Drone Technology in Ukraine - Automation, Lethality & The (Scary) Development Race

Perun
7 Apr 202471:42

Summary

TLDR无人机在乌克兰的演变和生产速度令人震惊,其数量和能力的飞跃式增长对战争产生了深远影响。本文深入探讨了乌克兰战场上小型无人机的类型、演变及其对全球未来战术的潜在影响。分析了无人机在目标识别、射程、制导、杀伤力和对抗电子战措施方面的进步,以及它们如何改变战场态势和战术部署。

Takeaways

  • 🚀 乌克兰战场上无人机的快速进化和生产量的提升,已经对战争的各个方面产生了巨大影响。
  • 🎯 无人机技术的进步,特别是在夜间作战和长距离打击方面的能力提升,使得战场更加透明且危险。
  • 💣 无人机的有效载荷和打击能力的提升,使得它们能够对更多类型的目标造成有效伤害。
  • 🔍 电子战和反无人机措施的不断演变,显示了无人机对抗技术的发展和战场适应性。
  • 🛠️ 无人机的生产方式从商业现成产品到专门定制,反映了战争需求对工业生产的影响。
  • 🔥 无人机在乌克兰战争中的战略影响虽然有争议,但它们的战术和作战价值不容忽视。
  • 📈 乌克兰和俄罗斯都在大量投资无人机生产,这表明无人机在现代战争中的重要性日益增加。
  • 🌐 无人机技术的发展和应用可能会对未来战场和军事战略产生深远的影响。
  • 🤖 随着无人机自主性的提高,未来无人机可能在更多领域发挥作用,同时带来新的挑战。
  • 🔄 无人机的广泛使用和生产可能会改变军事力量的平衡,特别是在资源有限的情况下。
  • 🚨 无人机的威胁和防御将继续成为军事研究和发展的重点,随着技术的进步,新的对策和战术将不断出现。

Q & A

  • 乌克兰战争中小型无人机的发展有哪些显著特点?

    -乌克兰战争中小型无人机的发展特点包括数量和能力的大幅提升,生产目标在某些情况下增加了一个数量级以上。技术方面,无人机的感知、射程、目标定位、致命性和对抗措施的能力都在不断进化。

  • 在乌克兰,小型无人机在战场上的影响有哪些?

    -小型无人机在乌克兰战场上提高了战场透明度,增加了战场危险性。它们有助于定位目标和校正炮火,同时对于没有足够掩护或电子战保护的防御位置构成了持续威胁。

  • 无人机在乌克兰战争中的战略影响如何?

    -虽然无人机在乌克兰战争中提供了战术和操作层面的成功,但有观点认为它们在战略层面上的影响有限,未能决定战争的胜负。然而,如果考虑到如果没有无人机的投资,战争可能会有不同的发展,无人机可能在某种程度上改变了战争的平衡。

  • 乌克兰和俄罗斯在无人机生产方面采取了哪些不同的模式?

    -俄罗斯采取了自上而下的集中式生产模式,重点放在规模化生产少数系统上。乌克兰则采取了更加分散的生产模式,涉及多个小型生产商、志愿者组织和个人,这种模式可能更具有创新性和抗打击能力。

  • 乌克兰计划在2024年生产多少无人机?

    -乌克兰计划在2024年生产至少一百万架无人机,并可能进口同样数量的无人机。

  • 无人机在乌克兰战争中扮演了哪些角色?

    -无人机在乌克兰战争中扮演了多种角色,包括情报、监视和侦察(ISR),以及作为攻击平台使用。它们被用来发现目标、校正炮火、进行夜间操作,并且能够携带各种类型的载荷进行攻击。

  • 乌克兰和俄罗斯如何应对无人机威胁?

    -乌克兰和俄罗斯采取了多种措施应对无人机威胁,包括使用电子战设备干扰无人机的信号、部署专门的无人机猎人进行拦截,以及开发和使用反无人机系统。

  • 无人机技术的未来发展趋势是什么?

    -无人机技术的未来发展趋势可能包括更强的自主性、更高级的导航和目标识别系统、以及更多样化的有效载荷。同时,随着对抗无人机的需求增加,反无人机技术和战术也将继续发展。

  • 无人机在乌克兰战争中的使用对全球军事战略有何影响?

    -无人机在乌克兰战争中的使用展示了其在全球军事战略中的潜力,特别是在提供战场感知、进行远程打击和执行精确任务方面。这可能会促使其他国家加大对无人机技术的投资,并重新评估其防御系统以应对无人机威胁。

  • 乌克兰战争中无人机的演变对未来战争有哪些启示?

    -乌克兰战争中无人机的演变表明,无人机技术可以在实战中迅速发展和适应,对未来战争的启示包括对无人机的依赖可能会增加,同时需要更加重视反无人机能力的发展,以及在战略规划中考虑无人机的潜在影响。

Outlines

00:00

🤖 无人机在乌克兰的演变与影响

本段落讨论了自2023年初以来乌克兰小型无人机技术的快速发展,以及它们在数量和能力上的显著提升。作者提出了关于这些小型无人机在乌克兰战争中的作用及其对全球未来战争可能产生的影响的重要问题。

05:00

🚀 无人机技术的进步与危险性增加

这一段分析了无人机技术如何变得更加危险,尤其是在过去一年中。作者通过所谓的“致死洋葱”模型,讨论了无人机系统为了最大化杀伤概率所需具备的特性,包括传感器、射程、制导和目标系统、有效载荷以及对抗对手防御的能力。

10:02

🌙 夜间作战能力的提高

本段讨论了无人机在夜间作战能力方面的技术进步,特别是在乌克兰战场上的应用。作者指出,尽管小型无人机在低光或夜间条件下的操作存在限制,但随着技术的发展,越来越多的无人机开始集成热像摄像头,提高了夜间作战能力。

15:03

📈 无人机射程的增加

这一段分析了无人机在射程方面的进步,特别是在乌克兰战争中。作者指出,随着时间的推移,小型无人机在更长的射程上发动攻击的能力有所提高,这得益于物理性能的提升和控制技术的进步,例如使用中继器来扩展信号范围。

20:04

🎯 无人机目标识别与制导的改进

本段讨论了无人机在目标识别和制导方面的技术进步,包括使用机器辅助进行目标识别和半自主/全自主的终端制导。作者分析了这些技术的优势和潜在弱点,并讨论了它们对未来战争的影响。

25:05

💣 无人机有效载荷的演变

这一段探讨了无人机有效载荷的演变,以及这对无人机在战场上的作用和目标选择性的影响。作者指出,有效载荷的增加不仅提高了无人机对特定目标的破坏力,还扩大了无人机可以执行的任务类型和目标范围。

30:06

🛠️ 无人机专用弹药的发展

本段讨论了为无人机操作专门设计的弹药的发展,以及这些弹药相对于传统弹药的优势和劣势。作者指出,专门的无人机弹药可能会提高操作的安全性和可靠性,同时提供更大的破坏力。

35:08

🔄 电子战与无人机对抗的演变

这一段分析了电子战在乌克兰战争中如何被用来对抗无人机,以及无人机制造商如何通过技术进步来抵抗这些对抗措施。作者讨论了电子战的频率变化、导航方法的改进以及其他古典导航方法的应用。

40:08

🏭 无人机生产的演变与挑战

本段讨论了乌克兰和俄罗斯在无人机生产方面的演变,包括从商业现成产品到定制系统的转变,以及这些生产方法对战争的影响。作者指出,无人机生产的增加对乌克兰和俄罗斯的战争努力至关重要。

45:10

🔍 无人机在乌克兰战争中的实际影响

这一段探讨了无人机在乌克兰战争中的实际影响,包括它们在战术、操作和战略层面的作用。作者分析了无人机如何改变了战场的透明度和危险性,并对无人机在未来冲突中的潜在影响提出了思考。

