New videos released in ‘A Method for Ethical AI in Defence’ series

Interviewees from the TAS CDLE video series

In a newly released series of videos on ethical AI, Trusted Autonomous Systems explores responsibility, governance, trust, law, and traceability for robotics, autonomous systems, and artificial intelligence.

These videos were produced by Trusted Autonomous Systems for the Centre for Defence Leadership & Ethics (CDLE) at the Australian Defence College.

The videos feature Chief Defence Science Prof Tanya Monro, ADF personnel, and representatives from Defence industries.

They explore topics of responsibility, governance, trust, law and traceability using a hypothetical science fiction scenario Striking Blind written by Australian Defence College Perry Group students in 2021. In the Striking Blind story, an Australian autonomy platform is deployed in a future operation with a fictional AI called ‘Mandela’.

The videos explore ethical and legal factors associated with this scenario. They highlight the need for maintaining robust oversight so that the ADF can benefit from AI and autonomous systems while addressing their complex challenges.

Designed for the full Defence learning continuum, the videos are based on the framework and pragmatic tools described in the Defence Science technical report A Method in Ethical AI in Defence (2021).

Screenshots from the animations produced in the video series

Using the framework the videos cover:

While the method does not represent the views of the Australian Government, it provides an evidence-based collaborative framework relevant to Australian Defence contexts of use as well as ethical and legal considerations aligned with international best practice.

These videos provide both an overview and an in-depth exploration of A Method for Ethical AI in Defence and can be used within professional military education and by external stakeholders of Defence, including academia.

How to use the videos
  1. Use animations as thought prompts within presentations and workshops on robotics, autonomous systems, and artificial intelligence amongst Defence stakeholders.
  2. Use longer videos in a ‘flipped classroom’ model of learning for professional education and training
  3. Use videos in multi-stakeholder meetings to establish a shared framework within which to identify ethical and legal risks for robotics, autonomous systems and artificial intelligence projects for Defence
  4. Use pragmatic tools videos to establish processes for the identification and management of ethical and legal risks on RAS-AI projects
How to cite the videos

Producer Tara Roberson (Trusted Autonomous Systems)

Creative Director Kate Devitt (Trusted Autonomous Systems)

Publisher Trusted Autonomous Systems

Production Company Explanimate

Sponsor Centre for Defence Leadership & Ethics, Australian Defence College

With thanks to all interviewees who appeared in the videos: Stephen Bornstein, Damian Copeland, Kate Devitt, Michael Gan, Chris Hall, Sean Hamilton, Lachlan Jones, Rebecca Marlow, Tanya Monro AC, Mick Ryan AM, Lauren Sanders, Jason Scholz & Dominic Tracey.

Cite as: Roberson, T. & Devitt, S.K. (2022). Ethics of Robotics, Autonomous Systems and Artificial Intelligence Videos for Defence. [14 Videos] Trusted Autonomous Systems. https://tasdcrc.com.au/ethical-ai-defence-videos/

The transcribed video series is available on the TAS website.

How to Plan a Successful Autonomous Systems Demonstration

Australian Droid & Robot demonstrating at EPE MILTECS Facility at CSIRO and Queensland Defence Science Alliance ‘Human Teaming and Response Robotics Standardised Testing, Evaluation and Certification Interactive Showcase’ Pullenvale 21 April 2022

Australian Droid & Robot demonstrating at EPE MILTECS Facility at CSIRO and Queensland Defence Science Alliance ‘Human Teaming and Response Robotics Standardised Testing, Evaluation and Certification Interactive Showcase’ Pullenvale 21 April 2022

Demonstrating the viability of artificial intelligence (AI) requires thoughtful construction and communication of both social and technical aspects.

Demonstrations are both scientific experiments and social events designed around achieving buy-in for the technology. When it comes to AI, demonstrators face the challenge of conveying the smarts of the system and the role of human intent.

