiple interoperating platforms. The use of standard operating systems and development languages is a move in the proper direction to foster more inter-company cooperation.

Sensor Technology

ASVs and AUVs require smaller, lower power sensors to achieve long mission durations. There are many good sensors available now, but some have too high a cost in power consumption and space. Not many radar units will fit in the palm of your hand and only draw one watt of power. Further advancement in the miniaturization of sensors is still needed. Advancements in sensor technology coupled to implementation of standards based operating environment with open APIs is needed to propel sensor integration and application development for ASVs/AUVs.

Inter-Vehicle Communications

A mixture of communications devices is required to manage telemetry, command, and inter-vehicle communications. Global satellite coverage is attractive but comes at a high cost and can be bandwidth limited. Cellular communications is fast and available close to shore in some areas. Multiple communication channels are required to serve the needs of autonomous vehicles.

Small, unobtrusive, and capable of collecting and communicating large amounts of data inconspicuously, the Wave Glider ASV could have numerous defense applications.
One of the distinguishing characteristics of the Wave Glider is its flexibility in accepting a wide variety of communications technologies such as Iridium, BGAN, Wi-Fi and cellular, and automatic switching from one device to another as circumstances change. For example, when a Wave Glider is close to shore it automatically switches from expensive and slow satellite communication to faster and cheaper cellular communication. As with the other electronics and sensor payloads, these are protected from the harsh environments by being placed inside watertight compartments with wet-mateable connectors leading to antennas on one of the masts.

There are hybrid solutions that employ several communication techniques at once: acoustic to communicate to subsurface assets, radios to aircraft and satellites, and cellular telephones to shore. By doing this, Wave Gliders can function as bridges that connect the sea floor to the shore or to aerial assets.

Navigation for Subsurface Vessels

Safe vehicle navigation is another challenge. Vehicles on the surface can use GPS to determine position, but for underwater vehicles, other methods such as a magnetic compass and/or Attitude and Heading Reference System (AHRS), are all that are available. But these dead-reckoning techniques are notoriously error-prone. By coupling with a persistent surface vessel carrying an acoustic modem and underwater position sensor, underwater assets can be given a firm frame of reference and communication path.

Conclusion:

The boundary between the sea and the sky is an important place to be. By employing modern technologies, ocean-going autonomous vehicles like the Wave Glider can provide critical information exchange from subsea to space, adding an important extension to modern national security operations.

This article was written by Dr. James Gosling, Chief Software Architect, and Mr. John Weeks, Distinguished Member of the Technical Staff, Liquid Robotics, Inc. (Sunnyvale, CA). For more information, Click Here .