《北約：2023-2043年下一代量子技術發展趨勢報告（英文版）（150頁）.pdf》由會員分享，可在線閱讀，更多相關《北約：2023-2043年下一代量子技術發展趨勢報告（英文版）（150頁）.pdf（150頁珍藏版）》請在三個皮匠報告上搜索。

1、Science&TechnologyTrends 2023-2043Across the Physical,Biological,and Information DomainsNATO Science&Technology OrganizationVOLUME 1:OverviewScience and Technology OrganizationThe research and analysis underlying this report and its conclusions were conducted by the NATO S&TOrganization(STO)drawing

2、upon the support of the Alliances defence S&T community,NATO AlliedCommand Transformation(ACT),and the NATO Communications and Information Agency(NCIA).This report does not represent the official opinion or position of NATO or individual governments butprovides considered advice to NATO and Nations

3、leadership on significant S&T issues.DISCLAIMERDale F.RedingAngelo De Lucialvaro Martn BlancoCol Laura A.Regan,USAF,PhDDaniel BaylissNATO Science&Technology OrganizationOffice of the Chief ScientistNATO HeadquartersB-1110 BrusselsBelgiumhttp:www.sto.nato.intDistributed free of charge for information

4、al purposes;hard copies may be obtained on request,subject to availability,from the NATO Office of the Chief Scientist.The sale and reproduction of this report for commercial purposes isprohibited.Extracts may be used for bona fide educational and informational purposes,subject to attribution to the

5、NATO S&T Organization.Unless otherwise credited,all non-quantitative graphics are used under Creative Commons licensing and generatedthrough text-to-image Generative AI models using Stable Diffusion 2.1.Copyright NATO Science&Technology Organization,2023First published,March 2023ForewordNATOcontin-u

6、es to strive forpeace,security,and stability inthe whole of theEuro-Atlanticarea.Centralto this task ismaintaining,andsharpening,ourtechnological edge.While we develop and adoptnew technologies,we also strive to preserve the ethi-cal,legal,and moral principles of the Alliance.Thisis what sets us apa

7、rt from our competitors.We wantto shape the standards and norms of emerging anddisruptive technologies,and lead in the ethical useof these technologies in defence.Further,we mustactively address the threats and challenges posed bysuch technological developments both now and inthe future.To do so req

8、uires us to understand theevolving S&T environment.This report constitutesan essential and evidence-based understanding offuture science and technology development,andwill act as a foundation for our future technological,defence,and investment decisions.Mircea GeoanaNATO Deputy Secretary GeneralThis

9、 report as-sesses the state,rate,and poten-tialimpactonthe Alliance ofemerging and dis-ruptive scientificandtechnologi-cal advances ex-pected over thenext 20 years.It updates and extends the previ-ous Science&Technology Trends:2020-2040,re-flecting the considerable geopolitical,technical,andscientif

10、ic developments that have occurred over thelast few years.This assessment is based on a reviewof selected national and international S&T foresightand futures studies,multi-national workshops,andtechnology watch activities conducted by the Sci-ence&Technology Organization.In addition,Igratefully ackn

11、owledge the collaboration and sup-port provided by Alliance and Partner defence R&Dcommunities,the NATO International Staff,AlliedCommand Transformation(ACT),and the NATOCommunication and Information Agency(NCIA).Dr Bryan WellsNATO Chief ScientistTable of ContentsForeword.iiiExecutive Summary-Volume

12、 1.vi1Introduction.11.1Context11.2Purpose31.3Approach31.4Overview52Science&Technology Trends.72.1S&T Development72.1.1Defence S&T Context.82.1.2Synergy.152.2Disruptive Technologies162.2.1Big Data,Information and Communication Technologies.162.2.2Artificial Intelligence.232.2.3Robotics and Autonomous

13、 Systems.302.2.4Space.352.2.5Hypersonics.412.2.6Energy and Propulsion.452.2.7Electronics and Electromagnetics.512.3Emergent Technologies572.3.1Quantum.572.3.2Bio and Human Enhancement Technologies.642.3.3(Novel)Materials and Advanced Manufacturing.712.4Convergence,Inter-Dependencies and Synergies772

14、.4.1Data-AI-Autonomy.772.4.2Data-Quantum.78v2.4.3Space-Hypersonics-Materials.782.4.4Space-Quantum.792.4.5Data-AI-Biotechnologies.792.4.6Data-AI-Materials.802.4.7Energy-Materials-AI.812.5Countering EDT Threats812.6Summary813Contextual Trends.833.1Introduction833.2Geostrategic833.3Innovation and Inves

15、tment863.4Strategic Drivers923.4.1The Operational Environment(Space,Info-sphere,Arctic,and Urban).923.4.2Culture,Ethics&Law.963.4.3The Environment.963.4.4Gender and Technology.973.4.5Mistrust in Science.993.4.6Miscellaneous.1003.5Contextual Trends-Conclusion1014Conclusion.102Appendices.104Bibliograp

16、hy.105Symbols,Abbreviations and Acronyms.134Executive Summary-Volume 1Science&Technology Trends:2023-2043 provides an assessment of Science&Technology(S&T)trends and their potential impact on NATO military operations,defence capabilities,enterprise functions,and political decision space.This assessm

17、ent draws upon the collective insights and research activities ofthe NATO Science&Technology Organization(STO),its collaborative network of over five thousandactive scientists,analysts,researchers,engineers,and associated research facilities.These insightshave been combined with an extensive review

18、of the open-source S&T literature,selected nationalresearch programmes,NATO STO technology watch activities,(serious)research games,STO CPoW(Collaborative Programme of Work)activities,and NATO innovation endeavours.The report is split intotwo volumes covering the overall conclusions(Volume 1)and the

19、 detailed analysis(Volume 2).This report builds upon,updates,and extends the previously published NATO Science&TechnologyTrends:2020-2040,following the same general structure and focus.In general,the conclusions andinsights from that report have stood the test of time.Nevertheless,noteworthy develop

20、ments have occurredin the broad technological and geostrategic environments,with COVID-19,Ukraine,climate change,anew Strategic Concept,and South-East Asia being of note.These changes have,in turn,driven relevantS&T developments and highlighted their implications for enterprise or military operation

21、s.The report aims to assist current and future military and civilian decision-makers in understandingemerging and disruptive technologies(EDTs),thereby guiding NATO R&D portfolio management,innovation activities,and capability planning.It focuses on the following:Why such EDTs are importantto future

22、 Alliance activities;How they are expected to develop over time;and What will this mean tothe Alliance from an operational,organisational and enterprise perspective?Over the next 20 years,we assert that four overarching characteristics will define advanced militarytechnologies.Technological developm

23、ents will be increasingly intelligent,interconnected,decentralised,and digital.These,in turn,will lead to military capabilities that are increasingly autonomous,networked,multi-domain,and precise.Technology will be increasingly dual use,i.e.developed and drawn from thecommercial sector.Emerging tech

24、nology-enabled capabilities will increase the Alliances operationaland organisational effectiveness by enabling the NATO Warfighting Capstone Concepts five WarfareDevelopment Imperatives(WDI):Cognitive Superiority;Integrated Multi-Domain Defence;Cross-Domain Command;Layered Resilience;and wide-rangi

25、ng Influence and Power Projection.At thesame time,such technologies will and indeed are presenting significant challenges to the Alliance,including operational,interoperability,ethical,legal,and moral concerns.NATO has approved an established set of EDTs of interest,also referred to as priority tech

26、nologyareas.Defence Ministers agreed to the first seven EDTs in October 2019.For the previous S&T Trendsreport,the STO added an eighth area(Materials)for future consideration and development.In 2022,viiat the Madrid summit,these EDTs were formally expanded to include Energy&Propulsion and NovelMater

27、ials and Manufacturing.In keeping with the STOs mandate to continue to monitor and evaluate thebroader technological landscape,this report also considers the status of recent developments in Electronics&Electromagnetics(E&EM)technologies.These S&T areas are either currently in nascent stages ofdevel

28、opment or are undergoing rapid revolutionary growth.The final list of EDTs considered in thisreport,along with the usual abbreviations or shorthand used(shown in bold),are:Data:Big Data,Information&Communication Tech-nologies AI:Artificial IntelligenceRAS(or Autonomy):Robotics&Autonomous Sys-tems Sp

29、ace:Space Technologies Hypersonics:Hypersonic Technologies Quantum:Quantum TechnologiesBHET(or Biotech):Bio&Human EnhancementTechnologies Materials:Materials&Advanced Manufacturing Energy:Energy&Propulsion E&EM:Electronic&Electromagnetic TechnologiesTechnological developments in AI,Autonomy,Space,so

30、me areas of Data,Energy,and E&EM areseen to be predominately(but not exclusively)disruptive in nature,as developments in these areas buildupon long histories of supporting scientific and technological development.The focus of action in theseareas revolves around their effective adoption and impact o

31、n other technologies.As such,significant orrevolutionary disruption of military capabilities is either already ongoing or will have a considerable effectover the next five to ten years.Emergent areas are to be found in Quantum,BHET,Materials,and someaspects of Data(e.g.6G technologies).Such developm

32、ents are perhaps better defined as re-emergent asprevious development cycles have significantly affected earlier technology revolutions.These will requiresubstantially more development time(ten to twenty years)before their disruptive natures are fully realisedas military capabilities.Since the last

33、report,several major EDT developments are worth noting,including:Data:The increased use of distributed ledger technologies,advanced analytics and visualisation,andthe development of new networking and wireless technologies(e.g.6G)are accelerating the needfor a well-thought-out digital data backbone

34、linking decentralised sensor and C2 nodes.This,inturn,is driving increased developments in the internet of things(IoT),edge computing,and new dataarchitectures(e.g.mesh,fabric,lake,etc.).AI:Disruptive AI applications and the role of AI as an S&T enabler or catalyst of other EDT devel-opments have be

35、en a significant factor in major developments across the physical,information andbiological sciences and associated technologies.At the same time,the limits of AI practice are leadingto the exploration of new,more robust and trusted methods,moving beyond deep learning and thedevelopment of the AIoT(

36、Artificial Intelligence of Things).RAS:Developments in AI and energy storage are driving the development of increasingly sophisticatedand robust autonomous systems.However,the most significant and impactful RAS development isits wide application and impact on current military operations and plans.Co

37、st reductions,widespreadavailability,and creative use on the battlefield have driven RAS innovation.Space:Commercial enterprises and strategic competitors are accelerating the development of spacetechnologies.The impact of sustained space-based communications(e.g.StarLink)in areas of hostilitiesis o

38、f special note.At the same time,counter-satellite research is of increasing concern,while on-orbitrepair capabilities promise cheaper and longer on-orbit operations.New propulsion systems alsopromise to increase the use of non-near-Earth(cislunar)space and reduce launch costs.Hypersonics:R&D is bein