50:12

📊 无人机技术的未来趋势与挑战

最后一段总结了无人机技术的未来趋势,包括生产能力、技术进步和战术应用的潜在变化。作者强调了无人机技术的不可逆性,以及它对未来战争的深远影响。同时,作者还提到了频道更新和其他相关内容。

Mindmap

生产目标的大幅增加
技术改进的持续性
生产率和技术水平的提升
DJI Mavic 3等商用四轴飞行器
小型ISR平台
携带小载荷的攻击型无人机
武装无人机
高速、需要高技能的遥控无人机
FPV无人机
固定翼设计,用于打击高价值目标
远程弹药无人机
用于打击战略目标的固定翼无人机
一次性远程攻击无人机
无人机的分类与应用
无人机的感知、射程、精确度和抵抗力的进化
无人机技术的发展
增加态势感知和目标定位能力
提高战场透明度
对战术和战略层面的影响
无人机对战争的影响
电子战和反无人机系统的使用
无人机的抵抗力和对抗措施
无人机在乌克兰的应用
自主目标识别和制导技术的发展
载荷和射程的增加
技术进步的持续推动
无人机与传统武器系统的结合
无人机在非对称战争中的潜在作用
无人机在军事战略中的角色
未来无人机的发展趋势
无人机的全球影响
从现成的商用无人机到定制的军事无人机
从商用到军用
从依赖外国供应链到国内生产
从手工作坊到规模化生产
生产方式的变革
对战场需求变化的快速适应
分布式生产网络的抗打击能力
无人机的可扩展性和适应性
无人机产业的演变
无人机在乌克兰的演变及其全球影响
Alert

Keywords

💡无人机

无人机(UAV)是指无需搭载操作员即可飞行的飞行器。在视频中,无人机被广泛用于乌克兰战争中,执行侦察、监视、目标定位和攻击等任务。它们由于成本低廉、操作灵活,成为现代战场的重要组成部分。

💡电子战

电子战是指利用电子设备进行的战斗行动,包括干扰、欺骗和破坏敌方电子设备的正常工作,以及保护己方电子设备不受敌方影响。在视频中,电子战是对抗无人机的关键手段之一,通过干扰无人机的通信和导航系统来阻止其执行任务。

💡自主导航

自主导航是指无人机或其他无人系统能够在没有外部指令或控制的情况下,根据预设的程序或实时感知环境来规划和执行飞行路径的能力。这一技术在视频中被提及,是无人机避免电子战干扰和提高生存能力的关键技术之一。

💡有效载荷

有效载荷是指无人机或其他运载工具携带的用于执行特定任务的设备或物品,如炸弹、侦察设备等。在视频中,有效载荷的种类和大小直接影响无人机的作战效能和任务范围,例如携带更大威力的弹头可以提高对装甲目标的破坏力。

💡生产能力

生产能力指的是制造无人机的效率和数量,这直接影响到战场上无人机的供应和使用规模。视频中提到乌克兰和俄罗斯都在大力提升无人机的生产能力,以满足战争中对无人机的大量需求。

💡战术和战略影响

战术和战略影响是指无人机在战场上的应用对军事行动和战争结果产生的直接和间接效果。视频中讨论了无人机如何改变战场态势,提供实时情报,以及在战术层面上对敌方造成压力,同时也探讨了无人机在战略层面上的作用和限制。

💡反无人机系统

反无人机系统是指专门设计用来对抗无人机威胁的技术和设备,包括干扰器、捕网枪、激光武器等。在视频中,反无人机系统是防御方为了抵御无人机攻击而采取的关键措施之一。

💡军事工业复合体

军事工业复合体是指军事和工业之间的紧密联合,它们共同推动军事技术的发展和军事装备的生产。在视频中,军事工业复合体在无人机的发展和生产中扮演了重要角色,特别是在提高生产效率和降低成本方面。

💡ISR

ISR是情报、监视和侦察(Intelligence, Surveillance and Reconnaissance)的缩写,指的是通过各种手段收集战场信息的活动。在视频中,无人机在执行ISR任务时,能够提供实时的战场情报,对作战行动至关重要。

💡自主或半自主制导

自主或半自主制导是指无人机在攻击目标时,能够在一定程度上或完全不需要人工干预的情况下,自动或半自动地完成目标识别和攻击的过程。这一技术在视频中被提及,是提高无人机作战效能的关键发展方向。

