⏰ Masterclass

From AI to Action: Advanced Workflows for Smarter Risk Management.

00 Days 00 Hours 00 Min 00 Sec
Discover more

Whitepaper

The 2 Factors that Unlock the Effectiveness of Unmanned & Autonomous Systems 

Oceans are vast, there are hundreds of thousands of vessels at sea, and security threats rapidly evolve and change. With too much sea, ground, and air to cover, and limited resources available to law enforcement and defense agencies, achieving effective and actionable maritime domain awareness (MDA) is daunting.

Manned assets – in the sea and air – have been the primary way to monitor maritime activities for decades. But these missions are inherently limited in scale and duration, due to concerns about safety and cost. The recent shift toward increasingly relying upon and utilizing unmanned systems (also frequently called “autonomous systems”) for MDA has been a natural response to overcoming the constraints and limited scalability of manned missions.

Unmanned systems offer numerous advantages, but they also come with a set of challenges that may compromise effectiveness. Whether these assets can be moved around during their missions, like drones, or are stationary, like drifters, their effective usage is totally dependent upon knowing both where and when to deploy them

Additionally, autonomous systems enable greater coverage of the seas, but they also produce a sea of data that creates a major challenge.

This white paper will examine the growing dependency on unmanned and autonomous systems by national governments and defense agencies and set out to address the two central challenges of this trend: pre-mission operational planning and real-time handling of incoming data.

The Race to Cover the Most Ground

Total, 100% coverage of the seas 100% of the time is impossible. But threats are continuously shifting and emerging, demanding attention and constant monitoring: 

  • The great power competition and illegal, unreported, and unregulated (IUU) fishing in the South China Sea
  • Houthi and Iranian disruptions in the Red Sea
  • The global network of narcotics smuggling
  • Arms trafficking in the Caribbean
  • Russia’s dark and gray fleets evading sanctions
  • Oil smuggling
  • Much more… 

In light of increasing risks worldwide, national governments are shifting efforts and resources to fund unmanned capabilities to enable surveillance and monitoring.

The global unmanned systems market size is expected to grow by 15.7% from 2023 to 2032. In 2022, the market’s size was projected to reach a valuation of $16.3 billion. By 2032, the valuation is set to reach $34.16 billion.

Graph
Source: Custom Market Insights

The U.S. Department of Defense is set to invest more than $2.6 billion in unmanned aircraft systems (UAS)  in 2023, with at least 29 programs fully dedicated to the procurement of UAS. The UK is expected to spend $5.8 billion by 2024. Globally, increasing the use of unmanned systems for combat and intelligence, surveillance, and reconnaissance (ISR) missions drives the unmanned systems market growth. Defense forces are highly invested in developing and procuring autonomous systems, to reduce the risks of soldier casualties and enhance accuracy during combat operations.

Why Now?

Unmanned and autonomous systems are having a moment. They have been around, in scale, for at least two decades, yet as the data above shows, there is unprecedented momentum around unmanned capabilities. This is only partially due to their inherent advantages. The core driver of this trend is the dramatic gap between the manufacturing power of the West and East.

China is building ships faster than the U.S. and its navy is expanding faster than any other navy on the planet. The manufacturing power and pace of the East, led by China, is creating an unbalanced battlefield, which holds true for surface, airborne, and underwater vehicles. Benefits of unmanned technology aside, the propulsion of unmanned systems is mainly a numbers game. 

An arms race that focuses on hardware (who has the most drones, aircraft, or ships) is more difficult to win now than ever before for the West, given the dizzying pace of military growth in China and other Eastern countries. 

Instead of an incessant and ineffectual hardware race, strategic advantage can be achieved by the West through being smarter and more agile. A shift from trying to have the largest number of weapons and the most expensive ones, to cheaper swarms that can be deployed cost-effectively and quickly. This means incorporating technology, specifically artificial intelligence (AI), to gain an edge. 

The more hardware that is developed, the more it is understood that battlefield victory likely depends on who has the best common operating picture to manage and coordinate unmanned/autonomous and manned assets. 