39、g conducted on mixed-mode engines for broad military applications(e.g.crewed aircraft),deployment of operational hypersonic missile systems and development of effectivecounter-measures.viiiEnergy:Developments in Energy,responding to climate change,and security challenges,are drivingincreased militar

40、y exploration and adoption of electric propulsion(air,land and maritime)and newbatterychemistriesforadvancedenergystorage.Researchintolarge-scaleglobalsolarenergyproductionand terrestrial and extra-terrestrial space-based small nuclear,thorium,and fusion reactors showconsiderable promise for safe an

41、d widely available energy production in the latter half of 2030 or early2040s.It is important to note that AI and novel Materials have been and will continue to be criticalenablers of such developments.Biotech:The near-miraculous rapid development of vaccines(such as that for COVID-19)and thepotenti

42、al development of engineered pathogens are of particular note.Advances in bio-manufacturing,synthetic biology and 3D bio-printing are accelerating.Materials:Research into room temperature superconductors,novel uses of graphene(and other 2Dmaterials such as graphyne),and new semiconductor materials h

43、old considerable promise for futuretechnologies.The applications of additive manufacturing and bio-printing are exploding,disruptingcurrent medical and logistics systems.E&EM:New non-silicon materials and semiconductor designs are pointing toward faster chips andspecialised processors(e.g.neuromorph

44、ic for AI).Truly disruptive effects will occur through technology convergence driven by combinations of EDTsand their complex relationships.The following synergies and inter-dependencies are projected to behighly influential in the development of future military capabilities:Data-AI-Autonomy:The syn

45、ergistic combination of Autonomy,Big Data and AI using intelligent,widely distributed,and cheap sensors alongside autonomous entities(physical or virtual)will leverageinnovative technologies and methods to yield a potential military strategic and operational decisionadvantage.Data-AI-BHET:AI,in conc

46、ert with Big Data,will contribute to designing new drugs,purposefulgenetic modifications,direct manipulation of biochemical reactions,new chemical and biologicalthreats,and living sensors.Data-AI-Materials:AI,in concert with Big Data,will contribute to the design of new materials withunique physical

47、 properties.This will support further developments in using 2-D materials and noveltechniques.Data-Quantum:Over a ten-to-fifteen-year horizon,quantum technologies will expand C4ISR datacollection,processing,and exploitation capabilities through significantly increased sensor capabilities,improved PN

48、T(positioning,navigation,and timing),secure communications,and computing.Energy-Materials-AI:New developments in energy storage,driven by novel materials such as grapheneand exotic battery chemistry,as well as stronger lightweight materials and novel designs(e.g.massivecastings,super-capacitors or 3

49、D printing),will continue to drive electrification or the use of greenfuels(e.g.hydrogen and biofuels)in military operations.AI to support these designs and materialdevelopments and optimise energy use will contribute to the greening of NATO forces.Space-Hypersonics-Materials:Development of exotic m

50、aterials,novel designs,miniaturisation,energystorage,manufacturing methods,and propulsion will be necessary to fully exploit Space and Hypersonicenvironments by reducing costs,increasing reliability,improving performance,and facilitating theproduction of inexpensive task-tailored on-demand systems.S

51、pace-Quantum:Space-based quantum sensors,facilitated by Quantum Key Distribution communi-cation,will lead to high-precision sensors suitable for satellite deployment.Increasingly commercial,smaller,lower power,more sensitive,and distributed space-based sensor networks enabled by quantumsensors will

52、be an essential aspect of the future military ISR architecture in ten to fifteen years.1.Introduction“Be not troubled about the future.You will come to it,if need be,with the same power to reason,as youuse upon your present business.”Marcus Aurelius Antoninus 1The Future1.1ContextSincethelastScience

53、andTechnology(S&T)trendsassessment2,theworldhasseenastaggeringamountof change and disruption.Over the previous three years,NATO,and the world as a whole,have beenchallenged by the withdrawal of NATO forces from Afghanistan,the pandemic of the century(COVID-19),the Russian-Ukrainian special military

54、operation,significant climate disruption,inflation,increasedtensions in the Asia-Pacific,authoritarianism,and disinformation.This strain has been compounded bythe relentless drive of commercial S&T developments,especially in the areas of quantum,biotechnology,and AI.NATO faces a strained and fluid s

55、ecurity environment,with existential challenges and threatsfrom all strategic directions,including state and non-state actors,near-peer military forces,cyberspace,space,and cognitive warfare.NATO is a unique consultative and collaborative military and po-litical framework.A key enabler of its succes

56、s for more than 70 yearshas been the NATO S&T community(the original NATO innovationengine),which has provided NATO with the intellectual and technolog-ical edge needed to ensure the Alliances success across the enterprise,operational,and diplomatic spectrum.Maintaining this S&T edgemeans understand

57、ing and adapting to emerging and disruptive technolo-gies(EDT).In turn,this means anticipating S&T developments andunderstanding their potential operational and strategic implications.The NATO S&T community enables the application of state-of-the-art validated knowledge for defence and security purp

58、oses.It embracesscientific research,technology development,quantitative analysis,capability-based planning,experimen-tation,and a wide range of related scientific activities 3.The NATO Science and Technology Organization(STO)plays a decisive role in supporting innovation;providing profound insights

59、into alliance challenges;ensuring the integration of Alliance capabilities;and making available an interconnected network of science and knowledge workers capable of providing2Chapter 1.Introductionevidence-based advice to NATO,as well as alliance members and partners.At its core,the role of NATOsS&

60、T community is to 3:“.maintain NATOs scientific and technological advantage by generating,sharing and util-ising advanced scientific knowledge,technological developments and innovation to supportthe alliances core tasks.”Figure 1.1:1958 NATO Science Committee(CREDIT:NATO).The importance of S&T to NA

61、TO and the Alliance was institutionalised early in the history of theAlliance 4,first through Article 2 of the North Atlantic Treaty 5 and later through the work and reportof the three wise men 6.Building on that bedrock and the foundational work of Theodore von Krmnand the NATO Science Committee(Fi

62、gure 1.1),the STO in its various incarnations has,over the past 70years,helped NATO effectively employ a strategy of technology 7,8,leveraging a decision and S&Tadvantage to significant intellectual,political,economic,and military effect.As noted by 9,“.inan Alliance united in purpose,extensive and

63、meaningful coordination,cooperation and collaborationof defence S&T adds significant value to national efforts,while establishing interoperability and thenecessary overarching command and control.”This S&T excellence has been a critical enabler of the technological edge that NATO has relied uponfor

64、more than seventy years.However,this advantage has been degraded in recent years due to variousstrategic,economic,social,and technical challenges.Regaining this edge will require NATO to engagewith and leverage the talents,capabilities,and creativity of the entire Alliance and national innovationsys

65、tems 10.Alliance leadership has awakened to the innovation challenge.As noted by leaders duringthe 2021 NATO summit in Brussels 11:“The speed of technological change has never been higher,creating both new opportunitiesand risks in the security environment and to the way NATO operates.We are determi

66、nedto preserve our technological edge and ensure Alliance interoperability to maintain thecredibility of our deterrence and defence posture.We have recently taken important steps tothat end,building on the Emerging and Disruptive Technologies(EDTs)Roadmap we agreedon in 2019.We have now adopted our

67、strategy to foster and protect EDTs.This strategyoutlines a clear approach for identifying,developing,and adopting EDTs at the speed ofrelevance,guided by principles of responsible use,in accordance with international law,andtaking into account discussions in relevant international fora.Moreover,thi

68、s strategy seeks topreserve our interoperability;safeguard our sensitive technologies;and actively address the1.2 Purpose3threats and challenges posed by technological developments by others,both now and in thefuture.Drawing on the extensive innovation expertise of all 30 Allies,we will further leve

69、rageour partnerships with the private sector and academia to maintain our technological edge.”1.2PurposeScience&Technology Trends(2023-2043)provides context and a foundation for the NATO EDT strategy,Alliance capability development and NATO S&T programmes of work.The core objective is to increasethe

70、 understanding within the Alliance of the potential for S&T outcomes to enhance or threaten Alliancemilitary operations.As such,the report is an aid to decision-makers in considering the following:Why such EDTs are important to future Alliance activities;How they are expected to develop over time;an

71、d,What this means to the Alliance from an operational,organisational and enterprise perspective?Anticipating the future security environment better than potential adversaries is one way the Alliancehas maintained a competitive advantage.S&T foresight is a critical aspect of this preparation.It does

72、notattempt to predict the future in detail(a difficult task at best and impossible at worst).Instead,it seeks toprovide a context for anticipating technologys potential development and impact on the Alliance.Thisreport will help guide the development of essential policies,standards and collaborative

73、 S&T developmentwhile supporting effective capability development.Analyses of technology trends and the associated process of technology watch are critical steps toidentify new militarily important technologies and communicate the potential impact of these technologieson NATO and national leadership

74、.Those recognised technologies hold the promise to enable thedevelopment of disruptive military capabilities for Alliance(BLUE)and potential adversarial(RED)forces.The report assesses S&T trends in priority technology areas projected to impact NATO militaryoperations,capability development and core

75、enterprise functions over the next twenty years.These S&Tareas are broad,have significant overlaps,do not cover all S&T development,and are expected to:Mature over 20 years;Be transformative or revolutionary;and,Be emergent or create generational shifts in S&T development.The STO provides these asse

76、ssments for NATO.As noted in the STO charter(2012)3:“To fulfil its mission,the STO will.provide advice to NATO and Nations leadershipon significant S&T issues,including the identification of emerging technologies,and theassessment of their impact on defence and security.”1.3ApproachThis report aims

77、to reach a wide audience inside and outside of NATO and its partners.A frank and opendiscussion of potential opportunities and risks presented by technological developments over the next 20years is essential if NATO is to deploy operationally effective military capabilities.As such,the report isbase

78、d strictly on the following:Technology trends discussed in the open literature;A global perspective on technological progress;Available scientometric analysis and surveys of technical experts;and,4Chapter 1.Introduction Logical reasoning informed by Alliance S&T expertise and technology watch activi

79、ties.Candidate S&T trends,within the context of EDTs,were identified considering which scientific andtechnological developments:Are likely to be realised in a non-cost prohibitive manner within the next twenty years;Will present a significant challenge to Alliance forces(e.g.survivability,costs,inte

80、roperability,legal,etc.);and,Will significantly impact Alliance capability or planning decisions(i.e.decision-making,counter-measures,etc.)Science&Technology Trends:2023-2043 supersedes the Science&Technology Trends:2020-2040report 2.Nevertheless,it draws heavily upon its foundations,structure,insig

81、hts and lessons learned.As before,the report exploits a broad range of open-source reports,internal assessments,NATO EDTand innovation activities,technology reviews,serious games,quantitative analysis and futures studies todevelop a comprehensive understanding of the future technology landscape.Thes

82、e sources include:Existing NATO S&T trend and future security environment studies,strategies,discussions andassessments;Technology watch activities conducted by the S&T Organisation,including existing TechnologyWatch Cards(TWC)(current as of October 2022),Chief Scientist Reports,serious(technology)g