Highlights

2023年初,乌克兰小型无人机的技术水平和生产速度已经相当惊人,仅仅一年多之后,无论是数量还是能力都有了巨大的飞跃。

乌克兰战场上小型无人机的使用已经改变了战场的透明度和危险性,为双方提供了前所未有的战场感知能力。

无人机的种类繁多,从小型的ISR平台到远程一次性攻击无人机,它们的功能和成本差异巨大。

无人机技术,尤其是针对乌克兰的战术发展迅速,一些最新的技术和战术发展通常比较敏感,因此本文主要关注一两个月前的例子。

无人机的传感器能力正在提升,尤其是在夜间操作方面,这使得无人机在更多时间段内能够识别目标。

无人机的射程正在增加,这使得它们能够打击更远的目标,扩展了无人机的战术和操作用途。

无人机的自主性正在提高,一些系统开始采用半自主或完全自主的导航和目标识别技术。

无人机的有效载荷正在变得更大和更专业,这使得它们能够对更多种类的目标造成更大的伤害。

乌克兰和俄罗斯都在大量投资无人机生产,这导致了无人机数量的大幅增加,改变了战场的平衡。

尽管无人机在战术和操作上取得了成功,但它们在战略层面上的影响仍有争议,一些观点认为无人机并未改变乌克兰战争的结果。

无人机的部署和使用正在不断进化,乌克兰的经验可能会对未来的战场无人机系统和战术产生深远影响。

无人机的生产和部署模式正在从简单的商业现成产品改装转向更加定制化和专业化的设计。

电子战是对抗无人机的主要手段之一,但无人机制造商也在不断开发新的抗干扰技术。

乌克兰和俄罗斯都在努力提高无人机的生产能力,预计未来无人机的数量和质量还将继续增长。

无人机在乌克兰战争中的作用不容忽视,它们已经成为战场上不可或缺的一部分,并将继续影响战争的走向。

无人机的发展和应用提出了新的挑战,包括如何防御这些系统以及如何适应它们带来的战场变化。

无人机技术的发展可能会迫使军事战略和战术发生根本性的变化,特别是在对抗同等级别的对手时。

Transcripts

00:00

in early 2023 I covered the evolution of

00:02

small drones in Ukraine when the

00:04

technology level and production rates

00:06

were already scary but now just over a

00:08

year later the leaps in quantity and

00:10

capability have been immense production

00:12

targets have ramped up in some cases by

00:15

more than an order of magnitude and even

00:17

as the quantities have ramped up the

00:18

technology involved has continued to get

00:20

better and with no sign of that

00:22

Breakneck engineering and Industrial

00:24

Evolution slowing down I think it's time

00:26

to ask some important questions how are

00:28

we seeing these small uas of in Ukraine

00:31

what impact are they having on the War

00:32

and what might it mean for the rest of

00:34

the world when this technology and the

00:35

tactics developed for its use inevitably

00:38

start to go Global to do that as normal

00:41

I'm going to break this up into a couple

00:42

of segments we'll start with an overview

00:44

of some of the most common types of

00:45

drones that are operating in Ukraine and

00:48

how they may be getting better looking

00:49

at how some of these drones are evolving

00:51

in terms of their senses range targeting

00:53

lethality or countermeasure resistance

00:56

then we'll go to the second scary part

00:57

of the picture because drones AR just

00:59

getting deadly they're getting more

01:01

common so I'll cover some elements

01:02

relating to production estimates and

01:04

techniques before closing with a zoomed

01:06

out look at just how impactful these

01:07

systems have been and what their role

01:09

might be going forward some quick

01:11

caveats then in terms of scope and

01:13

content firstly while there are a huge

01:15

array of drones out there to keep things

01:16

manageable I'm going to focus just on

01:18

smaller unmanned aerial systems that are

01:20

used in attack rols so bomber drones

01:23

fpvs loading Munitions and oneway attack

01:25

systems I've talked about dedicated ISR

01:28

drones like the Russian orine before so

01:30

they won't be in this episode nor will

01:31

Naval or ground drones that very much

01:33

deserve their own billing I also need to

01:35

flag that drone technology and the

01:36

tactics for using the me Ukraine are

01:38

evolving incredibly quickly and some of

01:40

the very latest Technical and tactical

01:42

developments are usually fairly

01:43

sensitive so I'll generally try and

01:45

focus on examples that are at least a

01:46

month or two old because I want to put

01:48

together a high level picture here not a

01:50

how-to guide for example on how to build

01:53

a more Jam resistant fpv drone okay so

01:56

let's do some quick categorization of

01:57

the sort of drones we're going to be

01:58

talking about today ranging from small

02:01

ISR platforms all the way through to

02:03

longrange oneway attack drones your

02:05

smallest ISR drones are usually going to

02:07

be quadcopters sometimes octocopters

02:10

think you're off-the-shelf DJI mavic 3

02:12

or equivalent for example and while

02:14

they're probably intended to be reusable

02:16

they are cheap relatively short ranged

02:19

and a lot of times they're being used

02:20

for things like increasing situational

02:21

awareness identifying attacks calling in

02:24

and correcting artillery and giving even

02:26

relatively low Echelon units the ability

02:28

to metaphorically see what is on the

02:30

other side of the next Hill weaponize

02:32

one of those quad octocopters and you

02:34

might have yourself what are sometimes

02:35

called drop drones or bomber drones

02:37

these keep the same fundamental

02:39

characteristics as the base system so

02:41

intended reusability for example but

02:43

they add a small payload to be used in

02:45

attacks for example knocking out

02:46

disabled Vehicles by dropping grenades

02:48

through hatches there is of course a

02:50

massive variation in scale and cost here

02:52

ranging from off-the-shelf hobby drones

02:54

dropping hand grenades all the way

02:55

through to purpose-built monsters

02:57

capable of cting around artillery

02:58

projectiles but the basic concept is

03:01

broadly similar fpvs or firstperson view

03:04

drones are also generally quadcopters

03:06

but with a different piloting method

03:08

these originally developed out of racing

03:10

drones so compared to the more

03:11

reconnaissance focused quadcopters they

03:13

tend to be much faster require a lot

03:15

more in terms of pilot skill and are

03:17

also often a lot cheaper a lot of these

03:19

systems tend to be intended as

03:21

single-use attack options against

03:23

tactical targets so you're still

03:25

operating in a relatively close range

03:26

bracket 10 km for example is a pretty

03:29

longrange FP PV strike but you're

03:31

probably doing it with a lower cost a

03:32

bigger boom single use only although as

03:35

always those points should come with a

03:36

giant asterisk cuz there are fpv designs

03:39

out there that are designed to operate

03:40

as reusable bombers Laing Munitions in

03:42

Ukraine like the Russian Lancet tend to

03:44

represent a bit of a step change from

03:46

the fpvs the designs are usually fixed

03:48

Wing not quad rotors so optimized for

03:51

Speed and range and are often going to

03:52

be used by specialized teams at ranges

03:54

of tens of kilometers to hit point

03:56

valuable targets Lance it in some of its

03:59

Ukrainian equivalent for example are

04:00

perhaps best known for their counter

04:01

battery role hunting enemy artillery

04:04

systems that aren't exactly going to be

04:05

parked in a Frontline trench finally you

04:07

have longrange strike or one-way attack

04:09

drones here you're still mostly looking

04:11

at fix wi designs that are intended to

04:13

be single use only but the range payload

04:15

and intended target type are all going

04:17

to be different we're talking ranges in

04:19

the hundreds of kilom not tens of kilom

04:21

targets that are operational or

04:23

strategic in nature think fuel

04:25

refineries or air bases for example and

04:27

usually a larger payload to reflect both

04:29

the different Target type and the fact

04:31

you're probably not going to get

04:32

pinpoint accuracy the Iranian design

04:34

shed 136 probably fits into this

04:36

category along with a number of

04:37

Ukrainian designs and you can see them

04:39

as representing a sort of bridge between

04:41

the other cheaper drone systems and

04:43

traditional long-range Munitions like

04:45

cruise missiles compared to a cruise

04:47

missile the oneway attack drones we've

04:48

seen so far tend to be much slower have

04:51

smaller warheads and be missing a lot of

04:53

the technological bells and whistles but

04:55

they also tend to be available at a

04:57

fraction of the cost while being much

04:58

easier to produce

05:00

okay so now we do get to the scary part

05:02

where we start talking about how these

05:03

systems which are increasingly

05:05

everywhere and very affordable are also

05:07

continuously evolving and have become

05:09

more dangerous even just over the last

05:11

year but in order to give an overview of

05:13

some of the factors that might be making

05:15

drones in Ukraine more dangerous over

05:17

time what I thought I'd do is take a

05:19

concept that we've looked at before the

05:21

survivability onion and flip it around a

05:24

little the survivability onion usually

05:26

just describes all the things that have

05:27

to go wrong sequentially for you to get

05:29

killed in your vehicle on the

05:31

battlefield you have to be somewhere the

05:33

enemy is a looking and B can reach and

05:35

then you have to be seen identified

05:37

targeted and hit by something that is

05:39

lethal enough to penetrate your defenses

05:42

and do significant damage so in order to

05:44

identify some changing Trends in drone

05:46

lethality we're going to take that same

05:47

concept and sort of flip it around

05:50

instead of a survivability onion we have

05:51

a much tastier and more traditionally

05:53

Australian lethality lamington a layered

05:56

list of the characteristics your system

05:57