Two challenges of demonstrating the potential of AI are:

  • Performing decision-making: working out how best to show cognitive and social decision making through complex autonomy demonstrations in a way that makes intelligent performance and errors understandable
  • Showcasing ability to abide by human values and intent: working out how best to design a demonstration that shows the ability of a system to abide by decision making norms, including commander’s intent, military objectives, and ethical, legal, and safety-focused frameworks

Demonstrations are performances that include social and technical elements, such as: actors (humans, UI, software, networked systems and machines); enabling technologies (mechanics, comms, screens, instructions, consoles, connectivity, controls, batteries/fuel, security, lights, tents, generators); rituals (cultural behaviours, safety processes, signals); a choreographed narrative (CONOPS, narrative flow including objectives, cause and effect, conflict and emotions); planned, opportunistic and incidental interactions (stories, networking); evaluative criteria (expectations, key performance indicators); and goals (training/education, buy-in, positive emotions, media and communication outcomes, future investments, lessons learnt).

The TAS Ethics Uplift Program is conducting a research project that will develop and test an Autonomous Systems Demonstration Canvas to help optimise human understanding and buy-in for technology developers and investors. The Canvas will help tackle the opacity problem facing AI, for instance the way AI can obfuscate the rules, reasoning, and justifications underlying human-machine decision making as it occurs.

The first iteration of the ‘Autonomous Systems Demonstration Canvas’ V.1.1 (ASDC) was launched at the Queensland Defence Science Alliance (QDSA) Human Teaming and Response Robotics Standardised Testing, Evaluation and Certification Interactive Showcase at CSIRO’s Queensland Centre for Advanced Technologies’ MILTECS facilities on 21 April 2022.

The Canvas provides a scaffolding framework to help technology developers efficiently prepare, practise, and perform impactful technology demonstrations to diverse stakeholders to achieve diverse goals including attracting investment, showing technical progress, and getting publicity.

The intention is for the Canvas to inform future trials as a tool to support impactful demonstration planning for TAS programs. For more information on the Autonomous Systems Demonstration Canvas contact Dr Kate Devitt kate.devitt@tasdcrc.com.au.

Release of the Australian Code of Practice for the Design, Construction, Survey and Operation of Autonomous & Remotely Operated Vessels

By Rachel Horne, Assurance of Autonomy Activity Lead, TAS

Trusted Autonomous Systems (TAS) has released the Australian Code of Practice for the Design, Construction, Survey and Operation of Autonomous & Remotely Operated Vessels, Edition 1 (‘Australian Code of Practice’).

The Australian Code of Practice provides a best practice standard tailored for autonomous and remotely operated vessels operating in Australia. It is a voluntary standard, developed in close consultation with the Australian Maritime Safety Authority (AMSA), intended to be used to support assurance, accreditation, and safe operations.

Access the Australian Code of Practice and Guidance Materials via the TAS website.

The development of the Australian Code of Practice was informed by an analysis of existing, publicly available codes and guidelines for autonomous and remotely operated vessels[1],  significant stakeholder engagement, and public consultation.

TAS, supported by Australian Maritime College Search, have also developed a suite of Guidance Materials to support the use of the Australian Code of Practice. These Guidance Materials explain how the Australian Code of Practice fits into the existing Australian maritime regulatory framework, how to use the Code, and what the requirements are for each category of vessel, together with providing examples and suggestions on where to access further information.

TAS encourages owners, operators, surveyors, regulators and other users of the Australian Code of Practice and Guidance Materials to provide feedback to TAS, to help inform future iterations.

Where can I get more information?

To access more information on the Australian Code of Practice, you can:

 

TAS would like to thank all parties who contributed to the development of the Australian Code of Practice, including particularly Maaike Vanderkooi of Vanderkooi Consulting who led the development of the Code on TAS’s behalf, Rob Dickie of Frazer Nash Consultancy who led the COLREGs project on TAS’s behalf, together with his team Marceline Overduin and Andrejs Jaudzems, and Chris White from AMC Search who led delivery of the Guidance Materials, together with his team Reuben Kent, Damien Guihen, and Nick Bonser.

This project received funding support from the Queensland Government through Trusted Autonomous Systems (TAS), a Defence Cooperative Research Centre funded through the Commonwealth Next Generation Technologies Fund and the Queensland Government.


[1] UK Code of Practice for Maritime Autonomous Surface Ships, the LR Code for Unmanned Marine Systems, and DNV GL’s Autonomous and Remotely-operated Ships Class Guideline

Release of COLREGs Operator Guidance Framework

By Rachel Horne, Assurance of Autonomy Activity Lead, TAS

TAS and Frazer-Nash Consultancy have developed a COLREGs Operator Guidance Framework to make it easier to understand and comply with International Regulations for Preventing Collisions at Sea (COLREGs) when operating autonomous and remotely operated vessels. This Framework is available for standalone use, or as an annex to the new Australian Code of Practice for the Design, Construction, Survey and Operation of Autonomous and Remotely Operated Vessels.