83、ames and Von Krmn Horizon Scans(vKHS);Meta-analyses and reviews of open source technology watch and futures research from defence,security and industry sources;Internal and external quantitative analysis of academic publications,patents and research activities;Surveys of the STO network,Panels and G

84、roup seeking insights into technological developments,readiness and maturity;NATO-sponsored EDT workshops and innovation system engagements;and,Alliance and Partners EDT studies and research programmes.Defence Ministers approved a canonical set of EDTs and an associated roadmap in October 2019.In202

85、2,the meeting of Alliance Heads of State and Government in Madrid authorised two more prioritytechnologies for consideration,Novel Materials and Agile Manufacturing and Energy and Propulsion.Inkeeping with the STOs mandate to continue to monitor and evaluate the broader technological landscape,this

86、report considers the status of recent developments in Electronics&Electromagnetics(E&EM)technologies,including developments in directed energy weapons.As such,ten EDTs are considered indetail in this report,each broken down into technology focus areas,highlighting specific areas of researchand devel

87、opment(R&D).Volume 2 discusses this decomposition in further detail.In reading this report,one should keep several caveats in mind:1.The accurate and detailed prediction of S&T trends is a difficult task,although there is someevidence that such studies have been successful at anticipating S&T develo

88、pment within broad timehorizons 12;2.The ultimate objective of any such forecast is to be impactful in a manner that ultimately drives thedevelopment or understanding of useful technologies 13;3.Technologies rarely evolve in a simple linear fashion,and complex synergies between EDTs areoften as crit

89、ical(if not more so)as the EDTs themselves;1.4 Overview5Figure 1.2:NATO Defence Ministers(2019)(CREDIT:NATO).4.The list of EDTs provides a grouping of related technologies capable of technological disruption.The development of sub-technologies may be very different from the aggregate.Further,such ag

90、rouping is not unique,and one finds many such taxonomies in the literature(e.g.14,15,16,17,18.All such clusters,or taxonomies,are simplifications;however,this particular clustering oftechnologies has proven useful for Alliance purposes;and,5.Technology has historically driven the changing nature of

91、human conflict,but not conflict itself19.More broadly,“technology is neither good nor bad;nor is it neutral”(Krazbergs FirstLaw of Technology 20).In future conflicts,NATO will inevitably use new technologies,and itis necessary to define how that should occur before any crisis.This understanding prov

92、ides anessential first step to supporting technology-policy decisions,potential capability development,andpreparing defensive countermeasures.As such,discussion of the possible application or impact ofS&T should not be taken as an indication of current or future NATO S&T research efforts.Further,suc

93、h a discussion represents the considered assessment of the authors and does not represent anofficial NATO position.1.4OverviewWithin the following chapters,an analysis is presented of identified and militarily relevant S&T trends,which may impact NATO capability development and operational challenge

94、s over the upcoming 20 years(2023-2043).The key data sources,methodology,and analysis used to conduct this assessment aredescribed in the Methodology chapter of Volume 2.The assessment is presented in three parts:1.An overview is provided of the general nature of S&T development.Specific EDT areas a

95、reidentified that are expected to impact significantly NATO from 2023 through 2043(Chapter 2).These EDTs are presented separately,broadly considering the state and rate of development andthe military implications.Tables and graphs simplify and summarise the more detailed analysissupporting this repo

96、rt.This overview is followed by a consideration of critical potential synergiesbetween EDTs,as it is in the overlap between these developments that significant disruptions andinnovation will occur;2.The broad strategic context and drivers are outlined that will impact defence S&T development(Chapter

97、 3);and,3.Volume 2 has separate appendices,which provide a more detailed exploration of each EDT,drawingheavily upon STO research and technology watch activities.This section also includes Conjecture6Chapter 1.IntroductionCards,short vignettes that describe the potential future application of these

98、technologies.Theyare included to help contextualise the potential impact of these technologies.In addition to thesedetailed assessments,Volume 2 also reviews the data sources and methodology used to generatethis report.The bibliography at the end of this document and Volume 2 provides an extensive l

99、ist of usefulreferences.These are also used throughout the body of the text where appropriate.When using the AdobePDF version of the report,clicking on a numbered reference will take the reader to the relevant entry inthe bibliography.If desired and available,clicking on the provided URL(i.e.web-lin

100、k)will allow thereader to open the source reference directly for further study and exploration of the topic.2.Science&Technology Trends“Dont worry about what anybody else is going to do.The best way to predict the future is to invent it.Really smart people with reasonable funding can do just about a

101、nything that doesnt violate too many ofNewtons Laws!”-Alan Kay 21Apprehension2.1S&T DevelopmentThis chapter summarises the development of S&T,deemed essential to NATO,and seeks to answerthe following:What do we mean by emerging or disruptive S&T?What are these emerging or disruptive technologies,or

102、what disruptive scientific discoveries do weanticipate?What is the state,rate of development and projected impact of these technologies?What synergies do we need to pay particular attention to?While intuitively,the terms emerging and disruptive may be self-evident,there is little consensusaround the

103、 actual definitions for emerging and disruptive technologies(for example,see 22,23,24).Nevertheless,for purposes of this report,we narrowly define technologies as:8Chapter 2.Science&Technology TrendsEmerging:Those embryonic technologies or scientific discoveries that are expected to reachmaturity in

104、 2023-2043;and are not widely used currently or whose effects on Alliance defence,security and enterprise functions are not entirely clear.Disruptive:Those technologies or scientific discoveries expected to have a major or revolutionaryeffect on NATO defence,security,or enterprise functions in 2023-

105、2043.Convergent:Combinations of technologies that are integrated in a synergistic and novel manner tocreate a disruptive effect.Sustaining:Incremental improvements to existing technologies and resulting capabilities focusedpredominately on reducing size,weight,power,and cost(SWaP-C)or improving effe

106、ctiveness.Not all technologies or scientific discoveries are emergent or disruptive,nor is disruption driven solelyby technology 25.Further,not all emerging technologies will be disruptive;not all are emergent,andnot all emerging ones will drive convergent technologies.For this report,we focus on th

107、ose technologiesassessed as most likely to be disruptive over twenty years,including those that have moved beyond theinitial exploration phase but have not yet become widely exploited.Understanding the natural pattern ofEDTs development is a prerequisite in understanding and assessing their potentia

108、l effects on NATO andthe Alliance.The methodology section of Volume 2 discusses this in more detail.2.1.1Defence S&T ContextOne of the critical lessons(re)learned during the conflicts fought over the last two or more decades,especially during the war on terror and NATO operations in Afghanistan,is t

109、he essential human-centricnature of armed conflict.However,as noted in 26,one may take this(important)insight too far andassume that technologys role is diminished in modern warfare or that this human-centric perspective isindependent of the technological ecosystem in which we all live and work:“It

110、has become in vogue for leaders to argue that one of the lessons of the past decade ofwar is that“technology doesnt matter in the human-centric wars we fight,”as one four-stargeneral put it to me.But that assumes a definition of“technology”as exotic and unworkable.To paraphrase the musician Brian En

111、o,technology is the name we give to things that wedo not yet use every day.Once we use it every day,we do not call it technology anymore.Whether a stone or a drone,it simply becomes a tool we apply to a task.”As this quote highlights,a risk in dealing with S&T developments is that one quickly become

112、s inuredwith such technologies and,therefore,greatly underestimates the impact they are or may be having(e.g.the rapid development of effective COVID-19 vaccines).Over the last five years,technologys role in military operations and geopolitical influence has becomeincreasingly more pervasive.There h

113、ave been many revolutionary episodes in the development of warfare27,28,and it is well recognised that the current seventh generation military revolution 8,29,30 isbeing driven(once again)by rapid changes in the technological landscape.In a Clausewitzian sense,2.1 S&T Development9human-organised con

114、flict(war in its most extreme case)is a fundamental clash of wills between largesocial groups(e.g.states,pseudo-states,communities,societies,etc.).During such conflicts,whetherwith peer competitors or asymmetric threats,technology is an edge 31 to be exploited and,in a nuclear-constrained world,disp

115、layed 8.As democratised technology becomes even more central to humanexistence,it will also play an out-sized role in shaping modern conflict.As noted in 32:This next wave of innovation will stitch together the physical,digital,and bio-technicalrealms.as a result,the character of warfare is changing

116、.Already,we are in a new era ofpersistent cyber,economic,and information conflicts below the level of overt combat thatrisks bringing us closer to direct confrontation.Emerging technologies are changing therange and specificity of effects,enabling the microtargeting of individuals,and qualitativelyc

117、hanging the way we communicate,perceive our environment,and make decisions.Thisapproach centres around several focuses,including distributed and networked operations,human-machine collaboration,human-machine teaming,primacy in software-centric warfareand greater technological interoperability and in

118、terchangeability with allies and partners.”Broad interest in such EDTs arises from the fear that such technologies may revolutionize aspects ofour societies or provide an adversary with the means to change the balance of power.As noted by manyauthors in one form or another,Technology shapes warfare,

119、not war.War is a condition in which astate might find itself;warfare is a physical activity conducted by armed forces in the context of war19.Within broad strategic and geopolitical context(see Chapter 3),the nature of conflicts(warfare)ischanging,with general agreement that the transforming technol

120、ogical environment is a significant factor24,32,33.This changing nature of conflict manifests itself in hybrid war 34,35,36,hyper-war 37,memetic warfare 38 or next-generation conflict 39.In each,disruptive technologies are merged withexisting technologies and military capabilities to create new ways

121、 and means of engaging in conflict.Theyrarely replace such existing technologies immediately but are integrated into existing plans,doctrine andoperations.In 2016,Klaus Schwab,Founder and Executive Chairman of the World Economic Forum,defined thefourth industrial revolution as one which creates a wo

122、rld in which virtual and physical manufacturingsystems cooperate with each other in a flexible way at the global level 40.Complex technologicalsynergies and a fusion of the physical,biological,and information domains mark such revolution 41.As noted by 42,this technological development is characteri

123、sed by evolving and expanding immersiveexperiences,accelerated AI automation,and optimised technologist delivery.Thus,the technologies ofthe fourth industrial revolution underlie the development of the seventh military revolution.Unlike the last 80 years,EDTs are no longer driven by the needs and in

124、terests of the Defence Sectornor by clear societal objectives 32.Civil sector investments dwarf those of the defence sector,especiallyin such areas as quantum technologies,AI,bio-technologies and electronics.That is not to say thatpublic sector investment is inconsequential but rather drives bespoke

125、 very high-risk technologies,such asexquisite and novel semiconductors technologies 43.10Chapter 2.Science&Technology TrendsAnother common theme for technology development is how impactful and pervasive the synergisticdevelopment of AI-Data-RAS has been.It is difficult to overstate this combinations

126、 impact on developingnovel technologies.This raises the importance of breadth in technology development.Such breadthis as important as the actual pace of development.This breadth challenges prioritisation.EDTs covereverything from Quantum to Biotechnology,from Novel Materials to AI.This range makes

127、prioritisationdifficult,particularly as developments within a specialised area may vary widely.Yet at the same time,this tsunami of mutually reinforcing technological developments leads to combinatorial trends wherethese technologies can create new possibilities when theyre used together 44.Finally,

128、a theme in this report that shall arise again and again is the increasing focus on human-machineinteractions,including teaming,interfaces and symbiosis.The increasing scientific understanding of thehuman mind(driven in no small measure by advances in sensors,analysis,data and AI)will acceleratehuman

129、-machine interfaces as a pervasive disruptor.Historically,technology has driven the battlespaces changing(subjective)nature.Nevertheless,ithas not fundamentally altered human conflicts(objective)nature.Innovative technologies are inevitablyused in conflict,whether designed initially for such or not.