is going to need to maximize the

05:59

probability of a kill you need the

06:01

sensors to see a Target the range to

06:03

reach it a guidance and targeting system

06:05

to hit it and enough of a payload to

06:07

make sure they know that they've been

06:09

hit and of course you want your system

06:11

to be as resistant as possible to

06:12

whatever countermeasures or defenses

06:14

your opponent might be using to maximize

06:16

the probability that you make it to the

06:17

end of that process we've observed

06:20

technical changes in Ukraine that impact

06:21

each layer of the lamington so now what

06:23

I'll do is go through some of them in

06:25

sequence okay so step one for our

06:27

hypothetical drone system is finding a

06:29

Target

06:30

and here the war in Ukraine has already

06:31

witnessed some pretty significant

06:33

technical improvements particularly when

06:34

it comes to night operations while we

06:37

talk about flocks of drones being an

06:38

omnipresent fact of life on the

06:40

Ukrainian Battlefield the reality is

06:42

it's actually a lot of conditions during

06:44

which a lot of the systems out there

06:46

just can't operate effectively bad

06:48

weather can be even more of a problem

06:50

for a lot of uas than it is for manned

06:51

aircraft and while already in 2022 a lot

06:54

of larger and more expensive drones were

06:56

equipped with things like thermal and

06:57

Night Vision Optics for the vast

06:59

majority of fpvs and quad rotors out

07:01

there the sensor Suite on board which

07:03

would often be just a simple visual

07:05

camera didn't operate effectively in low

07:07

light or nighttime conditions indeed

07:10

what you get with a lot of fpv drones is

07:12

visual Spectrum only just like the Mark

07:13

1 eyeball only with resolution levels

07:16

reminiscent of the Xbox 360 and a level

07:18

of grain and interference that wouldn't

07:20

look out of place in an old analog TV

07:22

broadcast the obvious tactical

07:24

adaptation that that sort of limitation

07:26

imposed is that in environments where

07:28

the opponent has the ability to fly a

07:29

lot of these drones we have seen reports

07:32

of Russian and Ukrainian forces doing

07:33

for example a lot of their low-level

07:35

Logistics and resupply operations and

07:37

maneuver at night now of course there

07:39

were still small drones and ISR assets

07:41

that could see you even if you're

07:43

maneuvering after Sundown as well as the

07:45

quite Advanced sensor packages on things

07:47

like the tb2 Drone smaller tactical

07:49

level uas also had some night vision

07:51

capability off the shelf DJI drones for

07:54

example have some thermal options the

07:56

30t and the 3T being some of the most

07:58

common and that was fine if your goal

08:00

was to see enemy units maneuvering at

08:01

night and do something like call in

08:03

artillery on them because unguided

08:05

artillery shells don't tend to care

08:06

about the ambient light level when

08:08

they're addressed to a particular set of

08:09

grid coordinates but that Arrangement

08:11

brings with it two limitations one if

08:14

there is a target type that you would

08:15

rather engage with something like an fpv

08:18

as opposed to artillery perhaps for

08:19

example because you're short of

08:21

artillery ammunition that might be

08:23

difficult to do when the fpvs can't fly

08:24

at night and so you either inefficiently

08:26

task artillery to deal with the problem

08:28

or you let the Target go the other

08:31

limitation was that thermal camera

08:32

equipped drones were generally just much

08:34

more expensive than those that didn't

08:36

have that level of bling which means

08:38

both a probably not everyone gets to

08:40

have one and secondly if you want to

08:42

make cheap fpvs thermal capable you

08:44

probably can't do it by adding those

08:46

sort of thermal optics otherwise your

08:48

cheap fpv isn't going to be so cheap

08:50

anymore make no mistake it'll probably

08:52

still be cheap compared to many of the

08:53

other weapon systems on the battlefield

08:55

but is it really the quintessential you

08:57

cran War experience if you're not tou

08:59

touching off multi-million dollar

09:00

armored vehicles with something that

09:02

cost a couple hundred bucks and yes I am

09:04

joking around a lot but it's mostly to

09:06

distract from the fact the next bits

09:07

kind of terrifying as we go into 2024 I

09:10

think there are plenty of signs that

09:11

more and more drone systems including

09:13

relatively cheap fpvs are going to be

09:15

capable of operating at times that were

09:17

previously at least if not safe maybe

09:20

safer we've seen more and more fpvs

09:22

Laing Munitions and drop drones that

09:24

integrate thermal cameras into their

09:26

designs and you also see some Innovative

09:28

approaches like the one mentioned on the

09:30

right there that's a post by the

09:32

Ukrainian organization wild Hornets that

09:34

manufacture fpv drones showing off what

09:36

they describe as quote an affordable

09:39

solution for nighttime fpv drone

09:41

operations end quote and when they say

09:43

that they don't mean it in the corporate

09:44

buzzword cense they actually mean

09:46

affordable as in $50 US per unit at

09:50

$11,000 for a thermal camera you might

09:52

be tripling the cost of your average fpv

09:55

at $50 per drone it's only a 10% premium

09:58

for giving something that ability to

09:59

operate at night and at that point

10:01

widescale adoption might start to make a

10:04

greater degree of sense so at the big

10:06

picture level what's changed and what

10:08

matters basically if in 2022 thermal or

10:11

night vision was an expensive rare

10:13

capability for small drones in 2024 the

10:16

direction of travel has been towards

10:17

cheaper and more common all else being

10:19

equal you should expect that means that

10:21

in the future there'll be more drones

10:22

out there capable of identifying more

10:24

targets during more hours of the day the

10:26

protection that nighttime provided which

10:28

was already highly imperfect is likely

10:30

to be reduced and the

10:32

military-industrial complex as a whole

10:33

will take one more gigantic step towards

10:36

what I can only assume is the ultimate

10:38

goal of equipping every drone with a

10:39

mechanical equivalent of the eye of

10:41

saour on for less than the cost of a

10:42

Happy Meal okay so increasingly drones

10:44

can solve problem number one of needing

10:46

to be able to see a Target what about

10:48

actually being able to reach it after

10:50

all just because I can see the top of a

10:51

mountain doesn't mean that I can climb

10:53

to it the war in Ukraine has arguably

10:55

really highlighted the value of weapons

10:57

with longer reach explo the limitations

11:00

of your opposing systems reach has also

11:02

been one way commonly used to protect

11:04

important targets when the Russians

11:05

pulled a lot of their ammunition depos

11:07

out of high Mars range that was a

11:09

range-based adaptation and while yes

11:11

there are some limitations on the

11:12

ability of forces to just move

11:14

everything of note out of opposing range

11:16

good luck shifting an oil refinery for

11:18

example particularly in the earlier

11:19

stage of the invasion a lot of the

11:21

cheaper drone systems out there had

11:23

comparatively limited endurance and

11:25

practical range now while there is

11:27

always a lot of uncertainty in this anal

11:29

Anis here I just want to flag there's

11:30

more uncertainty than normal but if

11:32

you've gone through some of the relevant

11:34

publicly released interviews and

11:35

reviewed a lot of the Drone footage out

11:37

there that has been geolocated from 2022

11:39

and 2023 one thing you might notice is

11:42

that as time has gone on the number of

11:44

attacks being launched by small uas like

11:47

fpvs at longer and longer ranges has

11:50

increased and we've seen that Trend

11:52

towards increased practical range play

11:54

out across a range of systems with

11:55

various effects for an fpv drone it

11:58

might mean being able to hear get

11:59

relatively short- range artillery

12:00

systems like the toss 1A I think we've

12:03

actually seen videos of two fpv strikes

12:05

on those systems in the last week at

12:06

time of recording but also they use in

12:08

more of an interdiction role where

12:10

you're attacking Supply trucks or

12:11

reinforcements trying to move up to the

12:13

front line meanwhile for a system like

12:15

lanet that started with a double digit

12:17

range enhancing it considerably might

12:19

mean putting longer range artillery

12:21

systems or things like Forward Air

12:23

strips that were out of range in range

12:25

with unfortunate results for example for

12:27

the occasional Ukrainian Mig 9 airframe

12:30

that's been caught out of position in a

12:32

sense when you see the range of these

12:33

systems improving often there are two