The COLREGs Operator Guidance Framework translates the stated and unstated capabilities described, and the terminology used, in COLREGs into a vocabulary and format that makes sense for autonomous and remotely operated vessels. It is intended to be an enabling framework to:

  • Help vessel designers understand what capabilities COLREGs requires vessels to have;
  • Help operators understand what capabilities COLREGs requires and how mission planning can mitigate or remove the need for solving some of the more complex elements of COLREGs; and
  • Help regulators apply a consistent methodology for assessing the capability of a vessel with regards to COLREGs.

The COLREGs framework

Download the COLREGs Operator Guidance Framework on the TAS website

The intent is that the information gathered using the COLREGs Operator Guidance Framework will be used to inform regulatory approval processes and operational planning.

The COLREGs Operator Guidance Framework is currently presented as a PDF, which is best used printed in A3. It is also being converted to a digital tool in a collaboration between TAS, Frazer-Nash, and Aginic over the coming months.

Using the COLREGs Operator Guidance Framework

The recommended use of the COLREGs Operator Guidance Framework for an operator with a specific vessel and proposed operation in mind is as follows:

  1. Download and print out the COLREGs Operator Guidance Framework in A3 colour
  2. Download and fill out the Design Record Template, to ensure you have documented the capabilities of your vessel
  3. With your vessel particulars and the details of your proposed operation in mind, review the framework document, reading from left to right, and identify:
    1. When each rule in COLREGs applies (i.e. some only apply in specific contexts like when in Narrow Channels)
    2. The capabilities required to comply with each specific rule, broken down into the categories of Sense and Perceive, Decide, and Act (noting that these could be in the vessel, the control centre, or a combination)
    3. Mission constraints that could be implemented if you don’t have the capabilities to comply with a specific rule, to remain in compliance (for example, if you don’t have the capabilities needed to comply with Rule 9 – Narrow Channels, you may plan to avoid narrow channels, and therefore remain in compliance with COLREGs)
    4. The suitable method of compliance for each rule (for example, for Rule 5 – Lookout, the proposed evidence of compliance is Design Checklist and Simulation)
  4. Review your analysis, and prepare for own records a list of applicable rules for your vessel and proposed operation, corresponding required capabilities, any operational limitations that need to be imposed, and the recommended evidence type. You may then wish to provide your filled out Design Record and your analysis against the COLREGs Operator Guidance Framework to AMSA to support your application for exemption and/or certification. You can also review it when conducting operational planning to ensure you remain COLREGs compliant.

Further guidance materials, examples, and an instructional video will be released to support the use of the COLREGs Operator Guidance Framework in the coming months.

Further information

Background information on the project to develop the COLREGs Operator Guidance Framework, including the process the team used, is made available in a Briefing Note prepared by Frazer-Nash.

Next steps

TAS will be working with Frazer-Nash and Aginic to develop a digital version of the COLREGs Operator Guidance Framework. TAS intends to make this digital version available through RAS-GATEWAY, a new online portal for assurance and accreditation information and support for autonomous and remotely operated vessels.

The TAS RAS-Gateway project is creating a digital hub to support the Australian autonomous systems sector, including operators and the testing and evaluation ecosystem. The Gateway will feature new methods, policies, practices, and expertise to support accreditation. It aims to address issues currently experienced by regulators, insurers, and technology developers by, for instance, filling gaps in standards and producing consistent (yet flexible) parameters for safe and trusted operations and improved agility to meet fast-changing technical and social licence needs.

In parallel with this digital development, the COLREGs Operator Guidance Framework will be tested through a trial at the Reefworks testing range in Townsville later in 2022.

TAS welcomes feedback on the COLREGs Operator Guidance Framework to info@tasdcrc.com.au

 

TAS would like to thank all parties who contributed to the development of the COLREGs Operator Guidance Framework, including particularly Rob Dickie, Marceline Overduin and Andrejs Jaudzems of Frazer-Nash Consultancy for their smarts and creativity in identifying the best way to turn an idea into a tangible enabling framework, and then doing the hard analytical, excel, and design work to make it happen.