130、At the same time,they may create entirelynew operational domains(e.g.air,cyber&space)or reinvigorate old ones(land,sea&cognitive).Rapidchanges in the technological landscape are driving the current seventh-generation military revolution.Yet,organized human conflict(war in its most extreme case)remai

131、ns a fundamental clash of willsbetween large social groups(e.g.states,pseudo-states,communities,societies,etc.).With the developmentof increasingly autonomous and AI-driven systems,the objective nature of warfare is changing.Modernmilitary forces will need to adapt to operational domains increasingl

132、y dominated by AI and Autonomyand the technologies they enable,potentially redefining what is meant to impose ones will on an opposingforce.Transformation of Military forces will be increasingly dominated by the need to understand andsuccessfully integrate technology-enabled capabilities(e.g.DOTMLPF

133、-I)into the order of battle.Thosewith creativity and imagination will determine future military success in designing,developing,andoperating information/autonomy-enabled technologies across the tactical-operational-strategic spectrumat all levels of technological sophistication.However,this does not

134、 necessarily imply the dominance ofnew and expensive technology.In this context,the old engineering adage of“good,fast,cheap picktwo”will continue to apply and may be used by adversaries on the battlefield to negate an expensivetechnological advantage.The common factors that link these Fourth Indust

135、rial Revolution technologies are that they are allin some way,shape,or form:intelligent,interconnected,decentralised and digital(I2D2)in nature.Over the next 20 years,these four overarching characteristics can be expected to define advancedmilitary technologies.In each,disruptive technologies are me

136、rged with existing technologies and militarycapabilities to create new ways and means of engaging in conflict.More specifically,we note that thefuture S&T landscape will be characterised(and at the same time driven)by the following qualities:1.Intelligent:Technologies will exploit integrated AI,know

137、ledge-focused analytic capabilities,and symbioticAI-human intelligence to provide disruptive applications across the technological spectrum.2.1 S&T Development11(a)Autonomy:Artificial intelligence-enabled autonomous systems will be capable of independentdecision-making.Such autonomous systems may be

138、 robotic,platform-based or(digital)agent-based.(b)Collaborative Intelligence:The seamless integration of psycho-social-techno systems willsupport enhanced human-machine teaming and synergistic behaviours.(c)Knowledge Analytics:Advanced analytical methods(including AI)exploring large data setsand exp

139、loiting advanced mathematics will provide unrealizable insights,knowledge,andadvice.2.Interconnected:Technologies will exploit massive networks in and across virtual,biological,and physical domains,including networks of sensors,organisations,individuals,and autonomous agents,linked via newencryption

140、 methods and distributed ledger technologies.(a)Trusted Communications:The use of distributed ledger technologies(e.g.blockchain),quantum key distribution(QKD),post-quantum cryptography,and AI cyber-agents will ensuretrusted interactions and information exchange.(b)Synergistic Systems:The developmen

141、t of mixed(physical or virtual)complex systems-of-systems will create novel ecosystems(e.g.smart cities).(c)Biological-Physical Systems:Technologies will exploit the human,machine,biological andphysical domains via AI,human-machine interfaces,and genome manipulation.3.Decentralised:There will be dis

142、tributed and ubiquitous large-scale sensing,storage,and computation.(a)Edge Computing:Data storage,computation and analytics/AI will be embedded into agentsand objects close to information sources.(b)Ubiquitous Sensing:Low(or lower cost)sensors will create large sensor networks across thehuman-physi

143、cal-information domains.12Chapter 2.Science&Technology Trends(c)Decentralised Production:Just-in-time local digital manufacturing and production will exploitAI-assisted design,novel materials,and(mixed material)3D/4D printing technologies.(d)Democratised S&T:The broad availability of S&T information

144、 and fostered innovation willresult from increased and widely available high-performance computational capabilities,reduced design and production costs and the resulting generation of novel science.4.Digital:Novel disruptive effects(both positive and negative)will result from digitally blended biolo

145、gical,physical and information domains.(a)Digital Twin:A digital simulacrum of physical,biological,or information entities digitallylinked(ofteninnearreal-time)totheoriginalwillsupportpredictiveanalytics,experimentationand assessment.(b)Synthetic Realities:New perceived cognitive or physical realiti

146、es would be created based onintegrating psycho-socio-techno systems.Such realities may be augmented,virtual,social orcultural.Each of the four S&T trends will create synergistic combinations of technologies that will drivespecific military capability development trends.These are as follows:1.Intelli

147、gent Autonomous Action:Intelligent+DecentralisedIntelligent and increasingly autonomous systems are already supplanting and exceeding the capabil-ities of human forces.However,autonomous systems have been quite limited,employing fixed rulesets and various levels of direct human control.The increased

148、 use of AI will enable autonomoussystems capable of significantly more sophisticated decision-making,self-directed activity and,atthe same time,increasingly complex human-machine teaming.Such increased use of intelligentagents will dramatically expand into our synthetic realities,including cyber,bat

149、tle,and digitalsocial networks.Autonomous agents will provide rapid analysis,advice and courses of action forstrategic-operational-tactical planning.In turn,this development will allow for increased OODA2.1 S&T Development13(Observe-Orient-Decide-Act)loop effectiveness,and bring an entirely differen

150、t perspective on oldproblems unconstrained by old strategies.Such intelligent battle networks can potentially increasedecision speeds to levels requiring new human-machine interaction and visualization methods.Theresulting competition between battle networks will generate increased evolutionary pres

151、sures onalgorithms,each seeking an edge or combination of effects that will lead to a decisive victory.Similarly,enabling autonomous vehicles will increase effectiveness across the conflict spectrum,creating large sensor and strike mesh networks.Recent events suggest that these developmentsmay turn

152、the future battlespace into one more akin to those found in the late nineteenth and earlytwentieth centuries.2.Cognitive Dominance:Interconnected+DigitalEvolving agile and adaptive mesh C4ISR battle networks will create deep military action-dependentoperational dependencies.Such evolving battle netw

153、orks will increasingly become targets in,ofthemselves,and subject to effects-based conflict.This increased reliance on seamless and ubiquitousconnectivity will increase the operational importance of targeting such networks(military or civilian)through disinformation and cyber or physical attacks.Suc

154、h attacks may be implemented long beforethe conflict is initiated and could indirectly strike at logistics,personnel,information,financial orothersupportingelementsofmodernoperationalandstrategicnetworks.Overa15-20-yearhorizon,quantum technologies will increase C4ISR data collection,processing and e

155、xploitation capabilitiesthrough significantly increased sensor capabilities,secure communications,and computing.Inconcert with Big Data,Quantum computing and AI will contribute to the design of new materialswith unique physical properties generated by increased S&T dominance.This will enable further

156、developments in 2-D materials and novel designs.Space-based quantum sensors,facilitated byQuantum Key Distribution communication,will lead to a different class of sensors suitable forsatellite deployment.Increasingly commercial,smaller,lower power,more sensitive and distributedspace-based quantum-en

157、abled sensor networks will be an essential aspect of the future military ISRarchitecture in 20 years.Finally,supporting technologies will enable decision dominance,allowingfaster action-reaction cycles,especially in the cognitive and cyber domains.3.Expanded Domains:Interconnected+Decentralised14Cha

158、pter 2.Science&Technology TrendsAs the operational environment expands to include space,cyber,and the broader information sphere,the need to think,plan,and operate in a widely dispersed,interconnected,and multi-domain mannerwill become even more critical to mission success.The growing numbers and wi

159、de distributionof multi-domain sensors,multi-domain missions,and the rising processing capabilities embeddedat the networks edges will present new demands for dominance,counter-domain capabilities,protection,counter-countermeasures,and other secondary functions.The increased exploitationof new domai

160、ns will inevitably lead to the search for domain superiority,with attendant costs andcapability demands.In concert with Big Data and Quantum,AI will expand our ability to exploitthe biological domain,contributing to new designer drugs,purposeful genetic modifications,directmanipulation of biochemica

161、l reactions,and living sensors.These technologies will enable moreeffective human-enhancement technologies across the cognitive,social,and physiological domains.They will also increase chemical-biological countermeasures effectiveness(speed and efficacy)while paradoxically enabling the development o

162、f even more virulent agents.4.Precision Warfare:Intelligent+DigitalIncreased digitisation across C4ISR capabilities,miniaturisation,edge processing and fallingcosts have been the underpinning technological developments enabling increasingly intelligent,interconnected,and distributed systems.This has

163、 dramatically increased the development ofprecision strikes and effects-orient capabilities.However,swarming and using lower-cost cheapprecision weaponry has and will continue to put large,high-value capabilities at risk.At the sametime,increased digitisation exposes new and hitherto unanticipated v

164、ulnerabilities.New sensors(e.g.quantum technology-enabled)and increased reliance on synthetic realities(virtual,social,mixed,twinned,etc.)will present risks and opportunities.Using more sophisticated analytical toolsand leveraging the increased volumes of digital data will lead to new operational ca

165、pabilities.Forexample,novel hypersonic weapon designs will be developed using increasingly higher-fidelitycomputational fluid dynamics models and embedded sensors.The development of exotic materials,novel designs,miniaturisation,energy storage,manufacturing methods,and propulsion will benecessary to

166、 exploit space and hypersonic operational environments by reducing costs,increasingreliability,improving performance,and facilitating the production of inexpensive task-tailoredon-demand systems.Directed energy(precision)weapons will become more widespread andeffective as power and energy storage is

167、sues are addressed.AI,as a linking technology,will change the landscape of warfare.At the same time,the availabilityof digital data will allow decentralised and interconnected(autonomous)systems to analyse,adapt,andrespond.In turn,these changes will potentially support better decision-making through

168、 predictive analytics45.All of this will occur in the context of synergistic and symbiotic systems,including sensors,societies,and organisations.In this way,EDTs will continue to change the ways and means of conflict for at least ageneration but,at the same time,will need to integrate and operate al