12:35

different elements that are moving

12:37

together first are the improvements to

12:39

the physical performance of the system

12:41

how far can it physically fly in that

12:43

sense the move from offthe shelf drones

12:45

to more dedicated military models has a

12:48

lot of potential compared to a civilian

12:50

version a military fpv might have a lot

12:52

of [ __ ] you don't strictly need stripped

12:54

off it doesn't need to last particularly

12:56

long or look particularly good so you

12:57

can make some economies there and to an

12:59

extent you can also always just add more

13:01

fuel tank Edge or battery problem two

13:03

then is trying to find a way to control

13:05

said drone when it does fly further here

13:08

again from a mechanical perspective

13:09

there are a couple of related but still

13:11

distinct problems firstly you just need

13:13

an underlying signal strength and

13:14

quality that lets you reach the

13:16

requisite distance especially in a

13:18

battlefield context where both sides are

13:19

likely to have ew guys roaming around

13:21

the place doing their best to do

13:23

selective violence to different parts of

13:24

the electromagnetic spectrum the second

13:26

problem is just maintaining signal line

13:28

of site with all due respect to the Flat

13:31

Earth is out there the Earth isn't so

13:33

all else being equal the further a drone

13:34

flies at low altitude the more likely it

13:36

is that a bunch of dirt Rock and burnt

13:38

out tanks get between it and the

13:40

transmitter and while you can compensate

13:42

to an extent by increasing altitude to

13:45

maintain line of sight longer that also

13:47

leads into a phenomena we've seen in

13:49

lots of videos of fpv and other drone

13:51

attacks where during the final terminal

13:54

phase of an attack the Drone Dives

13:55

towards the ground because that's where

13:57

the target is and you can pH ially see

13:59

the signs of the signal breaking up in

14:01

real time to an extent a good fpv

14:04

operator knows that disruption is likely

14:06

to come and can pre- aim the Drone so

14:08

it's going to make the hit regardless

14:10

but it's still a factor relating to

14:11

signal range and quality that can

14:13

significantly impact just how effective

14:15

these weapons practically are okay so

14:18

how do you then increase the range of

14:19

the signal well one approach we've seen

14:21

used is the employment of repeaters

14:23

where you might have one or more drones

14:25

that instead of carrying Munitions or a

14:27

sensor package is carrying a signal

14:29

repeater these then can form an aerial

14:31

relay noting there's no reason you have

14:33

to put them on the Drone you can have

14:34

groundbased repeaters as well but

14:36

putting them on the drones has an

14:37

obvious Advantage from a line of sight

14:38

perspective now instead of having to

14:41

send a signal to the Drone that's making

14:42

the attack you just have to reach the

14:44

repeater which then has to reach the

14:45

next repeater which can then reach the

14:47

Drone that is making the attack if you

14:49

think of this from a line of sight

14:50

perspective a repeater that is

14:52

relatively close to the side of the

14:53

attack is less likely all else being

14:55

equal to lose sight of an fpv making an

14:57

attack in its terminal phase and so an

14:59

operator might be more able to make

15:01

those last minute flight path

15:02

adjustments that lead to a hit rather

15:04

than a swing and a mess repeaters also

15:06

have potential implications for the

15:08

survivability of the Drone operators

15:10

themselves it means you might not need

15:12

the Drone operator metaphorically

15:13

kneeling in a Frontline trench in order

15:15

to get the maximum possible reach out of

15:17

their drones instead you can send the

15:18

repeaters forward while The Operators

15:20

are sheltered in a position further back

15:22

depending on the particular setup and

15:24

opposing capabilities you might still

15:25

have a signature problem from a

15:27

concealment perspective but you may

15:28

still end up being more survival as a

15:30

drone operator using this sort of system

15:32

another thing to note here is that the

15:34

repeater drone and the drones it's

15:35

controlling don't have to be the same

15:37

design in fact that might be a

15:39

disadvantage I feel like I can talk

15:41

about this case because it has a casual

15:42

732 th000 views on Twitter but recently

15:45

there was a case of a Russian serviceman

15:48

describing a situation where Ukrainian

15:50

drone operators allegedly attacked a

15:52

position using a larger so-called Queen

15:54

drone with a signal repeater controlling

15:56

a bunch of smaller cheaper fpvs the

15:59

possibilities of that sort of

16:00

configuration are obviously massive

16:02

because a lot of signals equipment you

16:03

could fit on a larger drone that you

16:05

can't and probably don't want to try and

16:07

fit onto an fpv a large enough quad or

16:10

octocopter for example might be able to

16:11

carry a satellite internet terminal

16:13

worth thousands of dollars which would

16:15

not exactly work from any perspective if

16:18

you tried it with an fpv plus of course

16:20

in this environment a lot of the most

16:22

expensive equipment is concentrated on a

16:24

reusable system the queen drone rather

16:27

than the Disposable systems the PVS that

16:29

are going and slamming into things if

16:31

the queen isn't shot down it can be

16:33

recalled and then sent out again with a

16:35

new wing of supporting attack drones

16:37

it's worth noting that this idea of a

16:38

queen system providing information and

16:40

controlling other systems in a flock is

16:42

not at all new You could argue that the

16:44

pairing of ISR drones and Laing

16:46

Munitions already to an extent fits that

16:49

definition with the ISR drones generally

16:51

carrying the more expensive sensor and

16:53

communication package and the loitering

16:55

munition carrying the Warhead and a mad

16:57

maxian desire to be witnessed back

16:59

during the Cold War for example the

17:00

Soviets introduced the p700 with the

17:02

highly creative NATO reporting name

17:04

shipwreck which had an attack mode where

17:06

a flight of missiles directed towards an

17:08

enemy Fleet would have a single missile

17:10

act as the Target designator that Queen

17:13

missile would climb to a higher altitude

17:15

so it could better see the potential

17:16

targets and communicate via data link

17:18

with the others that were hiding closer

17:19

to the deck if the queen perhaps owing

17:22

to its higher altitude and more exposed

17:23

position were shot down a drone would be

17:25

promoted to Queen Judy climb to altitude

17:28

and assume where the previous missile

17:30

left off here there was no physical

17:32

difference between the different p700 in

17:34

the Salvo but you still have a split

17:36

between attacking missiles and a

17:38

coordinating or Target designating

17:39

missile in order to help solve the

17:41

underlying line of side issues so we

17:43

probably shouldn't be that surprised

17:45

more than 2 years into the invasion of

17:46

Ukraine that this old concept is being

17:48

leveraged to deadly effect so in summary

17:51

and as always this is oversimplifying if

17:53

in the early stages of the war the norm

17:55

was for small drones to attack Frontline

17:57

targets and face significant line of

17:58

side issues in 2024 The Meta if you will

18:02

increasingly involves long range

18:03

interdicting strikes supported by

18:05

technologies that mean that range and

18:07

line of sight are no longer always as

18:09

much of an issue the reason that matters

18:11

is because it establishes more and more

18:13

tactical and operational use cases for

18:15

these systems at 2 or 3 km your anti-

18:18

armor fpvs can supplement the role of

18:20

something like a javelin missile but if

18:22

you start reaching out to 10 km you're

18:24

playing in territory that might normally

18:25

be serviced by indirect fire options

18:27

artillery and once you have loitering

18:29

Munitions that might be able to go out

18:31

70 km you're playing in territory that

18:33

was previously dominated by long range

18:36

Precision Rock and artillery and air

18:37

support that's obviously a factor on the

18:39

battlefield because you're putting more

18:41

and more of the Tactical and operational

18:42

depths at risk from systems that might

18:45

be able to economically engage targets

18:47

that aren't worth dumping an entire

18:49

battery of mrls on it may also raise

18:51

questions in some armies as to who

18:53

should control these longer range Fire

18:55

Systems because suddenly you might have

18:57

a cheap system whose Prim Target is

18:59

tanks and armored vehicles so you think

19:01

it belongs wherever the atgms go but it

19:03

has a range bracket that's much much

19:05

longer than your traditional platoon