This project received funding support from the Queensland Government through Trusted Autonomous Systems (TAS), a Defence Cooperative Research Centre funded through the Commonwealth Next Generation Technologies Fund and the Queensland Government.

 

 

Results of public consultation on the draft Australian Code of Practice for the Design, Construction, Survey and Operation of Autonomous & Remotely Operated Vessels

By Rachel Horne, Assurance of Autonomy Activity Lead, TAS 

Public consultation occurred on the draft Australian Code of Practice for the Design, Construction, Survey and Operation of Autonomous & Remotely Operated Vessels, (‘Australian Code of Practice’) between 15 November 2021 and 15 December 2021.

Background information on the development of the Australian Code of Practice and the public consultation process is available on the TAS website.

TAS received seven written submissions from a diverse range of stakeholders, include SMEs developing vessels, government departments and Recognised Organisations. TAS thanks all stakeholders for taking the time to review the draft Code and make submissions.

The submissions received were considered, and further advice was sought from third parties assisting with the project where needed, to determine where changes were required to the Code. Examples of the changes made to the Code post-consultation include:

  • the accuracy of sensors is now required to be determined and declared, and their performance is required to be monitored. This will help to ensure that vessels do not operate in conditions where the sensors are not sufficiently effective, or when sensors cease to be sufficiently effective;
  • the control system must now be able to be disabled and isolated to allow for inspection and maintenance activities;
  • for survey-exempt vessels and vessels in survey, the risk assessment of any novel system must now be reviewed by an accredited marine surveyor or Recognised Organisation. A note has been added which provides that review by a competent person may be sufficient for a survey-exempt vessels where the vessel, due to its size, speed and shape, poses a very low risk to the safety of persons and other vessels should a failure occur;
  • for survey-exempt vessels and vessels in survey, tests or trials must now be witnessed by an accredited marine surveyor or Recognised Organisation. A note has been added which provides that a competent person witnesses the tests or trials may be sufficient for a survey-exempt vessels where the vessel, due to its size, speed and shape, poses a very low risk to the safety of persons and other vessels should a failure occur; and
  • improved alignment of the Code with the AMSA Guidance Notice – Small unmanned autonomous vessels, including changing the guidance on the operational speed permitted for survey-exempt vessels from 12 knots to 10 knots.

A Consultation Feedback Report was prepared which summarises the consultation undertaken, the responses received, and the outcomes.

The Consultation Feedback Report is available to download.

Once the necessary changes were made to the Code, the updated draft was provided back to AMSA for further review, before confirming it was ready to be finalised as Edition 1.

TAS welcomes ongoing feedback from users of the Code, which will support further future iterations and improvements.

For further information please contact us at info@tasdcrc.com.au.

 

 

iDrogue – Royal Australian Navy Capability – Autonomous Systems

Through disruptive innovation, Warfare Innovation Navy (WIN) Branch enables the Royal Australian Navy to be at the forefront of asymmetric warfighting for joint integrated effects. The iDrogue project, initiated by Trusted Autonomous Systems, led by Ocius Technology, and funded by WIN Branch, was established to develop and demonstrate a novel Autonomous Underwater Vessel (AUV) launch and recovery system. Ocius, a leading Australian innovator, is partnered with the Australian Maritime College and University of New South Wales on this exciting project. This pilot project is being conducted over 12-months, through 2022.

  • The ultimate aim, with further funding, is to develop an intelligent robot based on biomimicry that can launch and recover ‘any AUV, from any platform in virtually any sea state’.
  • AUVs are in increasing use by modern navies. The current method of launching and recovering AUVs is undertaken by humans at the sea surface level.
  • This pilot program will exploit advanced robotics and autonomy to undertake functions at calm depth and without human involvement. In the next 6-months the iDrogue will be automated and the design reviewed.