169、ongside existing systems.Emerging technology-enabled capabilities will increase the Alliances operational and organisationaleffectiveness by enabling the NATO Warfighting Capstone Concepts five Warfare Development Impera-tives(WDI):Cognitive Superiority;Integrated Multi-Domain Defence;Cross-Domain C

170、ommand;Layered2.1 S&T Development15Resilience and,wide-ranging Influence and Power Projection 46.At the same time,such technologieswill and indeed are presenting significant challenges to the Alliance,including operational,interoperability,ethical,legal,and moral concerns.Figure 2.1 summarises the n

171、ature of this future NATO battlespace.The timelines associated withenabling this multi-domain future are difficult to define precisely,but the military and technologicaltrends are increasingly evident.It took almost sixty years(US Civil War to WW1)to move through theindustrialisation of warfare and

172、a similar length of time for its informatisation.We are in the middleof a similarly dramatic shift with the intelligentisation of conflict domains,with potentially analogousconsequences.These technologies will give military decision-makers options under ELM(ethical-legal-moral)norms and guidelines.T

173、hat said,NATO and its Allies must be ready to counter the malicious useof such technologies that go against these norms.Space Enable Prepare Project Sustain Enagage Protect Inform C3Cyber Enable Prepare Project Sustain Engage Protect Inform C3Air Enable Prepare Project Sustain Engage Protect Inform

174、C3Land Enable Prepare Project Sustain Engage Protect Inform C3Maritime Enable Prepare Project Sustain Engage Protect Inform C3Cognitive SuperiorityCross-Domain CommandMULTI-DOMAIN OPERATIONSLayered ResilienceInfluence&Power ProjectionIntegrated Multi-Domain Defence Figure 2.1:The future NATO Multi-D

175、omain intelligent-interconnected-decentralised-digital strategic andoperational battlespace.2.1.2SynergyTo maintain a military-technological edge and to prevail in future operations,NATO forces must continu-ally evolve,adapt,and innovate to be credible,networked,aware,agile,and resilient 47.Such ada

176、ptationis most rapid and disruptive where EDTs work to enable one another or where the human,information,orphysical domains overlap 48.This highlights the importance of technology breadth(more opportunitiesfor surprising combinations and unexpected synergies).Several such critical synergistic connec

177、tions areidentified later in this report.In addition to interconnections between EDTs,it should be noted that many of the issues driving andlimiting the effective development of new capabilities are non-technical.Murray 49,50 notes that:“What matters in the technological adaptation as well as techno

178、logical innovation is howwell new and improved technologies are incorporated into effective and intelligent conceptsof fighting:it is not the technological sophistication that matters,rather,it is the largerframework.”For active development of EDTs into Alliance capabilities,the implications of cult

179、ure,concepts,risk-tolerance,organisational structure,policies,treaties,human capital,and ethics must be fully appreciated.16Chapter 2.Science&Technology TrendsThese factors will need to evolve as much as the technology if EDTs are fully developed into newoperational capabilities.2.2Disruptive Techno

180、logiesThis section provides a brief overview of critical developments and trends associated with this disruptiveS&T.Volume 2 of this report provides a more comprehensive review of each EDT.The scientometrics section at the end of each EDT discussion presents a table with the assessedpotential impact

181、,state,and rate of development for sub-areas associated with an EDT.These resultsare drawn from a survey of the STOs research network.This is followed by charts representing globalEDT leadership and trends,as determined by assessing the last five years of English language scientificpublication data.

182、This data was integrated and processed in the S&T Ecosystem Analysis Model(STEAM),described in more detail in the methodology section of Volume 2.2.2.1Big Data,Information and Communication TechnologiesData describes Big Data(raw digital data)that presents significant volume,velocity,variety,veracit

183、y andvisualisation challenges.Increased digitalisation,a proliferation of new sensors,new communicationmodes,the internet-of-things and the virtualisation of socio-cognitive spaces(e.g.social media)have contributed significantly to the development of Big Data.Advanced(Data)Analytics describesadvance

184、d analytical methods for making sense of and visualising large volumes of such information.These techniques span various approaches drawn from research areas across the data and decisionsciences,including artificial intelligence,optimisation,modelling&simulation(M&S),human factorsengineering and ope

185、rational research.Two additional aspects are essential in considering the big datachallenge:Information and Communication Technologies,and sensors and sensing.This system-of-systems is necessary for an effective multi-domain C4ISR framework,reflecting the collection,processing,exploitation and disse

186、mination of information supporting decision-making and C2.Big Data,Advanced Analytics and Information Communication Technologies(orData)Vast quantities of data and the algorithms to make sense of it underlie the development of the fourthindustrial age.Our world of experience has become increasingly

187、digital and virtual.To put this in propercontext,the world created 94 Zettabytes(ZB)(1021)of data,with 80 to 90%of that data unstructured,in2.2 Disruptive Technologies172022 alone.That number is expected to double every two years,rising to nearly 100,000 ZB annuallyby 2042 if this trend continues un

188、checked.70%of global GDP is digital,and data(not in analogueform)created in 2021 and 2022 alone represent 90%of all data ever created.When the last technologytrends report was written in 2020,the worlds cloud data storage was 6800 Exabytes(1000 Exabytes(EB)makes a ZB).This is expected to grow to mor

189、e than 200 ZB by 2025.All this data must be collected,transmitted,and stored,creating a growing demand for energy and analytical tools to help make sense ofit.This demand also creates huge piles of waste products,including toxic e-waste.It is estimated that only32%of the information available to bus

190、inesses is leveraged in any form 51,and overall,only 0.5%ofdigital data is exploited.It is difficult to overstate just how important Data is in driving and enabling EDTdevelopment and ensuring that the digital backbone and analytical methods that NATO will use in theupcoming years will be up to the

191、task.AI is a critical complementary technology whose growth,in somesense,is a reaction to this data volume challenge and information opportunity.In turn,this combinationof Data and analytical technologies has created a global network with characteristics of neural systems,apoint noted originally by

192、the media theorist Marshall McLuhan 52,53.For the next 20 years,this growth in data willhave a fundamentally disruptive effect on Allianceoperations,capabilities,and technological devel-opment.Data sets of a magnitude and complexitythat are difficult to handle logistically(a defini-tion that it must

193、 be noted changes yearly)due toincreasing volume,velocity,variety,veracity andvisualisation issues will present significant techni-cal,organisational and interoperability challenges.Distributed sensors,autonomous systems,new communication technologies(e.g.6G),new antennadevelopments,improved spectru

194、m usage,increased use of space assets,virtual socio-cognitive spaces,digital twins,ever more power-efficient electronics and the development of new and expanding analyticalmethods will increase our ability to understand the human,physical and information spaces around us.Datais the enabling technolo

195、gy for all EDTs and is central to their exploitation for enhanced militarycapabilities.Moreover,AI requires high-quality curated training data to develop new algorithms andapplications,placing even greater demands for more and better data.For NATO,Data will enable increased operational efficiency,re

196、duced costs and improved logistics,real-time monitoring of assets and predictive assessments of campaign plans.At the same time,it willgenerate significantly greater situational awareness at strategic,operational,tactical and enterprise levels.These applications will lead to a deeper and broader app

197、lication of predictive analytics to support enhanceddecision-making at all levels,a point recognised by other near-peer competitors.It has the potential tocreate a knowledge and decision advantage,which will be a significant strategic disruptor across NATOsspectrum of capabilities.There is the poten

198、tial to significantly impact NATOs kinetic and non-kinetictargeting effectiveness using cheap,widely distributed sensors(as part of the internet-of-things(IoT),linked by new communication protocols(such as 6G),building on analyses and dissemination of criticalinformation in real-time.Potential peer

199、or near-peer adversaries will seek a similar technical edge,whileasymmetric threat actors will exploit increasingly open and available data sources for targeted effect ordisruption.The commercial sector invests heavily in Data and ICT.Over the next 20 years,it is expectedthis will continue,and comme

200、rcial interests will lead in the overall development and application.Theeffectiveness of this investment underlies the current knowledge economy.There are no indications thatthis will change.Nevertheless,the unique needs of NATO military forces will require developing methodsand standards for intero

201、perability,sharing,collection,modelling&simulation,analysis,classification,curation,communication,and data management.Finally,it is not a given that more data and advancedalgorithms will ultimately produce better decisions.Understanding the complex socio-cognitive-technicalcontext around decision-ma

202、king and the proper role and integration of Data in this context will be essentialto developing a NATO decision advantage.The human sciences will be particularly significant in guiding18Chapter 2.Science&Technology Trendsdevelopments in this area.To be understood properly,Data needs to be viewed as

203、a series of technical challenges coveringdigital data collection,processing,exploitation and dissemination and analytical methods.Consequently,developments in Data are best considered along the lines of effort identified in the subsections below.Advanced Computing and MethodsDevelopments in applied

204、mathematics and the information sciences continue to yield new approaches tooptimisation,modelling&simulation(M&S),management of uncertainty,and addressing complexityand chaos.Some methods explore foundational approaches to improving the modelling and simulationof complex and complicated systems.Oth

205、er areas of note are developing new statistical and analyticalmethods to assess data integrity and graph and sparse data analytics.These methods are empoweredby advanced computational methods such as multi-part computation(“analytics to the edge”),newprobabilistic programming languages,and accelerat

206、ing advances in quantum computing.Finally,emergentresults are coming to light in modelling and simulation,especially around data-driven models whereempirical models of natural,complex processes and phenomena use libraries of data modelling primitivesand human-model interfaces.Such developments are e

207、xpected to continue over the next 20 years.Tosupport this analytical need,new approaches,especially new computational paradigms,such as quantumcomputing,mathematical methods and developments in traditional supercomputers,will be essential,54,55,56.Concerning conventional supercomputers,biomimetic ap

208、proaches such as neuromorphicsupercomputers provide a promising approach to significantly increasing supercomputer processingcapabilities 57,58,59,60,61.New approaches will also impact operational decision-making using new computing paradigms suchas edge,fog,transparent,mobile,and dispersed computin

209、g.Novel Applications and Decision MakingData needs to be exploited if it is to support decision-making.Developments over the next years willfocus on three core aspects:group decision-making;improved interfaces and visualisation;and theapplication of data analytics to enhance aspects of situational a

210、wareness such as comprehension,datafusion,control,discovery,design,and M&S.In particular,the increased use of digital twins(e.g.avirtual digital simulacrum of a natural system)supports the prediction of future failures or performanceimprovements through AI,analytics,or M&S methods.The use of digital

211、 twins is expected to expanddramatically as AI/ML and advances in predictive analytics will help extend its application 62,63,64.Good advice is built upon good data and ap-propriate algorithms.Unfortunately,both are re-markably fragile,and research focuses on ensuringalgorithms fairness and intellig

212、ibility,associat-ing embedded logic,and protecting data integrity.The need for new and innovative developments inthis area cannot be overstated with the increasein data and an ever-greater reliance on analyticsand AI/ML to make sense of the data.While it isdifficult to predict the exact nature of su