19:07

level asset you have a family of weapon

19:10

systems here with the potential to for

19:11

lack of a better word democratize access

19:13

to longrange precision and it may not be

19:16

fully understood just yet what all the

19:18

implications of that are going to be but

19:20

moving on what all those changes have

19:22

basically done at this point has

19:23

increased the odds that our drones will

19:25

find a Target and then be able to get to

19:27

it the next step and a pretty important

19:29

one at that is actually hitting it even

19:32

though you can argue that drop drones

19:33

laery Munitions fpvs are all Precision

19:36

weapon systems they all after all have

19:38

guidance systems and tend to go after

19:40

Point targets even if the guidance

19:42

system in question is often just the

19:43

brain of the pilot of the controls it's

19:45

important to understand that in practice

19:47

the evidence suggests that these things

19:49

do miss a lot and it's an area where I

19:51

think it's pretty logical to assume that

19:52

the video evidence available to us is

19:54

probably a bit skewed sure we do get

19:56

videos of drones and Laing Munitions

19:58

appear to miss their targets fairly

20:00

often just like people on social media

20:02

who might love to post images of the

20:03

Fast Car the fancy vacation or gour

20:06

dinners while conveniently leaving out

20:08

images of their credit card statements

20:10

in drone Warfare it's the operator who

20:12

holds the footage and makes the decision

20:13

on releasing it and so you probably get

20:15

to see the $500 drone slamming into the

20:18

$20 million air defense system or

20:20

totaling a t9m and what you don't see

20:22

are the drones that Miss or are jammed

20:24

out or miss or suffer technical failure

20:27

or Miss remembering of course that a lot

20:29

of these systems are hard to fly built

20:31

by the lowest bidder and often provide

20:33

the operator with a solid potato tier

20:35

video feed one Trend we've seen which

20:38

might have an impact on that particular

20:39

problem is the increased utilization of

20:42

machine assistance for Target

20:43

identification and also for guidance

20:46

these sort of visual targeting solutions

20:47

for example might include a system that

20:49

looks at a video feed and flags things

20:51

that might be a potential Target Andor

20:54

once the target has been selected and

20:55

designated by a human operator allow the

20:58

weapon to guide in without further Human

21:00

Assistance Russian sources for example

21:03

often claim that the latest versions of

21:04

the Lance at loitering munition are

21:06

capable of autonomous or semi-autonomous

21:08

guidance basically the system can look

21:10

at its surroundings identify things that

21:12

look like a tank an infantry fighting

21:14

vehicle or a piece of farm machinery say

21:16

hey I think that thing over there might

21:18

be a Target and then depending on the

21:20

exact autonomy settings the information

21:22

available is very vague at the moment

21:24

either with or without a thumbs up from

21:26

a human go in and engage one of those

21:28

targets

21:29

the exact answer to the question of how

21:31

autonomous systems like lan can be is

21:33

still very much contested you have

21:35

claims that run the full gamut from the

21:37

system is just highlighting potential

21:38

targets to give the operator a clue as

21:40

to what he might want to fly into all

21:42

the way up to this is an AI powered

21:44

super drone that can recognize the

21:46

Ukrainian tank that the human eye would

21:47

never spot and then autonomously guide

21:49

to it but while there's a lot of debate

21:51

about how autonomous exactly systems

21:54

like Lancet are we do know we've seen a

21:56

significant amount of footage from

21:57

Russian sources of lanced attacks being

22:00

launched with some sort of mechanical

22:02

assistance appearing to be in play at

22:04

the moment at least the evidence seems

22:05

to suggest that the systems that are out

22:07

there including the one on laner isn't

22:10

exactly perfect think more 10% Skynet

22:13

50% chat GPT and 30% that New Zealand

22:16

Supermarket AI that suggested people put

22:18

mosquito repellent in their potatoes or

22:20

turpentine in their French toast okay so

22:23

to make this just a little bit more real

22:24

what I want to do is now dissect some

22:26

Stills taken from the footage of a

22:28

recently documented Russian Lancet

22:30

attack this is footage released by a

22:32

pro-russian source that appears to show

22:33

at least some autonomous targeting

22:35

assistance being used and I think it

22:37

demonstrates both the strengths and also

22:39

potential weaknesses of systems like

22:40

this at least right now we'll start with

22:43

this image I have on screen there from

22:44

relatively early in the video and while

22:46

those of you on mobile phone screens

22:48

might be struggling a little bit I think

22:49

most humans would be able to pass this

22:51

image pretty successfully that large

22:54

object at the end of the treadmarks that

22:55

looks awfully like an armored vehicle is

22:57

in fact an armored vehicle I don't know

23:00

about you but for me it was the turret

23:01

and Main gun system that kind of gave it

23:03

away all of the stuff immediately

23:05

surrounding the vehicle including the

23:07

bit I've highlighted for reasons which

23:09

will become obvious in a moment lack

23:11

those same obvious visually identifying

23:13

features and are probably not the

23:14

primary target for this sort of system

23:16

next we have a still seemingly taken

23:18

from the sequence where the loitering

23:19

munition is approaching the potential

23:21

Target at this point you see a green

23:23

bounding box appear on the right hand

23:24

side of the screen which suggests that

23:26

maybe the automated targeting system is

23:28

identifying another potential Target off

23:30

to the right but the munition still

23:32

potentially under operator control

23:34

continues towards that original Target

23:36

we were just looking at a top down image

23:38

of before and indeed you can see that

23:40

Central Crosshair over that Target at

23:42

this point and as the munition

23:44

approaches that Target the expectation

23:46

is probably that the automated system

23:47

will kick in identify the Target and

23:50

assist with the Final Approach so of

23:51

course the munition races towards the

23:53

Target and then proceeds to not identify

23:56

it instead you can see it throws a

23:58

bounding box around what appears to be

24:00

an object that the potential Target is

24:02

hiding behind we can go into theories

24:04

about why that misidentification may

24:06

have been made but the fact is it

24:08

happens with negative consequences for

24:10

the engagement basically the machine was

24:12

fed a real life version of one of those

24:14

visual capture problems and it appears

24:16

to have proceeded to fail in spectacular

24:18

fashion other footage I've looked at has

24:20

shown cases of vehicles being correctly

24:22

identified as potential targets but also

24:25

bounding boxes being thrown around

24:26

things that are very obviously not

24:28

military equipment and other cases where

24:30

military equipment that is fairly

24:32

obvious doesn't get the automated Target

24:35

recognition treatment and some observers

24:37

have noted that after a rush of videos

24:39

it appeared to show lanets being used

24:40

with this sort of autonomous or

24:42

semi-autonomous targeting system it's

24:44

being comparatively rarer or absent in

24:47

more recent releases potentially

24:48

suggesting although this is very much

24:50

straight guesswork that potentially the

24:52

feature got some Battlefield beta

24:54

testing attracted a little bit of

24:56

customer feedback so to speak and may

24:58

now be in a revision and itation phase

25:01

but here's the thing there is every

25:03

reason to believe that better versions

25:04

of this technology are coming and coming

25:06

quickly we'll probably talk about this

25:08

more in the future but when you're

25:09

trying to make an AI driven visual

25:11

targeting solution work there's a couple

25:12

of components that go into it and some

25:14

of those components are the literal

25:16

components the things that give it the

25:18

computing power it needs to make this

25:19

solution work that compute is already

25:21

lighter and cheaper than it has been at

25:23

any other point in human history and

25:25

it's only going to get lighter and

25:26

cheaper there was some reporting I saw

25:28

that suggested that some lancets might

25:30

be using Nvidia Jetson tx2 modules to

25:32

provide some of that compute I did ask

25:34

some people who keep a down Lancet as a

25:36

sort of office pet whether or not it

25:38

contains such a module but the version

25:40

they have appears to be a pre- autonomy

25:41

version so I can't confirm one way or