___________

  • This project contributes to RAN sea superiority with a capability that integrates with current and future fleets and allied capabilities.
  • The graphics on the stand represent human machine teaming and human control.
  • It is an industry led (Ocius) project funded by WIN Branch and overseen through Trusted Autonomous Systems.
  • Ocius partners include AMC Search, UNSW and Southern Ocean Subsea (SOSUB).
  • WIN – Through disruptive innovation, Warfare Innovation Navy (WIN) Branch enables the Royal Australian Navy to be at the forefront of asymmetric warfighting for joint integrated effects.
  • Ocius Technology have developed a range of uncrewed platforms. More on their range is available here.
  • Trusted Autonomous Systems (TAS) were established though the Next Generation Technologies Fund (NGTF) to accelerate autonomous systems development for Defence. The TAS vision is ‘Smart, small & many’ and projects cover all domains.

Advance Queensland funds new Trusted Autonomous Systems Projects

Queensland’s robotics, artificial intelligence and autonomous systems sector has been boosted thanks to significant funding from the state government.

The Advance Queensland – Trusted Autonomous Systems (TAS) grant has been awarded to three game-changing and researched-backed projects.

These industry-led projects will increase the state’s capacity to build robotics, autonomous systems and artificial intelligence hardware and software.

Two of the projects involve the preservation of cultural art and language in Indigenous communities in north Queensland.

TAS CEO Professor Jason Scholz said its ongoing partnership with the state government highlighted both organisations’ continued leadership in drone and AI technology to grow small businesses in regional areas.

Professor Scholz said, “TAS and Advance Queensland is investing in RAS-AI projects with industry, non-profit organisations and universities to develop data and AI project methodologies for secure and trusted AI.”

TAS Chief Scientist Dr Kate Devitt said, “The technologies and methods developed, as well as the AI governance mechanisms applied, would place the state at the forefront of international best practice.”

The first project, led by KJR with the Western Yalanji Aboriginal Corporation, will work on human-machine teaming to identify and protect high-value cultural assets. Other partners on the project are Athena AI, Emesent, Flyfreely, MaxusAI, the Australian National University, University of Queensland, and Griffith University.

The second project is a collaboration between Revolution Aerospace and the Queensland University of Technology. The team will work on a low-cost cognitive electronic system hosted on an Uncrewed Aerial Vehicle (UAV).

The third project from Pama Language Centre and the University of Queensland, will focus on developing AI and automation in language technologies, with speech communities and providing training.

The grants were awarded after an extensive competitive process. Each will run for two years, until December 2023.

Project details

Human-machine teaming to identify and protect high value cultural assets

KJR and partners (Western Yalanji Aboriginal Corporation, Athena AI, Emesent, Flyfreely, MaxusAI, World of Drones Education Pty Ltd, and Griffith University) will develop a secure multi-platform human-machine teaming capability in Queensland through using semi-autonomous drones for data capture and machine learning for image classification to identify and protect Western Yalanji rock art.

Cognitive Payloads for Small UAV 

Revolution Aerospace and Queensland University of Technology are teaming with Queensland University of Technology (QUT) to build a low-cost cognitive electronic system hosted on an Uncrewed Aerial Vehicle (UAV).

AI and automation in language technologies: securing Queensland’s data sovereignty.

Pama Language Centre (PLC) and Janet Wiles, Ben Foley, and Ben Matthews at UQ will collaborate on a series of projects with speech communities. They will be designing, developing and evaluating new language technologies including a digital asset manager for language resources, tools to support sharing of augmented reality assets, and workshops to build capacity and resource creation. It will also extend the ARC Centre of Excellence for the Dynamics of Language (CoEDL) Transcription Acceleration Project (TAP) application of ‘transfer learning’ for speech recognition, aimed at reducing the training data and time needed to develop novel speech recognition systems. An example of an existing PLC project utilising Augmented Reality is available here.

Building QLD capability

The industry-led projects will build robotics, autonomous systems and artificial intelligence hardware and software. They will achieve:

  • Transdisciplinary research impact and sovereign capability in smart, resilient, and deployable systems in congested and variable data environments.
  • Next generation AI/Machine Learning (ML) methods and apply research into deployable systems
  • New models of data governance, data sovereignty, and assurance of ML pipelines
  • Technology integration to build trusted autonomy
  • Best practice ethical AI through participatory design
  • Digital regulatory approvals

Outputs from the projects will include:

  • AI Integration, augmented reality, advanced signal processing, AI classifier acceleration, AI language technologies, AI to aid in preserving cultural assets, and education materials/programs to train regional workers in use of next gen technologies
  • AI for Sovereign capability, Defence and regional Queensland communities
  • Technologies, frameworks, and methodologies developed in these projects are applicable to Defence AI and autonomy requirements as per the 2020 Strategic Update and Defence Data Strategy 2021-2023.