213、ch advances,the pressure for better and fast analytical methodsbased on new and evolving mathematics and programming paradigms will grow with the increase inavailable data.Distributed Ledger TechnologiesDistributed ledger technologies(DLT)hold promise for supporting logistics,personnel and financial

214、transactions.Any digital transaction,which requires tracking and assurance,will benefit from DLT.Thatsaid,DLTs are only as strong as their encryption protocols,and the need to implement post-quantumencryption will become acute(e.g.the development of quantum-resistant ledgers(QRL).The use ofquantum p

215、roperties is also being explored,specifically in the context of quantum money,which exploitsquantum properties to work around issues of block-chain scalability 65,66.2.2 Disruptive Technologies19Advanced NetworksData must be moved to where it is needed to be useful.Therefore,developing new network t

216、echnologies,such as the exploitation of 5G and the maturation of 6G technologies,will be essential.These will becomeeven more critical aspects of military networks,with challenges of low-probability of intercept,hidingtraffic in commercial signals,SWaP-C and rapid creation of mission networks critic

217、al challenges over thenext decade.Integrated sensor networks,communicationlinks,and processing will continue to drive theneed for C3 warfare or multi-domain battle net-works.Exploiting these evolving networks fortactical,operational,or strategic purposes is anarea of considerable R&D and their impor

218、tance inthe upcoming decades is only expected to grow.Much of this work involves understanding how tolink new sensors and capabilities into a seamlessand compelling whole.Given their importance on the modern battlefield,the role of offensive EW is alsogrowing.This,in turn,has led to the development

219、of such concepts as mosaic warfare,which seeks to bringtogether current concepts of edge networks,data lakes,networks of low-cost sensors,multi-domaininformation networks,and autonomous and manned systems to create an asymmetric decision advantagethrough the leveraging of complexity 67,68.In additio

220、n,the use of AI to both rapidly exploits raw dataand act as an information glue,e.g.linking incompatible systems into a cohesive whole.Developmentsare also expected to occur in the use of very low-power networks will allow surveillance networks tooperate clandestinely for years without intervention.

221、The internet of things,cloud computing,power to the edge,etc.,are all driven by the availability ofadvanced networks.There is considerable commercial activity in this area.Ultimately,this will increasethe reliance on a mesh of nodes that facilitate greater situational awareness and exploitation of i

222、nformation.SensorsSensors collecting data are critical elements of any data strategy.Significant advances in quantumtechnologies(e.g.magnetic anomaly detection,gravimetric,etc.)are enabling novel sensors.Oldertechnologies such as passive coherent location radars 69 are also growing in importance and

223、 will enablestealth surveillance.Another technology of note is the development of computational imaging systems,which are lens-free optical systems employing direct computational methods to create images 70.Suchapproaches may also be extended to radar 71.Finally,research on HF radars and 3D LiDAR sh

224、owconsiderable promise and are expected to be refined further over the upcoming decades.Developmentsfor wide-area surveillance and long-range communication 72 will be driven by concerns about thevulnerability of space-based sensors,innovative antenna designs and the challenges of operations indiffic

225、ult environments 73,74.Bioinformatics and the sensors that supportit are already important health and performancemonitoring tools.Combined with smartphones,continuous monitoring of personnel and readinessis already possible,while new diagnostic and as-sessment technologies will greatly increase CBRN

226、countermeasure options and detection.StorageAn often-overlooked aspect of big data is the needfor storage.Technologies such as helium drives,shingled magnetic recording(SMR),DNA storage,largememory serves NVRAM,Rack scale design and 5D Optical storage 75,76 hold promise but the energyand access spee

227、d challenges are not insignificant.DNA storage seems to be the most promising for large-20Chapter 2.Science&Technology Trendsscale storage of information 77,and developments in biotechnologies will enable developments in thisarea.As critical as technologies are,developing new data architectures will

228、 also improve the managementand storage of immense volumes of data.Improvements to data warehousing,lakes and fabrics will berequired over the next five to 10 years,as will data architectures that implement(biomimetic)forgettingstrategies 78,79 to reduce data storage and access issues.CyberThe incre

229、ased reliance on data,networks,sensors,and analytics that may drive future success on thebattlefield also presents significant cyber vulnerabilities.Cyberspace,as a domain that must be protectedand manoeuvred within,is expected to grow in even greater importance and challenge.Solving the cyberdefenc

230、e dilemma will require bringing together hardware,humans and data 80,81.In particular,developments in zero trust security 82,and the role of trust in general will support more robust cyber-physical systems 83,84,85.The use of cyber security meshes has also been identified as an area ofinterest and g

231、rowth 86.ScientometricsTable 2.1:Big Data,Advanced Analytics and Information Communication Technologies(Data)2023-2043.EDTTechnology Focus AreasImpactTRLHorizonDataAdvanced Computing&MethodsHigh7-82025-2030Novel Applications&Decision MakingHigh5-62030-2035Distributed Ledger TechnologiesHigh5-62025-2

232、030Innovative NetworksHigh5-62030-2035Networked Sensors&SensingHigh5-62025-2030Data StorageHigh7-82022-2025CyberHigh5-62025-20302.2 Disruptive Technologies210.00 1.00(a)Data-Leading Countries(Map)(STEAM Analysis).Leading Countries 2018-2021 DataNATONon-NATO%of Data Publications per Nation44%13%12%10

233、%7%6%5%5%5%4%United States of AmericaUnited KingdomGermanyChinaFranceItalyCanadaJapanAustraliaSpain(b)Data-Leading Countries(STEAM Analysis).Figure 2.2:Big Data,Advanced Analytics and Information Communication Technologies(Data)-STEAMResults-Countries22Chapter 2.Science&Technology TrendsEDT Trends o

234、ver 4 Years Data1.011.451.712.050.150.20.250.280.230.280.310.330.120.170.190.210.380.630.730.840.540.680.740.732.783.814.354.830123456789102018201920202021Advanced Computing&SoftwareCyberData StorageDistributed Ledger TechnologiesInnovative NetworksNetworked Sensors&SensingNovel Applications&Decisio

235、n Making%of Data Publications per Technology Area over Total Number of EDT Publications83,555106,616139,110173,7641,605,174 1,476,978 1,680,228 1,877,170 2018201920202021Data PublicationsTotal EDT Publications(a)Data-Topic Trends(STEAM Analysis).Leading Institutions 2018-2021 DataNATONon-NATO%of Dat

236、a Publications per Academic Institution3.6%2.3%1.7%1.7%1.6%1.6%1.6%1.5%1.5%1.4%Max Planck SocietyChinese Academy of SciencesHarvard UniversityUniversity of CambridgeSpanish National Research CouncilUniversity of OxfordCalifornia Institute of TechnologyMassachusetts Institute of TechnologyUniversity

237、of TokyoINAF(b)Data-Top Institutions(STEAM Analysis).Figure 2.3:Big Data,Advanced Analytics and Information Communication Technologies(Data)-STEAMResults-Trends and Top Institutions2.2 Disruptive Technologies232.2.2Artificial IntelligenceAI refers to the ability of machines to perform tasks that nor

238、mally require human intelligence forexample,recognizing patterns,learning from experience,drawing conclusions,making predictions,ortaking action whether digitally or as the smart software behind autonomous physical systems 87.Artificial Intelligence(AI)OverviewArtificial Intelligence(AI)is the abili

239、ty of machines to perform tasks that traditionally require humanintelligence for example,recognising patterns,learning from experience,drawing conclusions,makingpredictions,or acting whether digitally or as the intelligent software behind autonomous physicalsystems 88.There are several branches of A

240、I,but the two of most interest to us are Machine Learning(ML),a branch of AI that exploits data and statistical methods,and deep learning based on neural networks,both algorithmic and biological 89.AI has been called the most impactful technology ever invented and is expected to play a significantdi

241、sruptive role in defence and security.While AIs practical and theoretical challenges have beenexplored since the mid-1950s,developments over the last decade have given new impetus to researchand development.However,as AI techniques are widely used,their limitations become more apparent,leading to co

242、ncerns about returning to an“AI winter”and highlighting the need to revisit the enablingsciences mathematical,algorithmic,and technical foundations.As a critical Alliance and national resource,Data may be the new“oil”,but AI is probably themost important modern tool for refining this data into actio

243、nable information.The need to automate theanalysis,clustering,exploitation,and interpretation of increasingly large and complex datasets has beenthe primary driver for developments in AI.Each AI development cycle has expanded the methodologiesand application of AI to real-world problems while drivin

244、g the need for more curated and validated data.Today,AI/ML is deeply embedded in modern technology,and this dependency has exploded over the last3 years with new astounding applications arising almost every day.New methods and the increasinglywide availability of sizeable(often)publicly available tr

245、aining sets are particularly important to thisdevelopment.More critically,AI has become the essential driver of massive technological developmentsin other EDT areas,an often-overlooked aspect.AI is a priority R&D area in the commercial world,with many nations making significant investments.Business

246、is the primary driving force behind AI,although research is often based on widely availableopen-source tools and publicly available data 90,91.The brittle nature of most existing applications and24Chapter 2.Science&Technology Trendsthe need for explainable AI are two serious technical challenges tha

247、t remain to be overcome.Complexproblems associated with human-AI teaming and psycho-socio-technical issues will also need to beconsidered but hold the promise of revolutionary applications.Notwithstanding these limitations,by 2030,it is estimated that the contribution of AI to the global economy wil

248、l be$15.7 trillion(USD)92.The integration of AI into business has lagged in recent years,even as the pace of new applicationsgrows ever more 93,94.Largely,this reflects the difficulties of fully exploiting any new technologywithin business,but also highlights the limitations of the AI talent pool.Th

249、is has led some researchers tonote that there appears to be a strong pull from the developers of new methods towards new applications.Over time,this separation between the state and rate of development of new methods and applicationsshould be expected to lead to disillusionment with AI applications,

250、given the brittleness and obtuse natureof many applications.This possible slowdown in research,the resulting exodus from the labs to business,concerns at the algorithmic level,abuse and legal issues on data and other AI challenges are all a recipefor another winter within 5 to 10 years,followed by t

251、he inevitable spring.The holy grail of AI innovation is the development of Artificial General Intelligence(AGI,e.g.human-level generalized intelligent behaviour).Such an evolution would present a significant(and potentiallyimpossible)technical challenge despite over 60 years of AI research.It is unl

252、ikely that AI systems willmeet this level of cognitive ability within the next 20 years.Undoubtedly,AI is and will have a revolu-tionary impact on NATO operations and capabil-ities.AI is the fulcrum around which big data willbe leveraged into actionable knowledge and,ulti-mately,a NATO decision adva

253、ntage.Integrationof AI into combat models&simulation,enterprisesystems,decision support systems,cyber defencesystems and autonomous vehicles will allow forrapid and more effective human-machine decision-making.The use of AI on sensors to pre-processinformation and provide adaptive use of frequencies