25:43

the other but as a hypothetical exercise

25:46

that sort of embedded AI Computing

25:47

device only goes for a couple of hundred

25:49

US and the hardware that's available and

25:51

the price point it's available at is

25:53

improving quickly another main input to

25:55

getting good outcomes out of AI is to

25:57

have a lot of good data to train it on

25:59

and the opportunity to iterate and

26:01

develop and so whether the rest of the

26:03

world likes it or not the war in Ukraine

26:05

where drones are being used on a massive

26:07

scale in combat conditions might become

26:09

a sort of Nursery trining ground for the

26:12

targeting AI of the future it's also

26:14

concentrating a lot of resources human

26:16

and financial Capital potentially

26:18

creating that sort of Precinct of

26:20

capability that sort of Rapid

26:21

development environment where you have

26:23

the full array of Engineers Specialists

26:25

and end users concentrated communic and

26:28

capable of Rapid development and

26:30

iteration so in Broad summary Strokes if

26:33

in 2022 and early 2023 drone targeting

26:36

Ukraine was mostly a matter of pilot

26:38

skill experience intuition and

26:41

occasionally just hoping that you didn't

26:42

lose signal at an inopportune moment in

26:44

2024 more of the options were seeing

26:46

deployed include at least some degree of

26:48

computer assistance would that just be

26:51

in the area of Target identification and

26:53

recognition or in the field of semi or

26:55

fully autonomous terminal guidance and

26:57

as to why High developments like that

26:59

might matter I think we're only now

27:00

starting to unpack the potential

27:02

implications camouflage and concealment

27:04

may become even more difficult if you

27:06

need your scheme to fool both the human

27:08

operator and the automatic recognition

27:09

system the need to have a highly skilled

27:11

loitering munition or fpv operator might

27:14

be somewhat reduced which matters a lot

27:16

given just how difficult training can be

27:18

and the premium that's currently being

27:19

placed on skilled operators plus and

27:22

perhaps most frighteningly it might

27:24

offer a partial answer to jamming

27:26

because even if electronic warfare can

27:28

successfully disrupt communication

27:30

between an operator and their drone

27:32

that's not going to stop the attack if

27:33

the Drone is capable of autonomously

27:35

identifying its own targets and guiding

27:37

into them okay so at this point the

27:39

drones have been able to fire a Target

27:41

reach the Target and then with a little

27:43

bit of mechanical assistance hit the

27:44

target let's now talk about the element

27:46

that actually makes that hit mean

27:47

something the payload the system is

27:49

carrying and just how lethal it is after

27:52

all without an effective payload

27:54

slamming a drone into something like a

27:56

tank is probably going to be about as

27:57

effective as throwing a bike at a

27:58

battleship and here's where lethality

28:00

ties into the purpose that drones can

28:02

actually fill with just about any weapon

28:04

system there are going to be certain

28:05

targets it's designed to service and

28:07

others that it doesn't really provide an

28:09

effective answer for a machine gun isn't

28:11

an anti-tank weapon and a javelin

28:13

missile is going to struggle against

28:14

infantry in the open so an evolution in

28:16

drone payloads isn't just about making

28:18

them more effective against certain

28:19

sorts of targets it's about broadening

28:21

the sort of missions they can do and

28:23

targets they can service at 1 kg of

28:25

payload you're probably talking about

28:27

infantry or vehicles with the hatches

28:28

left open 2 and 1/2 kilos plus and you

28:31

start to have more anti- armor options

28:32

available to you and once you start

28:34

getting into serious double- digit

28:35

figures your drone might be able to do

28:37

its bit for the global emissions

28:38

reduction Movement by proactively

28:40

imposing a output restriction or two on

28:43

opposing oil refineries in 2022 while

28:46

there were some units operating

28:47

specialized heavier drop drones even at

28:49

that time most of the drop drones we saw

28:52

operating and the first generations of

28:54

fpvs that started to come through were

28:56

mostly using very light payload there

28:58

were a lot of systems I saw at that time

29:00

that were often using effectors in the 1

29:02

kg or under range perhaps the

29:04

quintessential examples here were either

29:05

literal hand grenades or adaptations of

29:07

the famous vog the vog often

29:10

specifically the vog 17 version is a

29:12

lightweight 30 mm high explosive

29:14

projectile primarily intended for the

29:16

AGS series automatic grenade launchers a

29:19

wooden case of these things from some

29:20

manufacturers would contain about 108

29:22

grenades and weigh 55 kilos with a lot

29:24

of that being packaging the actual

29:26

rounds themselves from memory are about

29:28

350 G so about 3/4 of a pound and

29:31

particularly when you're only dropping

29:32

them in singles you're only really going

29:34

to do damage to light targets and even

29:36

then only with a direct or very close

29:39

hit but everyone us them because with a

29:41

3D printed Tail kit they could be

29:43

dropped relatively accurately and

29:45

because the Soviet Union was kind enough

29:47

to leave an absolute [ __ ] ton of these

29:49

things behind American 40 mm grenades

29:51

would be another more lethal option but

29:53

we're only available in much smaller

29:55

numbers in 2022 through 23 the path of

29:58

drone Evolution created a number of fpv

30:00

designs that were capable of carrying

30:01

heavier payloads some of the most common

30:04

payloads here for obvious reasons were

30:06

rpg7 Warheads things like the pg7 VL so

30:09

now you have a munition that is yes

30:11

heavier and requires a chunkier drone to

30:13

carry but is a designed to act as an

30:16

anti- armor weapon meaning you probably

30:18

can graduate from killing golf carts to

30:20

bmps and of course like the vog 30 mm

30:24

the Soviet Union built a [ __ ] ton of

30:25

these things and there's plenty of them

30:27

available that availability was

30:29

important not just in a quantity sense

30:30

it was also important in terms of making

30:32

sure that drone operators could actually

30:34

get access to the Munitions this was

30:36

stuff that might already be being pushed

30:37

down the logistics train anyway there

30:39

are a lot of units with RPG SS out there

30:42

and so you just have to nicely ask the

30:43

supply officer to let the Drone units

30:45

borrow a few boxes a heavier option

30:48

again that we saw for the chunkiest of

30:49

fpvs and also some loitering Munitions

30:52

were tandem charges this would be

30:54

something a bit heavier again like the

30:55

pg7 VR which basically gives you two

30:58

charges and one of the main useful

31:00

benefits of that is making the Warhead

31:02

much more effective against explosive

31:04

reactive armor which obviously has

31:06

utility in Ukraine where 90% of every

31:08

vehicle upgrade package seems to consist

31:10

of covering an object in as many ER

31:12

blocks as will fit so that's already a

31:14

scary enough Evolution but obviously

31:16

designers don't stand still and so I'd

31:18

argue we've seen at least three parallel

31:20

tracks of continued development when it

31:21

comes to increasing The lethality of

31:23

combat drones in Ukraine the first is

31:25

just continued experimentation with

31:27

larger and larger payloads because one

31:29

of the simplest expedients for

31:30

increasing the damage something does is

31:32

either making it make a bigger boom or

31:34

having it make more booms so one of the

31:37

responses might be to come up with

31:38

somewhat chunkier drones that can handle

31:40

the additional payload and some designs

31:42

capable of doing so already existed

31:44

before the full scale Invasion that

31:46

image you see there on the right for

31:48

example is a shot that was put up by the

31:49

Russians of a captured Ukrainian drone

31:51

allegedly there it looks like the

31:53

payload is actually three p tabs which

31:55

is a roughly 2 kg anti-tank bomb

31:58

originally designed during the second

31:59

world war that implies a payload of at

32:01

least 6 kg and we're pretty confident

32:04

there are designs out there that can do

32:05

considerably more and this movement

32:07

towards higher payload capacity isn't

32:09

just something we've seen with drop

32:10

drones or fpvs we've seen it in other

32:13

systems categories as well the Russians

32:15

have increased the Warhead size on the

32:16

average Lance that you see used in

32:18

Ukraine by several times over since 2022

32:21

and if you look at some of the longrange

32:22

strike drones Ukraine is using again

32:25

heavier Warheads have become