Possessing the most robust data analytics tool to scan an endless churn of data from numerous feeds to make faster and more informed decisions is critical as are systems that create high-quality security leads.

The Benefits of Unmanned Technology

The comparative benefits of unmanned systems are:

  • Enhanced safety and risk reduction: avoiding putting people at risk is the most fundamental and crucial benefit of unmanned systems 
  • Operational endurance and reach: unmanned systems and assets are not reliant on supplies, can endure extreme weather conditions, and can go farther, for longer
  • Cost efficiency: unmanned systems are easier to produce and operate. Their deployment is considerably less expensive – it uses less fuel and requires less logistical support and ongoing maintenance. The lower operational costs of unmanned systems make them a more sustainable and efficient option for long-duration maritime surveillance missions
  • Scalability: autonomous systems can be built for scalability, especially when you broaden the definition to include sensors, buoys, and drifters.

Unmanned systems offer significant advantages over manned aircraft for MDA, including increased efficiency, lower risk, and lower environmental impact. As technology continues to advance, unmanned systems are also becoming more precise and reliable, especially with the introduction of AI, allowing real-time analysis and agility.

Recognizing the value of unmanned systems for MDA, many armies worldwide have adopted a “hybrid fleet” model that incorporates various unmanned systems into their operations. As this approach becomes more common, military and defense organizations confront challenges regarding the increased number of assets ready for deployment. 

Unmanned systems allow for more missions. More missions inevitably increase the cadence with which organizations must confront the question: where do we deploy? And more deployments produce a lot more data, causing government and law enforcement agencies to frantically search for the proverbial needle in the haystack.

Two Critical Factors for Maximizing Unmanned Technology

Don’t Get Lost in a Sea of Data

Unmanned systems are cheaper to produce and deploy, allowing for more of them to be operational at any given time. But more is better only if the deployment is thought out and effective. 

The first step before deploying an asset, manned or unmanned, is to decide where to deploy it. In the dauntingly expansive oceans, how do you choose where to look? Pre-mission planning is often based on historical data – deployed where illicit or threatening behavior has been known to happen before, or in known hubs of activity. The key word here is known. Deployment planning often falls into the trap of habit or inertia, rather than proactive operational planning meant to discover the unknown or to discover how the maritime ecosystem is constantly evolving.

As the number of deployed assets grows, data increases in parallel. First, there is more data about what is happening near and around these assets, and it is crucial to ensure the assets’ ongoing protection by constantly being aware of which vessels are currently near or approaching the asset. 

Second, the reason to have more operational assets is to obtain better visibility into a wider area. This also means more data – more dots on the map at any given time. This is both the objective but also the challenge – how do you transform data into information, information into knowledge, and knowledge into action?

AIS data comes with a great deal of “noise” and effective operation involves taking the data, processing it, and producing actionable insights. Effective pattern recognition and adapting to changes is difficult without a good base of historical data, as noted in our executive brief on Windward’s AI technology, but obtaining ten years of historical AIS data costs millions of dollars. 

And working with less than that puts you at risk of not having sufficient data to optimally train and refine the AI data models that have become mandatory for the complex maritime ecosystem. 

Effective Pre-Mission Planning 

Effective use of unmanned and autonomous systems relies on pre-deployment operational planning – knowing where and when to focus attention. As mentioned, oftentimes unmanned systems are sent, at best, to where they have traditionally been deployed or, at worst, based on hunches. 

An effective approach to pre-mission planning relies on accurate data and insights fueling the decision-making process. Windward’s Maritime AI™ platform can be integrated with unmanned systems to optimize strategic operations with predictive intelligence, comprehensive risk insights, and operational recommendations.

Windward assesses risk levels using open-source data and a ship’s historical data. Ships are classified into red (high risk), yellow (low risk), and blue (low risk) categories. These classifications are based on historical data, including deviations from normal economical behavior, past illegal activity, instances of turning off their transponder, and other relevant factors providing valuable insights for maritime surveillance operations.

