Sunday, June 11, 2023
Bringing the Latest in News Straight to Your Screen

Sea Buoys: Solid Templates In Navy’s Hyped Race For Maritime Autonomy

By News Creatives Authors , in Business , at August 31, 2021

Despite a whole lot of hype, autonomous watercraft remain developmental assets, subject to lower mission success rates and far higher loss rates than Washington’s vocal autonomy advocates care to admit. Outside of some highly classified “one-time” missions and shorter-range, more controlled “tool-like” activities, the routine, day-to-day life of a mariner at sea is still a bit too much for most robots to take. 

The autonomy gold rush is understandable. It’s the hot new thing, and every single Navy in the world wants to unlock the potential cost-saving and performance-enhancing benefits of maritime autonomy. But getting autonomous or un-crewed platforms ready for ocean-spanning naval duties demands more investment. It’s going to cost a lot. For robots to move forward in any concrete way as waterborne national security assets, autonomous technologists need bigger support networks, more prototypes, and many, many more platforms than conventional Pentagon thinkers expect. 

And that, if acknowledged, is fine.

For long-endurance autonomous vessels, America already has a good understanding of the challenge ahead. Simple buoys—the first “autonomous” maritime systems—still, after decades of design refinement, suffer high loss rates and are plagued by partial mission fulfillment. As of early August 2021, ten percent of the National Oceanic and Atmospheric Administration’s moored weather and sea condition buoys are completely inoperative, while another 25 percent currently suffer one or more sensor or system failures (you can check real-time performance here). Vandalism alone accounts for a continuous 15-20 percent drain on buoy performance. And these statistics are for simple weather-reporting stations with just a basic communication system and eight sensors apiece! Far more complex unmanned warfighting systems that strategic visionaries want to send on long missions into busy, contested or strategically important waters are likely to suffer far more wear-and-tear and exhibit far worse performance.

America’s naval tastemakers do not appear to be building this constant “loss rate” into their procurement models. Rather than be forthright about the challenge ahead, modifying the procurement approach to address the need for continual iterative development amid constant levels of platform attrition and loss, too many in the Department of Defense are—at least in models of future Navy fleets—just embracing the hype, and presenting un-crewed maritime platforms as complete, fully-functional packages, while asking America to sacrifice an enormous swath of the Navy’s existing fleet for what are, right now, just a bunch of science projects. 

When Hype Hits The Pentagon’s Procurement System:

Hype sells. Getting the Navy to embrace any new product requires a breezy blend of confidence and marketing of capabilities that are just “around the corner”. For unmanned advocates, the use case is relatively easy; unmanned aircraft (and some unmanned undersea craft) have done wonderful things, so why not move to a similar model at sea? Sure! After the operational and technical hiccups are worked through, the potential is certainly there—autonomous vessels and submarines are, at some point, going to be a far larger component of the maritime world. 

The only problem is that the technology is just not ready yet and the Pentagon is moving too slowly to make it ready anytime soon. Compounding the problem is that influential people have already drunk too much of the autonomous Whiz-Bang Kool-Aid and are making bad choices. Too few Washington, DC decision-makers understand that most of today’s unmanned watercraft only, at this point, offer unrealized potential. Some checks are in place, and a handful of widely-disparaged efforts to drive real-world land-based testing for components aboard large unmanned platforms have moved forward, but America’s defense tastemakers need to recognize that unmanned platforms require a lot more integration and real-world at sea work before autonomous platforms truly become mainstream, day-to-day defense assets. The focus must be on getting basic platforms into the fleet and at sea, so we can understand if these platforms are going to need more or less support than America’s humble buoy fleet.

Fueled by the field’s relentless boosterism, coupled with an odd willingness of those within and without the Navy to wave away inconvenient and unglamorous logistical truths, the limited capability of present-day autonomous systems to support the Navy at sea not well understood. Without procurement plans that account for continuous platform iteration, constant attrition, and support and salvage, un-crewed platforms will fail. At a basic level, to autonomy-boosters, robots that don’t have humans out there maintaining them in real time don’t need to think deeply about strategy, or account for logistics or even worry about the wear and tear of the ocean. It’s this sort of “overselling”—the drumbeat of exciting tactical achievements without the underlying context or information—that risks autonomy’s future at sea, endangering America’s national security interests. 

Autonomy at sea is at an early stage. Take Saildrone, an innovative California company that is busy marketing the military potential of a sensor-laden marriage between a sailboat and paddleboard. Saildrone’s webpage and marketing materials detail one successful mission after another. But Saildrones are still early in their technological development. A deeper dive shows that these unmanned craft are, like any other piece of maritime gear that is deprived of onboard maintenance, prone to breaking down under the weight of weather and the maritime environment. 

The headlines obscure the redundancy and support required to keep robots in the fight. In 2019—after several failed attempts—three saildrones were deployed to circumnavigate Antarctica. Two failed, and while the last remaining robot hit an iceberg, damaging the sensor suite onboard, it survived. In 2020, four saildrones were sent to complete an Arctic mapping mission for the National Oceanic and Atmospheric Administration. One failed on the way, becoming a “mobile weather buoy,” and “all but one” of the saildrones suffered “some level of equipment failure”. And with Fall approaching, the saildrone’s solar-powered sensors didn’t get enough sunlight to operate, and the mission was aborted on September 23. 