Public consultation commences on draft Australian Code of Practice for the Design, Construction, Survey and Operation of Autonomous & Remotely Operated Vessels

New future thinking paper: Autonomous and Remotely Operated Vessels: 2021 to 2040

By Rachel Horne, Assurance of Autonomy Activity Lead, TAS

“Autonomous and Remotely Operated Vessels: 2021 to 2040” [1] is a newly published paper considering the future of autonomous vessels. It forms part of the Maritime Industry Australia Limited (MIAL) Future Leaders White Paper: Predictions for the Australian Maritime Industry 2040.

The paper considers: What are autonomous vessels? How are they regulated? Why is regulation so difficult and what can we change to make it easier? A central focus for the discussion is how we might accelerate development of autonomous vessels with a focus on building sovereign capability and what success in this sector would mean for the Australian autonomous systems ecosystem in 2040.

New future thinking paper: Autonomous and Remotely Operated Vessels: 2021 to 2040

New future thinking paper: Autonomous and Remotely Operated Vessels: 2021 to 2040

The paper aims to introduce autonomous vessels and the opportunities and challenges the technology presents for Australia, in a way that is accessible and thought-provoking. It explores a broad range of topics and issues related to autonomous vessels, starting with background on what these vessels are and why they are used, what their benefits are, and the features we will likely see by 2040 commercially and within Defence, and then moving on to impact on workforce, the current regulatory framework and conceptually difficult areas, and concluding with a range of proposed solutions.

 

Predictions for the Australian maritime industry in 2040

The paper includes six predictions for the Australian maritime industry in 2040, extracted below:

  1. Minimally crewed vessels with a spectrum of autonomous capabilities will be a normalised part of the commercial vessel fleet operating for routine passenger transport, movement of goods, scientific research, and tourism. The police, border protection and Defence agencies will have significant numbers of semi-autonomous and autonomous vessels in their fleets.
  2. A Maritime Water Space Management System (MWSPS) will have been implemented to manage allocation of surface and subsurface water space and interaction between smart vessels.
  3. The deconfliction service offered through the MWSPS, together with advanced navigation, sensing, and inter-vessel communication technologies, will enable minimal-crewing, and multiple semi-or fully-autonomous vessels to be supervised remotely by single operators, due to the significant reduction in collision risk.
  4. A new Commonwealth Government entity, ‘Australian Complex Autonomous Systems Safety Authority’ (ACASSA) will set the standards and conduct assurance activities for the “black box” behind autonomous and semi-autonomous systems, for each of the air, land and maritime domains. Two way secondments between ACASSA and traditional regulators will ensure a seamless experience for stakeholders, consistent regulatory and policy development, and the upskilling of staff.
  5. “RegTech” concepts will be implemented by the ACASSA to enable continuous background monitoring of AI-based autonomous systems, using risk thresholds to determine input required by the operator, and enabling non-intrusive compliance checks.
  6. Australian Ports are able to accommodate large international trading vessels with advanced autonomy on board, and the integration of Vessel Traffic Services with the Maritime Water Space Management System have reduced the workload of VTS operators and vessel crew, reduced incidents, and improved efficiency. [2]

 

How can the 2040 vision be achieved?

The paper argues that change, led by Government and supported by industry, is needed to ensure the maritime industry can access the range of benefits autonomous technology offers into the future. Collaboration, and a focus on new regulatory approaches, upskilling the maritime workforce, and smart ports, are all central to achieving the 2040 vision.

The paper concludes:

Autonomous and remotely operated vessels are already in operation in Australia and around the world, and their capability and availability are rapidly growing. It is predicted that, by 2040, these vessels will be an integrated, integral part of the Australian maritime industry, leading to safer, more efficient, maritime operations, with less environmental impact. However, to achieve that vision, significant effort from Government, the maritime industry, and other stakeholders must be invested to put in place the regulatory frameworks, qualifications frameworks, skills base, and port facilities, that are required.

Transitioning from 2021 to the vision for 2040 will require the advancements contained within the diagram below.