254、(e.g.cognitive radar)and bandwidth will paradoxi-cally lead to decreased communication traffic.AI will also significantly affect the conduct of NATO S&Tefforts as meta-analyses of existing research will expose discoveries,identify promising research areas,and provide improved S&T tools to support fu

255、rther research.Policy,legal,and interoperability challenges will be serious challenges for NATO.In recognition ofthis enabling role and the associated challenges,NATO has developed an AI implementation strategydesigned to guide NATO in the effective and ethical use of AI within operations and the en

256、terprise 95.Many nations are doing the same 96.Much has been written on the growth of AI and its importance for the fourth industrial age,as AI isessentially an enabler for other technology areas.Developments in AI are clustered into four areas drivenby a need to reduce data requirements,improve agi

257、lity,increase resilience,support generalisation to otherproblem sets,and work more synergistically with“wetware”systems(e.g.humans).Such developmentsspan various research areas and applications in the defence and security sphere.These are broken into thetopics below.Advanced AIOver the next 20 years

258、,exploring new mathematical and statistical AI approaches is essential to long-termdevelopment and application.Such AI methods research is broad in scope.While the use of AI methodsis expanding rapidly,there remain critical limitations to current approaches.Trust and transparency(orexplainability)ar

259、e two of the most pressing challenges.Still,some researchers have argued that currentmethods are deeply flawed,requiring a complete rethink of our approach to AI.Two unique approaches underlie AI,symbolic and sub-symbolic 97,with each mostly exploredseparately.Research into identifying ways in which

260、 symbolic and sub-symbolic solutions may interacttechnically may improve the accuracy of systems and enhance military decision-making.Symbolicapproaches to AI characterise the first wave of AI research.Work in this area would greatly increase the2.2 Disruptive Technologies25explainability and reprod

261、ucibility of AI results,and is expected to be a growth area in the upcoming years.Advances in neural networks have been oneof the most critical enablers of effective AI overthe last decade.They are a subset of ML and un-derlie modern deep-learning approaches.Neuralnetworks are biomimetic algorithms

262、suitable forpattern recognition and correlation problems withbig data.They are exceptionally agile and adapt-able,capable of continuous learning and broadapplicability.Innovative developments focus onoptimising their use for specific classes of prediction,classification,and processing problems.One a

263、reaof interest is improving the effectiveness and efficiency of neural networks through such methods aslong short-term memory,hierarchical recurrent,residual,convolutional,and quaternion-valued methods.Competitive strategies such as Siamese Networks and generative adversarial networks(GAN)have alsod

264、emonstrated considerable promise.Another area of research focuses on learning architectures to makeAI more agile and resilient while using less data.Methods such as deep Q learning,federated learning,and“machine common sense”hold promise.Significant embryonic developments are taking place inConvolut

265、ional Neural Networks 98(optimised for 3D image recognition),Convolutional Autoencoders99,and Dilated Convolutional Neural Networks 100,101.Additional areas of development are in:Federated Learning:ML methods are being developed to leverage local processing and federateddata collected at the edge(e.

266、g.mobile devices and the Internet of Things(IoT).Such methodstypically do not share collected data with centralized systems,exploiting the data locally,and arepart of an AI to the Edge trend.Systems employing such an approach are more robust due to theincreased volume of data used and are inherently

267、 more secure and private 102,103.Learning with Reduced Labelling:Labelling data is a critical first step in training supervisedmachine learning(ML)algorithms.This approach is a time-consuming,resource-intensive,andcostly step,leading to inflexible solutions.New methods require considerably less labe

268、lled data,and ML algorithms are far more adaptable to real-world conditions.These approaches providesystems able to use sparse data sets with methods that are easier to train,more versatile,able togeneralize from related datasets and ultimately are more broadly applicable or valuable 104.Machine Com

269、mon Sense:New AI algorithms seek to integrate“common sense”into AI reasoning.This means AI reasoning employs a baseline understanding of situations,perception,behaviour,motor functions,memory,and fundamental physics.As a result,such systems are expected to bemore effective,robust,and agile,demonstra

270、ting greatly improved human-like reasoning 105,106.ApplicationsArguably,the most exciting developments in AI are those associated with their application.Perhaps themost exciting development in AI is the wide-ranging and high impact of its application.AI continues tointegrate many systems,processes,a

271、nd defence capabilities.Some of the more exciting and potentiallydisruptive applications are artificial social intelligence,automating disinformation and cognitive warfarestrategies,producing and identifying“deep fakes,”supplementing air or air-weapons control,aide inhigh-resolution image recognitio

272、n,creating images from text descriptions,navigating human terrain,andproviding universal(low usage)language translation.One of the least appreciated and yet most significant area where AI disruption may occur is as anenabler of new scientific,mathematical,and engineering discoveries and as a forcing

273、 function for a newscientific paradigm 107,108,109,110,111.The ability of AI to enable truly disruptive S&T is difficultto overstate as recent examples of solving protein folding(a fifty-year-old challenge)or the developmentof new materials highlight the yet fully untapped nature of AI for scientifi

274、c discovery.26Chapter 2.Science&Technology TrendsAnother area of current and future disruption in the development of AI is to create systems that cangenerate images or video based on textual data.Significant breakthroughs have occurred within 2022,withthe development of generative AI methods with th

275、e release of Stable Diffusion 2.1(used to generate manyof the images used in this report),DALL-E 2 and Midjourney 112.Similarly,the artificial intelligencechatbot program ChatGPT can produce sophisticated text interaction with users and may even be used fordiagnostic purposes 113.The wide-spread use

276、 of these tools is expected over the next few years,withever-growing sophistication and growing social challenges 114.Indeed,the use of GPT-3 has raised astorm of concerns.GPT-3 is 115:GPT-3(Generative Pretrained Transformer 3)is a state-of-the-art language processingAI model developed by OpenAI.It

277、can generate human-like text and has a wide range ofapplications,including language translation,language modelling,and generating text forapplications such as chatbots.It is one of the largest and most powerful language processingAI models,with 175 billion parameters.GPT-3 was used to build ChatGPT,

278、which has surprised many with its ability to hold a conversation orwrite a paper.While the concerns are grossly overstated,it is clear that the development of increasinglysophisticated generative AI will be destabilizing over the next few years,and engender a burst of creativityat the same time.Coun

279、ter-AIAs Alliance and competitor forces increasingly use AI,it becomes increasingly critical to detect,deflect,and limit the impact of attacks on Alliance AI while undermining adversarial AI-enabled systems.AI-on-AI engagements are emerging(mainly in the context of disinformation),and the role of AI

280、 to enablecriminal behaviour is underappreciated 116,117.Like any operational domain,there is a need to detect,deflect and limit the impact of attacks on Alliance AI.Countering adversary AI and associated decisionprocesses through AI manipulation and deception is a difficult technical challenge.Simi

281、larly,detectingAI manipulation is not straightforward.Finally,AI-on-AI conflict is not just a possibility,but also anevolving reality in cognitive warfare and disinformation 118.Therefore,these engagements will becomeincreasingly common,and it is necessary to identify adversarial AI,model it,assess

282、its weaknesses,anddevelop counter-AI strategies 68.The importance of exploring AI countermeasures was noted in theD3TX exercise as a weak signal technology prime for increased R&D 119.Similarly,the role of AI inenabling criminal behaviour is underappreciated 116,117.In addition,AI methods are often

283、remarkablybrittle,and estimating such brittleness is a difficulttask 120,121,122.Improving the robustnessand tools for validation and assurance will be anarea of focused development.Ensuring AI adviceis robust,trusted,ethical and consistent with na-tional rules-of-engagement(ROE)will require AIappro

284、aches with a strong emphasis on robustness,agility,explainability,trust,and human-AI collab-oration.Further,it will be necessary,especially in Alliance operations,to define processes and standardsfor verification,validation and accreditation(VV&A)of such AI systems.Human-Machine SymbiosisA core AI c

285、hallenge is to marry AI with human-driven systems to create an effective psycho-social-technical collaborative system.Research is moving forward rapidly,focused on developing explainable AI,understanding trust-building and ensuring confidence(validation and assurance)in the operationalizationof AI-e

286、nabled systems.Analysis of published data suggests that this is a growing concern as systemsmove away from focusing on the technical side of AI to considering the broader socio-technical aspectsand optimal usage.Explainability,reproducibility,and trust are all critical research areas 123 andmajor le

287、aps forward are expected over the next 5 to 10 years,as faith in AI methods will depend on2.2 Disruptive Technologies27such advancements.New AI/ML techniques will need to be developed to support their assessments andexplicitly communicate associated constraints and limitations.These human-interpreta

288、ble ML models arecombined with visualisation and communication modes that are more conducive to human understandingand evaluation 105,124.Finally,rethinking human interaction with AI to bring about and impact acquired knowledge preciselywhen practical and applicable via user-friendly interfaces will

289、 be necessary to build trust in such systems.TrustistheAchillesheelofAIandhumaninteraction.Understandinghowtogenerateatrustedrelationshipbetween humans and AI and understanding what it means for AI to trust human judgment 105 a difficulttechnical challenge.ScientometricsTable 2.2:Artificial Intellig

290、ence(AI)2023-2043.EDTTechnology Focus AreasImpactTRLHorizonAIAdvanced AIRevolutionary3-42035 or(+)ApplicationsHigh5-62025-2030Counter AIRevolutionary3-42030-2035Human-Machine SymbiosisRevolutionary3-42035 or(+)28Chapter 2.Science&Technology Trends0.00 1.00(a)AI-Leading Countries(Map)(STEAM Analysis)

291、.Leading Countries 2018-2021 Artificial Intelligence(AI)NATONon-NATO%of AI Publications per Nation41%16%11%10%5%5%5%4%4%4%United States of AmericaChinaUnited KingdomGermanyFranceJapanCanadaItalyAustraliaIndia(b)AI-Leading Countries(STEAM Analysis).Figure 2.4:Artificial Intelligence(AI)-STEAM Results

292、-Countries2.2 Disruptive Technologies29EDT Trends over 4 Years Artificial Intelligence(AI)2.943.774.445.374.645.956.797.840.270.440.680.850.040.060.080.1102468101214162018201920202021Advanced AIApplicationsCounter AIHuman Machine Symbiosis%of AI Publications per Technology Area over Total Number of

293、EDT Publications126,482151,004201,233265,9111,605,174 1,476,978 1,680,228 1,877,170 2018201920202021AI PublicationsTotal EDT Publications(a)AI-Topic Trends(STEAM Analysis).Leading Institutions 2018-2021 Artificial Intelligence(AI)NATONon-NATO%of AI Publications per Academic Institution2.4%2.2%1.7%1.

