Supporting the Royal Australian Navy's Strategy for Robotics and Autonomous Systems

Building an Evidence Base

by Peter Dortmans, Joanne Nicholson, James Black, Marigold Black, Carl Rhodes, Scott Savitz, Linda Slapakova, Victoria M. Smith

Full Document

FormatFile SizeNotes
PDF file 1.7 MB

Use Adobe Acrobat Reader version 10 or higher for the best experience.

Research Questions

  1. What objectives should frame the Royal Australian Navy’s RAS-AI strategy?
  2. What opportunities and threats do RAS-AI systems pose to the Royal Australian Navy, and how might these change how and where the service operates?
  3. Which RAS-AI capability options are available to the Royal Australian Navy, and how might these change the way the service operates?
  4. What lines of effort should the Royal Australian Navy pursue to implement its RAS-AI strategy?

The Royal Australian Navy has embarked on an ambitious plan to modernise its maritime capabilities to support Australia's defence strategy. The 2020 Defence Strategic Update calls for Australia to be ready to shape the strategic environment, deter actions against its interests and respond with military force when required.

Maritime capabilities feature heavily in the update, including those related to robotics, autonomous systems and artificial intelligence (RAS-AI). The Navy recently established the RAS-AI Directorate, giving it the responsibility of developing a maritime RAS-AI strategic roadmap to provide a path for developing and employing RAS-AI out to 2040.

In this report, the authors provide an evidence base to inform the Navy's thinking as it develops its RAS-AI Strategy 2040. Analysing a range of information captured through a literature review, environmental scan, interviews and workshops, the authors make observations that should shape the evolution of the strategy. A framework for the strategy, consisting of the future operational context, potential RAS-AI effects and a high-level technology roadmap, is developed and populated, and objectives for RAS-AI and implementation lines of effort are identified and discussed.

For the Navy's RAS-AI strategy to succeed, its implementation needs to be planned in a manner that recognises the evolving environment that the service will contend with over the next two decades.

Key Findings

Seven objectives can drive the integration of RAS-AI into the current and planned Fleet

  • Maintain undersea advantage.
  • Grow mass on the surface.
  • Posture to gain a strategic advantage through data.
  • Coordinate through a common control system.
  • Normalise human-machine teaming to create effects.
  • Adapt acquisition processes to optimise investment in RAS-AI.
  • Mobilise academia and industry as part of the total maritime capability.

Given these objectives, Navy's strategy for RAS-AI must recognise key trends and requirements

  • AI will be increasingly pervasive and RAS capabilities increasingly numerous in the future operational environment.
  • Agile and distributed command, control and communications is necessary to support increasingly rapid decisionmaking.
  • Navy must adapt concepts, practices and training to optimise the complex array of interactions between humans and machines.
  • RAS-AI capabilities offer new mission sets that have the potential to change how effects are delivered.
  • Along with seaworthiness and cyberworthiness, trustworthiness is an essential attribute for future capabilities.
  • Although replacing crewed systems with uncrewed ones offers significant benefits, there are hidden costs that are not consistently recognised.
  • An evergreening acquisition approach is needed to ensure the rapid advances in RAS-AI technologies can be readily identified, developed and fielded.

Recommendations

  • Identify and create a transition path for RAS-AI technologies to move from research to experimental systems to fieldable capabilities.
  • Identify future force requirements in which RAS-AI can play a significant role, both in providing new capability options and as an adversary capability that needs to be countered.
  • Develop and implement a recruitment and training regime that recognises RAS-AI capabilities, which will require a different set of workforce skills and competencies.
  • Evaluate both the direct and indirect effects of fielding RAS-AI capabilities on the broader Navy system.
  • Manage RAS-AI across the capability development, acquisition and sustainment cycle in a way that keeps up with technological advances while supporting good governance practices.
  • Design an experimentation program that recognises the need for different approaches that consider the maturity of the RAS-AI system.

Table of Contents

  • Chapter One

    Introduction

  • Chapter Two

    Australia's Strategic Environment to 2040

  • Chapter Three

    Technology Enablers and Transition for RAS-AI

  • Chapter Four

    Lessons for Developing RAS-AI Capabilities

  • Chapter Five

    Building a Strategic Roadmap for Maritime RAS-AI

  • Chapter Six

    Summary

  • Appendix A

    Literature Review Summary

  • Appendix B

    Stakeholder Interview Summaries

  • Appendix C

    U.S. Workshops

Research conducted by

This research was sponsored by the RAS-AI Directorate and conducted by RAND Australia.

This report is part of the RAND Corporation Research report series. RAND reports present research findings and objective analysis that address the challenges facing the public and private sectors. All RAND reports undergo rigorous peer review to ensure high standards for research quality and objectivity.

Permission is given to duplicate this electronic document for personal use only, as long as it is unaltered and complete. Copies may not be duplicated for commercial purposes. Unauthorized posting of RAND PDFs to a non-RAND Web site is prohibited. RAND PDFs are protected under copyright law. For information on reprint and linking permissions, please visit the RAND Permissions page.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.