Australian Autonomous Vessel Ecosystem in 2040

These advancements are within Australia’s reach, if a proactive, coordinated effort, led by Government and incorporating industry and the community is enacted.

If this effort is not put in now, for example because of distrust for new technology, fear about the impact on jobs, an inability to depart from ‘the way it has always been done’, or simply disinterest from the Australian Government, other countries, particularly those with more developed technological capability, will seize the advantage, and monopolise the opportunities on the table. Leveraging Australia’s talented technologists and innovators, maintaining a strong focus on building sovereign capability through multidisciplinary activities, and a Government-led, multi-domain effort to revamp Australia’s regulatory approach to emerging technology, will position the Australian maritime industry to take full advantage of the spectrum of safety, environment, efficiency, and economic benefits of autonomous systems technology. [3]

“In other words, to fully realise the potential of autonomous shipping, the development technologies must be deemed valuable by the wider marine industry as well as the society as a whole.” (Advanced Autonomous Waterborne Applications – AAWA initiative)

 

What is TAS doing that supports the vision identified in this paper?

In addition to facilitating the development of game changing trusted autonomous systems technology, Trusted Autonomous Systems (TAS) also has two common-good activities: A1 Ethics and Law of Trusted Autonomous Systems and A2 Assurance of Autonomy. These activities, funded by Queensland Government, provide support and resources to TAS participants and broader commercial and Defence stakeholders.

Under the A2 Assurance of Autonomy Activity, the National Accreditation Support Facility Pathfinder Project (NASF-P) is delivering the following:

The NASF-P has already:

Under the A2 Assurance of Autonomy Activity, the Enabling Agile Assurance of Drones in Queensland project is delivering smart digital regulatory tools to enhance efficiency and communication amongst operators and regulators, facilitating innovation and driving growth in industry.

To find out more about the projects underway at TAS please visit our website, or contact info@tasdcrc.com.au.

 

References

[1] Horne, R. (2021). Autonomous and remotely operated vessels 2021 to 2040. MIAL Future Leaders White Paper. Predictions for the Australian Maritime Industry 2040. Maritime Industry Australia Limited. pp.12-27

[2] Ibid.,

[3] Ibid.,

[4] Ibid.,

 

TAS Report: Analysis of available standards and codes for autonomous and remotely operated vessels

Rachel Horne, Assurance of Autonomy Activity Lead, Trusted Autonomous Systems (TAS)

Autonomous systems offer the ability to increase safety and efficiency while lowering environmental and economic costs. In the last five years, there has been a rapid acceleration in the capacity and availability of autonomous and remotely operated vessels. Autonomous systems need to be trusted by government, regulators, operators and the broader community if this rapid acceleration is to continue and to ensure this technology can integrate into commercial and defence operations. An integral part of gaining trust is having consistent assurance requirements, and a clear, tailored regulatory framework.

The Queensland Government has identified the safety, environmental, and economic opportunities of addressing the assurance and accreditation challenges for autonomous systems, resulting in funding for the Assurance of Autonomy Activity through Trusted Autonomous Systems (TAS)[1]. This Activity: “…aims to unlock Queensland’s, and by extension Australia’s, capacity for translating autonomous system innovation into operational capability, leveraging regulatory and technical expertise and strong stakeholder relationships to support industry and regulators.”[2]

To support the development of a clear, tailored, regulatory framework, TAS is leading the development of an Australian Code of Practice for the Design, Construction, Survey and Operation of Autonomous and Remotely Operated Vessels (Australian Code of Practice), supported by Maaike Vanderkooi of Vanderkooi Consulting. The Australian Code of Practice will represent best practice and is intended to provide certainty for industry by providing clear standards, tailored for the type of vessels and operations common in Australia. The Australian Code of Practice will be voluntary and will be updated periodically. It is hoped that the Australian Maritime Safety Authority (AMSA), which has been closely consulted throughout this project, will recognise the Code and in the future consider incorporating it into their regulatory framework.

There are several codes, standards, and guidelines already available internationally for autonomous and remotely operated vessels. The first step in the development of an Australian Code of Practice is understanding the leading existing codes, standards and guidelines, considering them in an Australian context, and then determining whether any of these documents, or specific approaches they take, could be tailored for use in an Australian context. Recognising the importance of this step, TAS engaged Maaike Vanderkooi of Vanderkooi Consulting to prepare a Report (“Analysis of Available Standards and Codes for Autonomous and Remotely Operated Vessels”), as the first stage in the project to develop the Australian Code of Practice.