This isn’t to attack Saildrone. Saildrone is doing exactly the right things to move their innovative platforms forward—the company is constantly pushing for real-world projects, building and testing prototypes and seeing what may work. They’re transforming modest funding from any research and development project they can get into real sea time—and making great strides as they go.  

But things change as unmanned craft get more complex and enter the Pentagon’s procurement cycle. Development slows to a crawl while capability-intoxicated acquisition engineers put interesting but ancillary tactical concerns over boring operational realities and the overarching imperative to refine basic at-sea utility.

The slow progress of the General Dynamics

Knifefish unmanned undersea vehicle shows how, for a complex military platform, the iterative test and modification cycle seen in the civilian world—a cycle that Saildrone is exploiting—slows to a crawl. Increased complexity drives price, reduces the number of procured platforms, and constant capability-enhancing refits limit the lessons that only real sea-time can teach. The first Knifefish was delivered in March, and rather than rush to get the few newly-delivered Knifefish out to sea, the Navy, in May, awarded General Dynamics a 73 million dollar contract to enhance the capability of their five newly-delivered Knifefish systems. At this rate, as more and more requirements creep into the platform, it’ll be years before a Knifefish ever gets “operational”, out to sea, and into a sailor’s hands.

While the risk/reward dynamics in fielding more advanced systems are complex and often classified, the Pentagon can move faster. The focus for some platforms must shift from accountant-minded efficiency to emphasize sea time. With many unmanned platforms bound for military service, even though it may be more “cost-effective” for the Pentagon to organize work by production/modification “blocks” or “flights”, it may be better to initially focus on numbers, getting basic production models out and working at sea and chalking up operational lessons while others in the fleet are modified to the latest capability standard. Either that, or focus on keeping the production/modification lines operating at some modest and constant level, busy obtaining a sufficient number of platforms so that both the manufacturer and user can continually iterate lessons learned from at-sea use into newer robotic platforms.

Defuse The Hype And Use The Humble Buoy As Template 

To keep hype in check, the Pentagon can move to incentivize demonstrated performance. Veracity is important. Reward contractors—and the Navy overseers who accurately predict platform performance, maintenance expectations and sensor availability, while penalizing the ones whose products fail to live up to their marketing hype. In a frothy field like autonomous systems, overall contractor veracity needs to be treated as a key performance indicator, on par with Congressional influence, industrial base concerns and parochial interests.

That’s where buoys can help. Buoys may be simple, but they are the original autonomous vessels and they have quietly chalked up millions of hours of operational time. Un-hyped and almost ignored, they offer an invaluable amount of honest and simple ground truth. As such, the operational trends exhibited by buoy constellations offer the unmanned community an enormous amount of information about maritime autonomy’s likely logistical requirements and unmanned autonomous systems performance going forward. It would be foolish to ignore available performance data.

At a fundamental level, data show that buoys have high failure rates, suffer a substantial amount of sensor attrition, and suggest that strategically important buoys often suffer above-average loss rates. For unmanned craft, that means sea-time—moving as fast as possible to understand what, exactly, the sea will do to our various unmanned platforms—is going to be a critical thing. And if robotic performance aligns with America’s buoy constellations, then redundancy—in both platform numbers and in onboard systems—will need to be a major focus.

Buoys also need substantial logistical backup—often requiring their operational and technical helpers to travel out to very remote, isolated areas for extended periods. For buoys, post-breakdown salvage and recovery is often complex, requiring far more resources than expected. The same will be true of autonomous craft, and unmanned vessels with engines, munitions and complex communication gear will likely need even more overwatch, maintenance, resupply and repair support than the average meteorological buoy.

Outside of salvage, buoys suggest that more sailors and ships will be needed for robot support purposes than autonomy advocates like to admit. We’ve done this before. A few years ago, when small ships were the hot new thing in the maritime, Navy modelers would leave pricey logistical support out of their money-saving outlines of potential future small-ship heavy fleets. Again, the logistical needs of a robot fleet may be substantial; to help keep America’s various buoy constellations operational, the Coast Guard dedicates a vast portion of their fleet—upwards of fifty cutters—towards buoy tending. The Navy, as it goes in a similar direction, will need to make similar investments, but, given the Navy’s strange aversion to funding “boring” tankers, salvage ships, tenders, repair craft and other unglamorous logistical assets, the simple, hype-tinged visions of autonomy evangelists is far more appetizing than the rough reality of keeping robots in service at sea.

At sea, robots certainly offer the potential for lots of new and exciting capabilities. But that will mean nothing, if, in the race to chase new capabilities, the Pentagon overlooks the harsh operational realities facing any brave robot sent out to cruise our big and increasingly contested oceans all alone.


Leave a Reply

Your email address will not be published. Required fields are marked *