SEE US AT...

Issues Archive

January/February 2018 Vol. 22
No. 1
 Confined area survey  
  leveraging multiple  
  vehicles operating as one  
  entity. Image: Teledyne  
  Marine 
 Teledyne  
  Oceanscience's  
  Z-Boat. Photo: Teledyne  
  Marine 
 A Marine Advanced  
  Research WAM-V USV  
  configured with a  
  Teledyne SeaBotix ROV.  
  Photo: The Maritime  
  Alliance 
 Z-Boat and WAM-V with  
  integrated ROV  
  coordinating efforts  
  during a routine  
  shallow-water survey.  
  Photo: Teledyne Marine 
 WAM-V platform configured  
  with Plank Aerosystems  
  Shearwater UAV. Photo:  
  The Maritime Alliance 
 An unmanned System of  
  Systems approach. Image:  
  Teledyne Marine 
Cover story: Improving unmanned vehicle  
  collaboration
  


Vitad Pradith, Matt Burdyny and Sean Newsome, Teledyne  
  Marine, USA
  

An exciting opportunity now exists for the unmanned community to cooperate and develop not only better and more robust unmanned systems, but fully autonomous systems that can work together toward a common task

The current state of the unmanned vehicles industry can be best summarised by the domain that they operate in: underwater, on the surface, or in the air. Although advances in unmanned technologies have increased the robustness and effectiveness of these vehicles, these gains typically only apply to their respective community (e.g. advances in acoustics that primarily benefit the underwater environment). This is not the fault of industry, but rather it is an accepted progression based on our understanding of the laws of physics that guide engineers to design vehicles based on our current assumptions of the natural world. This progression has created a diaspora of unmanned vehicles often incapable of communicating with one another and therefore requires the innate ability of humans to shepherd these tools to tackle a task or provide a solution toward an outcome.

To further clarify the issue: there is no intercommunication or interplay between the vehicles outside of their domain, meaning, an unmanned/autonomous underwater vehicle (UUV/AUV) operates as a completely independent system from an unmanned/autonomous surface vessel which may be in close proximity to the UUV/AUV. The lack of spatial awareness highlights the technical gap in the integration and communication between the vehicles that inhibit the use of these systems toward solving real world problems or to perform increasingly complex operations. The present debates on these vehicles are no longer on the viability of using these platforms, but rather how to transition them from unmanned to autonomous systems that can work together. For that reason, a “System of Systems” approach is needed to manage the different vehicle platforms allowing them to perform in concert with one another. In many cases, users are attempting to solve various tasks across domains in a seamless and transparent way.

The term System of Systems is not a new concept and has been prevalent in the military vernacular for quite some time. The application of the term in the commercial domain however is relatively new and is currently being socialised into the commercial industry to tackle industry specific problems. The purpose of this article is to describe the potential of collaborating vehicles and our progression toward autonomy along with the potential impact to the industry using scenarios that stand to benefit from these advances.

A PARADOX: TAKING THE MANNED OUT OF UNMANNED

Currently, operations that utilise unmanned vehicles (underwater, surface and aerial) are resource intensive. The number of operator(s) dedicated per vehicle could be best represented as an n:1 ratio where n < 1 (i.e. requires more than one operator at all times). This model inhibits the unmanned vehicle’s full potential as a cost effective tool. This cost is not only associated with the economics of operating these systems but also diminishes the promise of increased safety of operations.

Although the increased use of unmanned systems has greatly reduced the number of personnel involved in tasks associated with them, all of these vehicles require personnel to deploy and retrieve, monitor and intervene and, depending on the vehicle, manipulate the onboard sensors during operation. These processes are onerous to the participants and do not factor in the added complexity of monitoring their own interactions with other vehicles that may be operating in the vicinity. Moreover, unmanned vehicle operators may not appreciate the characteristics of how different vehicles respond in an unfamiliar domain. For example, an UUV/AUV loitering versus a tethered remotely operated vehicle (ROV) hovering and holding station underwater. This human choreographed operation can introduce a level of complexity and uncertainty that creates an awkward dance between the platforms and their operators. This discussion also recognises the ongoing development of “modular autonomy” systems that bolt onto existing manned vessels and is not included with the autonomous/unmanned surface vessels (ASVs/USVs) scope of this article.