Report: Analysis of Available Standards and Codes for Autonomous and Remotely Operated Vessels

The Report analyses:

  • the UK Code of Practice for Maritime Autonomous Surface Ships[3];
  • the LR Code for Unmanned Marine Systems[4]; and
  • DNV GL’s Autonomous and Remotely-operated Ships Class Guideline[5].

As part of the analysis of these documents, the Report seeks to:

  • understand the structure and requirements of each of the codes;
  • identify the differences and similarities between the codes; and
  • consider the codes in the Australian regulatory context.

The below is extracted from pages 81-82 of the Report

“The report finds that:

  • an Australian Code of Practice for autonomous and remotely operated vessels should align with the regulatory framework that already exists for conventional domestic vessels;
  • the three available codes focus largely on vessels which comply with international conventions or Class Rules; and
  • this is different to the context for an Australian Code of Practice, which will be tailored towards commercial vessels operating only in Australian waters.

For this reason, none of the available codes and standards considered in this report provide a template that could be tailored for use in Australia with only minor modifications.

However, each of the three available codes will significantly influence the content of the Australian code. This report uses the analysis of the three available codes and standards to identify the standards or requirements that should apply to autonomous vessels, beyond the requirements of conventional vessel standards. This will include tailored requirements for:

  • situational awareness;
  • control systems;
  • software integrity and testing; and
  • safe states.

This report also finds that the operational requirements that apply to conventional vessels in Australia should apply to autonomous and remotely operated vessels, with some differences:

  • the safety management system requirements need to be tailored to autonomous and remote vessel operations;
  • the minimum crew and crew competency requirements will need to be modified; and
  • there will be additional requirements for contingency planning and control hierarchies, which should be informed by the content of the three available codes and standards.

In line with the available codes and standards, a risk analysis approach, which focuses on the impact of potential failures, should apply to the development and testing of novel systems on the vessel, including the systems for situational awareness and control and all systems which do not meet the requirements of the conventional vessel standards.

Finally, this report notes that the baseline requirement of each of the available codes and standards is for an autonomous or remotely operated vessel to be ‘as safe as’ a conventional vessel. Given that the scope of the Australian code will include very small, low risk autonomous marine equipment, whether or not this baseline approach is appropriate for all vessels subject to the Australian code will need to be considered as part of the consultation process on the development of the Australian code.”

Since this report has been completed, the project has identified the underpinning principles for the draft Australian Code of Practice, using stakeholder consultation to refine and ensure they are fit for purpose[6] (complete) and drafted the Australian Code of Practice (complete).

Next steps

1.     Undertake consultation on the draft Australian Code of Practice with the Australian Maritime Safety Authority (underway);

2.     Undertake public consultation on the draft Australian Code of Practice to ensure it is fit for purpose and will be as useful as possible for Australian industry (expected mid-November 2021); and

3.     Finalise and release edition one of the Australian Code of Practice with accompanying guidance material (expected early 2022).

TAS intends to release edition one of the Australian Code of Practice with accompanying guidance material via our website and it will likely also appear directly on AMSA’s website. Feedback on the Australian Code of Practice can be provided directly to either TAS or AMSA to inform continued iterative development. 

If you would like to contact us in relation to the TAS Code of Practice project, to offer feedback, suggestions, or request more information, please email us at info@tasdcrc.com.au.

[1] Trusted Autonomous Systems, 2021, Assurance of Autonomy Activity

[2] Trusted Autonomous Systems, 2021, Assurance of Autonomy Activity

[3] Maritime UK, 2020, Being a Responsible Industry, Maritime Autonomous Ship Systems (MASS) UK Industry Conduct, Principles and Code of Practice, A Voluntary Code (Version 4)

[4] Lloyd’s Register, 2017, LR Code for Unmanned Marine Systems

[5] DNV GL, 2018, DNV class guidelines: Autonomous and remotely operated ships, DNVGL-CG-0264

[6] Information on these principles and the consultation undertaken available here: Vanderkooi, M. and Horne, R., 2021, Outcomes of a successful webinar on TAS’s project to develop an Australian Code of Practice in 2021