At any given moment, there are hundreds of thousands of vessels at sea, some transmitting with their automatic identification system (AIS) while others do not. Even when focusing on a specific area of interest, such as the Persian Gulf, the sheer number of vessels makes it difficult to decide where to focus attention. The image below shows 13,450 active vessels in the Persian Gulf – too many to serve as the basis for operational decisions about where to focus attention.

Map of system
All vessels in the Persian Gulf on June 23, 2024. Source: Windward’s AI platform.

To be more effective and get a laser-focused picture of where assets should be deployed, as well as to map the threat landscape and identify targets, two steps can be taken. First, build a vessel profile to narrow down results even further. For example, the vessel’s past activities (previous port calls, deceptive shipping practices, etc.), flags of interest, or historical voyages. Second, apply Windward’s predictive behavior risk models, which can identify vessels that are more likely involved in smuggling events or illicit behavior.

Refining the search to only show vessels flagged by the Windward’s system as high or moderate risk for smuggling yields a considerably more manageable picture – showing a total of 633 vessels, 411 of them high risk. But there are still too many vessels that are too scattered to serve as the basis for a clear patrol route…

System screenshot
High and moderate-risk vessels in the Persian Gulf on June 23, 2024. Source: Windward’s AI platform.

To narrow the search even further, these high and moderate-risk vessels can be queried based on: 

  • The vessels’ flag – looking at flags of convenience, or a specific nationality
  • Previous port or port of destination
  • Behavioral history – behaviors known as likely indicators of illicit activities, such as dark activity, slow-speed, or location tampering

A known sequence of activities – for example, a port call in Iran followed by dark activity in the Persian Gulf

System screenshot
High or moderate-risk vessels that conducted ID & location tampering in the Persian Gulf in June 2024. Source: Windward’s AI platform.

The above image shows vessels marked as high or moderate risk that have conducted ID & location tampering in the Persian Gulf in June 2024. From 584 high and moderate-risk vessels, there are now only 75 vessels to inspect – and it is simple to see where in the Persian Gulf they are concentrated.

Armed with this information, it is much easier to know where to look and build an explainable, justified, and intentional patrol route based on specific targets of interest, known hubs of illicit activities, or suspicious behavioral patterns.

Creation of Common Operating Pictures

Security and defense missions often involve dozens of different agencies and organizations, and collaboration between troops or unmanned systems on the water, land, and in the air is paramount to mission success. A common operating picture (COP) that receives data from multiple sources and shares it with relevant stakeholders in real time can be as challenging to create as it is critical.

Information is power and the simple, easy, and fast sharing of information between one agency and another can be the thing on which the mission depends. 

Having yet another platform to operate during a mission is not conducive to success. Insights must be integrated directly into existing workflows and systems, ensuring that the data being collected and transmitted by the unmanned assets – and all other sources of information – is fed into and analyzed in the same space. API technology that allows for insights to be integrated directly into existing workflows and systems is more important than ever in this new unmanned era. 

Mission Management Streamlined

The sea of data generated in today’s digital ecosystem makes it impossible to see through the noise and quickly overwhelms available resources  – there are hundreds of millions of raw AIS messages transmitted daily, before adding in all the unmanned data! Simplification increases the accuracy of vessel movement mapping and makes it possible to process massive amounts of data.

Sharing data with other teams and making strategic decisions in real time is also extremely difficult without a large base of fused and clean data.  

The process of taking raw data being collected by unmanned assets, applying automatic risk assessments to prioritize investigation resources and human expertise, and yielding actionable knowledge in the form of new suspicious behavioral patterns, risk profiles, and automatic lead generation is how missions should be managed.

This is certainly not limited to unmanned missions. But the proliferation of unmanned systems and autonomous assets, and the growing reliance on them for ongoing maritime domain awareness, means that the number of missions being carried out at any given time is exponentially increasing. 

To scale up, reliance on limited human expertise and insight has to be narrowed and applied only when it is most needed. The heavy lift of prioritizing the intervention of human analysts or forces has to be abandoned, and replaced by AI-based insights.

I Want to Learn More About Unmanned Tech & Lead Gen