294、4%1.3%1.3%1.2%1.1%1.1%1.0%Max Planck SocietyChinese Academy of SciencesHarvard UniversityUniversity of CambridgeStanford UniversityUniversity of OxfordMassachusetts Institute of TechnologyUniversity of TokyoPeking UniversityUniversity of Michigan(b)AI-Top Institutions(STEAM Analysis).Figure 2.5:Arti

295、ficial Intelligence-STEAM Results-Trends and Top Institutions30Chapter 2.Science&Technology Trends2.2.3Robotics and Autonomous SystemsAutonomy is the ability of a system to respond to uncertain situations by independently composing andselecting among different courses of action to accomplish goals b

296、ased on knowledge and a contextualunderstanding of the world,itself,and the situation.Autonomy is characterized by degrees of self-directed behaviour(levels of Autonomy)ranging from fully manual to fully autonomous 125,126,127.Robotics is the study of designing and building autonomous systems spanni

297、ng all levels of Autonomy(including full human control).Unmanned Vehicles may be remotely controlled by a person or actautonomously depending on the mission.Applications include access to unreachable areas,persistentsurveillance,long endurance,robots in support of soldiers,cheaper capabilities,and a

298、utomated logisticsdeliveries.Robotics and Autonomous Systems(RAS,or Autonomy)OverviewRobotics,autonomous systems,and uncrewed vehicles have become commonplace on the modern battle-field.Likewise,Data and AI have enabled operations in the information space.Robotic and AutonomousSystems(RAS)have exten

299、ded these operations as effectors in both the physical and information domain.The history of autonomous systems in defence is a long one going back to at least 1898 with NikolaTeslas demonstration of a wireless remotely operated crewless boat 128.However,building uponadvances in Data and AI,there ha

300、s been a significant push over the last 20 years to use system autonomyacross various physical and virtual environments.Collaborative autonomy,SWaP-C(size,weight,power,and costs)reductions and significant improvements in on-board AI have made RAS an effective forcemultiplier in operations across the

301、 spectrum.The success of these efforts is seen in the increased use ofplatform autonomy(e.g.unmanned vehicles(UxVs),with ISTAR(intelligence,surveillance,targetingand reconnaissance)and precision strike platforms being increasingly common in operations.One needonly look at recent operations in Ukrain

302、e 129,130,Syria 131,and Nagorno-Karabakh 132 to seehow profound the impact of RAS has been in altering the battlespace.The ultimate objective of integrating RAS into operations has always been to unite the human andautonomous system(at whatever level of independence)into a formidable team,allowing t

303、he automatedsystem to take on dull,dirty,dangerous and dear tasks(the four Ds of robotisation)133,while atthe same time exploiting the unique talents of the human system.The underlying motivation is todecrease costs,reduce manning,improve operational effectiveness,and reduce casualties.The increasin

304、gimportance of RAS in operations is driven by the creativity and availability engendered by SWaP-Creduction.2.2 Disruptive Technologies31Approaches to autonomy may range from fully autonomous to semi-autonomous or even unmannedsystems.Specific levels of independence depend on sensors,mission type,co

305、mmunication links,on-boardprocessing,legal and policy constraints.The drive for more semi-autonomous and fully autonomoussystems in operations will dramatically expand future NATO capabilities into an environment where everysoldier acts as a company,every ship as a task group and every aircraft as a

306、 squadron.RAS development isprimarily driven by operational needs such as high-altitude-long-endurance(HALE),increasing levels ofintegrated AI,decision speed and human-machine factors(e.g.how to make the overall human-machineteam/system more effective while retaining necessary human oversight and de

307、cision-making).In addition,legal,policy and interoperability considerations will challenge the use of autonomous systems across thekill chain.Nevertheless,given the operational advantages to both NATO and potential adversaries,thereis little doubt that autonomous systems will significantly and incre

308、asingly enhance,threaten,and enablecurrent and future operational capabilities over the next 20 years 134.Developmental areas in RAS may be grouped into four broad areas,like those for AI.We will considereach in the following sub-sections.Advanced RASMajor advances in RAS,both current and foreseen,a

309、re predominately based on improvements to SWaP-C,manufacturing,AI,and swarming behaviours.SWaP-C reductions are and will be due to improvedminiaturisation,manufacturing methods,novel low-power sensors,and advances in digital communicationtechnologies.Improvements in this area are expected to be stea

310、dy,resulting in costs continuing to declinewhile availability on the battlefield will increase.In addition,advances in printable electronics and otherhardware will increase the use of 3D-printed task-tailored disposable systems(e.g.135,136,137)forboats,aerial swarms,autonomous submarines,missiles,lo

311、itering munitions,and logistics.The most exciting area of advanced RAS isthe integration of AI.Agile learning-based auton-omy is an evolving technology,a critical certifica-tion challenge,and is expected to develop quicklywith developments in AI.Technical challenges arebeing driven by advances in AI

312、,especially deepneural nets for perception,reinforcement learningfor control,and online model learning.Support-ing such adaptability are significant developmentsin spatial tracking of multiple moving objects inreal time.Some of this work is underpinned bysophisticated neural networks.However,other p

313、romising biomimetic approaches(e.g.modelled on insectneurons 138)are being explored.Improving integrated AI systems performance and power overheadwill improve SWaP-C and system adaptability.Military development is dwarfed by commercial interestsin this area,particularly in the development of self-dr

314、iving vehicles 139,140,141.However,despite thehype around such systems,significant technical challenges remain,and widespread military use on thebattlefield is still a long way into the future.However,more limited use in support functions may comequicker 142.RAS ApplicationThe use of RAS across the

315、battlespace,operational domains,enterprise and industry(especially inmanufacturing and logistics)is increasing as costs,adaptability and miniaturisation improve.Thereare three trends to note within this space:the increased use and development of collaborative systems,biomimetic micro-and mini-system

316、s,and increased usage across the operational spectrum.Collaborativesystems research of interest explores advanced mix-domain autonomous teams and improved humaninterfaces for control and collaboration.Other research explores the application of non-homogeneousRAS teaming and air-launched systems for

317、independent or cooperative operations.In addition,excitingdevelopments are arising in using RAS systems for specialised environments,such as urban ISR,long-duration underwater vehicles,crewless naval vessels,assisted communication,and supporting operations32Chapter 2.Science&Technology Trendsin subt

318、erranean environments.Finally,using RAS systems for 3D printing holds significant promise fordeployed military operations in the future 143,144.C-RASWith the increasingly widespread use of UxVs and swarms,research is ongoing to understand how toanticipate and defeat these threats through kinetic and

319、 non-kinetic means.Swarm-on-swarm engagementis an open area of study.It is clear from recent operations in Ukraine and elsewhere that improvement toC-RAS capabilities is an area that will need priority development over the next decade.Failure to counterever cheaper and widely available UxVs can and

320、will have undesired strategic consequences.RAS Man-Machine TeamingResearch expands human-machine symbiosis,al-lowing humans and machines to work as col-leagues,partners,and teammates.This researchrecognises that humans cannot fully assimilate,un-derstand,and act on the volume of informationpresented

321、 nor control the autonomous collectionof that data.Therefore,more effective and naturalinteraction of AI and RAS systems with humanswill significantly enhance operations.Interestingaspects of this work consider determining and sig-nalling the need for human control and how theembedded AI can assess

322、whether the human operators input can be trusted.ScientometricsTable 2.3:Robotics and Autonomous Systems(RAS):2023-2043.EDTTechnology Focus AreasImpactTRLHorizonRASAdvanced RASHigh5-62025-2030ApplicationHigh5-62025-2030Counter RASHigh3-42030-2035Human-Machine TeamingHigh3-42030-20352.2 Disruptive Te

323、chnologies330.00 1.00(a)RAS-Leading Countries(Map)(STEAM Analysis).Leading Countries 2018-2021 Robotics&Autonomous Systems(RAS)NATONon-NATO%of RAS Publications per Nation35%15%12%8%7%7%5%5%5%4%United States of AmericaGermanyUnited KingdomFranceItalyChinaJapanSpainCanadaSwitzerland(b)RAS-Leading Coun

324、tries(STEAM Analysis).Figure 2.6:RAS-STEAM Results-Countries34Chapter 2.Science&Technology TrendsEDT Trends over 4 Years Robotics&Autonomous Systems(RAS)0.040.050.050.050.050.090.120.151.361.721.851.6600.511.522.52018201920202021Counter RASEnhanced RASHuman-Machine Teaming%of RAS Publications per Te

325、chnology Area over Total Number of EDT Publications23,30827,42133,96034,9741,605,1741,476,9781,680,2281,877,1702018201920202021RAS PublicationsTotal EDT Publications(a)RAS-Topic Trends(STEAM Analysis).Leading Institutions 2018-2021 Robotics&Autonomous Systems(RAS)NATONon-NATO%of RAS Publications per

326、 Academic Institution2.8%1.8%1.5%1.3%1.3%1.3%1.3%1.3%1.2%1.2%Max Planck SocietyMassachusetts Institute of TechnologyUniversity of OxfordUniversity of CambridgeETH ZurichCalifornia Institute of TechnologyChinese Academy of SciencesHarvard UniversityImperial College LondonUniversity of California Berk

327、eley(b)RAS-Top Institutions(STEAM Analysis).Figure 2.7:RAS-STEAM Results-Trends and Top Institutions2.2 Disruptive Technologies352.2.4SpaceSpace is a unique physical domain challenging Allies traditional perceptions of time,distance andgeography.It is generally considered to begin 90-100 km(the Karm

328、an line 145)above sea-level.Space Technologies exploit or must contend with the unique operational environment of space,whichincludes:freedom of action,global field of view,speed,freedom of access;a near-vacuum;micro-gravity;isolation;and extreme environments(temperature,vibration,sound and pressure

329、).Space Technologies(ST,or Space)OverviewHumankind has been making effective use of space for over 60 years.Still,over the last decade,therehas been a renewed interest in exploring and exploiting space and the technologies that enable it.In thiscontext,two interrelated and intersecting trends have l

330、ed to an explosion in the exploitation of space andspace-based assets.First,commercial space has only grown stronger in the last few years,leading to the developmentof satellites,sensors,communication,and launch.Further,the assets that commercial companies havebrought to the table have,in many cases

331、,changed the economics of space exploration(e.g.Acceleration,Reconnection,Turbulence and Electrodynamics of the Moons Interaction with the Sun(ARTEMIS).They have also contributed to a changing battlefield(e.g.the Ukraine-Russian war and the use of Starlink146,and the availability of space-derived im

332、agery has complicated military operations placing themunder a global spotlight 147).Finally,the global commercial space industry continues to lead on thepath to dramatic decreases in launch costs,new options for deploying space-based assets,and the nearreal-time commercial availability of high-quali

333、ty space-derived information(e.g.Electro-Optical(EO)and Infra-Red,Synthetic Aperture Radar(SAR)and Electronic Intelligence(ELINT)Satellites).Second,new technologies and manufacturing methods have changed the nature,availability,andcosts of using space(e.g.3-D printing 148,149,150,151).Such technologies include new propulsionoptions such as advanced electric propulsion systems,spin launch,on-board