In a recent real world demonstration, the Maritime Alliance’s BlueTech Week event in San Diego, California, USA, (November 2017) provided a promising glimpse of what a coordinated effort between various unmanned systems could look like. This demonstration scenario utilised a Teledyne Oceanscience Z-Boat USV performing a routine hydrographic survey within a shallow-water area of a port facility (and simultaneous LiDAR) around critical infrastructure. The USV detected an anomaly during the survey and reported the location of the anomaly in real-time for further investigation and interrogation to another fleet of awaiting USVs provided by Marine Advanced Research’s (USA) WAM-V platform. Each of the WAM-V USV platforms are configured with either a Teledyne SeaBotix ROV or Plank Aerosystems, USA, Shearwater unmanned aerial vehicle (UAV) and responded to the anomaly providing an aerial overview of the area via the UAV while the second WAM-V holds station deploying an ROV to perform a close inspection and interrogation. This scenario could have easily adapted an unmanned/autonomous underwater vehicle for complete littoral coverage and accommodate the need to monitor an area with deeper waters. In full disclosure, many of these vehicles were coordinated with multiple humans operating in the background for this demonstration; however, the ability to operate, adapt and coordinate these vehicles in real-time during a live scenario represented the status of leading-edge technologies. The enthusiasm generated from this demonstration provided a tangible look into the possibilities of these vehicles as a coordinated system to solve and adapt to a particular task.

Another potential application tackles the inspection of confined space areas. These inherently dangerous operations stand to gain from continuing improvements related to unmanned systems. Leveraging the different vehicle systems as one system provides better actionable information to the operators allowing them to effectively respond as a coordinated unit.

ESTABLISHING A COLLABORATIVE ENVIRONMENT

Current vehicles operate as singular entities based on predefined algorithms (deterministic). Most vehicles are programmed with a rigid control systems architecture that is only capable of executing a single task, making it difficult to adapt during an active mission. The gap in artificial intelligence of these vehicles is not yet sufficient to automatically recognise and report on anomalous events nor are they connected or networked in such a way to provide this information heuristically in a fluid manner that is transparent to the operator. The aforementioned scenario(s) however do highlight the industry’s transition toward the integration and coordination of multiple vehicles to execute a common mission and represent forward progress.

IMPROVING COLLABORATION

Looking into the future, the unmanned community will require an operational ratio of n:n (i.e. an elastic number of human operators to a larger number of vehicles that may be added/subtracted at any time). In order for these platforms to be effective, there will need to be an emphasis on the overall command of the system (i.e. a system of multiple vehicles across domains) and less on the control of the individual platform. Much of these advancements will require participation and cooperation amongst academia, public/ private entities and industry partners. Undoubtedly, all of the participants presently engaged in this community will appreciate the complexity that a System of Systems approach entails; however, the community should also be engaged and amenable to partnerships across all fields of expertise.

Collaboration should also extend to other industries. The automotive community, for example, is arguably the leading industry focused on improving vehicle autonomy. Definitions on the levels of autonomy (Levels 0: Basic to 5: Fully Autonomous) have been clearly defined by the United States Department of Transportation’s National Highway Traffic Safety Administration (NHTSA) and provide a common framework for the automotive industry to move forward and could perhaps be applied to the maritime industry as a potential model. For instance, one such definition could be applied contextually to the current status of unmanned vehicles in the maritime domain:

“Level 1: This driver-assistance level means that most functions are still controlled by the driver, but a specific function (like steering or accelerating) can be done automatically by the car.”

And for the few organisations leading the innovation into the maritime industry, the following definition might apply:

“Level 2: At least one driver assistance system of ‘both steering and acceleration/ deceleration using information about the driving environment’ is automated, like cruise control and lane-centring. It means that the ‘driver is disengaged from physically operating the vehicle by having his or her hands off the steering wheel AND foot off pedal at the same time’, according to the SAE. The driver must still always be ready to take control of the vehicle, however.”

ONE SYSTEM TO RULE THEM ALLl

The opportunity now exists for the unmanned community to cooperate and develop not only better and more robust unmanned systems, but fully autonomous systems that can work together toward a common task or as an entity that can adaptively respond to situational changes. The aforementioned demonstration between Teledyne, Marine Advanced Research and Planck Aerosystems represents a glimpse into this exciting future and will undoubtedly lead to new innovations that will redefine current operational paradigms.

About the authors

Vitad Pradith is global technical sales manager at Teledyne Oceanscience; vitad.pradith@teledyne.com

Matt Burdyny is Teledyne Marine’s director of business development; matt.burdyny@teledyne.com

Sean Newsome is development manager Middle East for Teledyne Marine Business; sean.newsome@teledyne.com

 

 

 

Back Issues

Article Search

© 2019 Underwater World Publications Ltd

Privacy Policy

Online Dispute Resolution

W3C XHTML Validated W3C CSS Validated Visit www.justice-roche.com