32:26

increasingly available and recently it's

32:28

been reported the Russians may even be

32:30

following this same evolutionary path

32:32

with some of their cruise missiles when

32:34

the ukrainians recently broke open a

32:35

crashed Russian x101 cruise missile NATO

32:38

reporting name as15 they found that it

32:40

looked like the Russians had pulled out

32:42

some of the fuel tank Edge to replace it

32:44

with a second Warhead adding roughly

32:46

another 400 kg of payload to the thing

32:49

while sacrificing range that

32:51

realistically the system didn't need to

32:53

hit Targets in Ukraine the logic in all

32:56

of these cases is at least somewhat

32:57

similar if you're going to invest a lot

32:59

of money and resources into either

33:01

smashing an object into something or

33:03

using it to bomb something you want it

33:05

to be effective when it does so and all

33:07

else being equal bigger boom often

33:09

equals bigger effect of course for any

33:12

given Target type there is an optimate

33:14

amount of boom and it is possible to go

33:15

too far recently Russian TV showed

33:18

Sergey Sher touring an arms plant during

33:20

which one of the weapons showed off was

33:22

the Fab 3000 3 ton aerial bomb this

33:26

fragile masculinity special is obviously

33:28

a very dangerous weapon but there are

33:30

serious questions over whether or not it

33:32

would actually be practical or efficient

33:34

its range as a Glide bomb is

33:35

questionable the ability to mount this

33:37

thing under anything less than a

33:39

dedicated bomber aircraft likewise

33:41

questionable and it's likely to run face

33:43

first into that basic physics problem

33:45

wherein doubling the amount of explosive

33:46

yield does not in fact double the blast

33:48

radius so the trend towards increasing

33:51

payloads probably has a limit and the

33:52

dedication of resources to extremely

33:55

large systems like Fab 3000 is probably

33:57

questionable although to be fair

33:59

probably not historically surprising

34:01

given that after America spent a portion

34:03

of its very large military budget to

34:05

development field the 9.8 ton GBU 43

34:09

Moab sometimes referred to as the mother

34:11

of all bombs Russia dedicated a portion

34:13

of its much smaller military budget to

34:15

building an even larger one which it

34:17

dubbed the father of all bombs a totally

34:20

efficient and not at all wasteful

34:22

exercise the second big trend is not so

34:24

much to increase the size of the

34:25

payloads but rather to change the nature

34:28

of the payload this potentially offers

34:30

the option to take a drone and optimize

34:32

it better for certain Target types

34:34

without increasing its size or cost if

34:36

you swap out the anti-tank RPG Warhead

34:38

on an fpv with a thermobaric payload for

34:41

example that might be much more

34:43

effective against Targets in buildings

34:44

or enclosed spaces without any increase

34:47

in weight air bursting and enhanced

34:49

fragmentation projectiles might be more

34:51

effective against infantry Targets in

34:52

the open and of course there are a

34:54

variety of specialized anti- armor

34:56

Munitions available there's always been

34:58

a degree of this sort of munition to

35:00

Target matching going on and there are a

35:02

bunch of off-the-shelf Munitions that

35:03

you can potentially bolt to a drone to

35:05

get alternative effects but as the war

35:07

has gone on we've seen more Focus placed

35:09

on dedicated Munitions for drone

35:11

operations this is the process where we

35:13

see more and more Munitions that have

35:15

either been substantially modified for

35:17

drone use or designed and built from

35:19

scratch for drone use although in some

35:21

cases even those scratch built in

35:23

inverted commas Munitions will include

35:25

some recycled or repurposed components

35:28

in 2023 Ukraine announced that they were

35:30

creating a new category of ammunition

35:32

for drone operations which sort of puts

35:34

the organizational Scaffolding in place

35:36

for getting better at designing and

35:38

procuring drone Munitions and allowing

35:40

drone units to order and be supplied

35:42

with them military supply systems are

35:44

often like engaging with customer

35:46

support which consists of nothing other

35:48

than the automated options you can press

35:50

one for vogs two for RPG 7s three for

35:53

152 mm shells or press four to hear

35:56

options 1 through 3 again now it might

35:59

be more possible to streamline that

36:00

process from an organizational

36:02

perspective but we'll see how things

36:04

actually play out in 2024 the move over

36:06

to purpose-built or more significantly

36:08

modified payloads has a number of

36:10

potential benefits for drone operations

36:13

and it must be said potentially a lot of

36:14

drawbacks for those being targeted by

36:16

them purpose-built Munitions might be

36:18

more stable and predictable if they're

36:20

being used for a drop drone for example

36:22

they might improve safety and handling

36:24

so you don't have troops having to cut

36:26

open cluster Munitions in order to ad

36:27

hoc Munitions for their drones it's

36:29

worth noting that converting regular

36:31

Munitions into drone carry Munitions can

36:33

be a dangerous and difficult process

36:36

sometimes people are injured or killed

36:38

sometimes Munitions fail on impact and

36:40

so the extra safety and reliability you

36:42

might get with a purpose-built option

36:43

really does mean something and for a lot

36:46

of systems you might just see increased

36:47

yields for no increase in weight to

36:50

illustrate that imagine an early ad hoc

36:52

strike drone concept where the way you

36:54

gave the thing its boom was by

36:56

integrating an arery shell you have lots

36:59

of shells so you just create an empty

37:00

space plop in the Shell fuse it

37:02

appropriately and off you go the issue

37:04

however is that most of the weight of an

37:06

artillery projectile isn't the explosive

37:08

charge a lot of the mass in your average

37:11

152 or 155 mm projectile is just metal

37:15

the structure of the projectile that is

37:17

designed to survive the very violent

37:19

experience of being fired out of a metal

37:21

tube at very high velocity which as a

37:24

propulsion method for a warhead is much

37:26

much more traumatic than being gently

37:28

carried aoft and then to target by a

37:30

drone so a purpose-built warhead for a

37:32

longrange strike drone might have a lot

37:34

less casing and a lot more actual charge

37:37

there also just might be a push because

37:39

of the sheer scale of the number of

37:41

Munitions being demanded the Soviet

37:43

Union and other producers did leave

37:45

behind a truly mindboggling amount of

37:47

Munitions but as hard as burning through

37:49

the old Soviet stocks is in Ukraine

37:52

they're certainly giving it a red hot go

37:54

so you have producers and organizations

37:55

springing up to try and do at drone

37:58

munition production at larger scale

38:00

recently for example there was an

38:01

article covering the steel Hornets which

38:03

is an organization in Ukraine which is

38:05

described is functioning like a sort of

38:06

Amazon for drone Munitions units can

38:09

reportedly Place orders for all sorts of

38:11

specialized drone payloads including

38:13

some very nasty looking anti-personnel

38:14

options and those are then after some

38:17

obvious Security checks boxed up and

38:19

shipped out so to summarize if the old

38:22

Arch type for drone Munitions was a

38:24

small existing general purpose munition

38:27

the new overly generalized archetype

38:29

would be payloads that are larger more

38:31

specialized and purpose-built and as for

38:34

why that matters hopefully it's obvious

38:35

for both Ukraine and The Wider World in

38:38

2022 23 and early 24 we have already

38:41

seen drones be terrifyingly effective

38:43

drones and loitering Munitions account

38:45

for a significant proportion of visually

38:46

confirmed artillery losses and have

38:48

inflicted a lot of visually confirmed

38:50

damage against a lot of Target types but

38:53

a lot of that observational Baseline

38:54

we're getting was probably established

38:56

using munitions that were not Optimum

38:58

for purpose I think it's logical to

39:00

assume that Russian Lancer teams would

39:02

have inflicted greater losses on Ukraine

39:04

for example if they had started in

39:06

February 2022 universally equipped with

39:08

the larger Warhead version you can

39:10

filter through huge volumes of Russian

39:12

and Ukrainian video of drone attacks

39:14

where with a different payload the

39:16

result might have been different I think

39:19

it's also safe to assume that when other

39:20

major militaries start equipping with

39:22

these sorts of systems in Earnest Uncle

39:24

Sam's not going to be buying a version

39:26

where someone's duct AP a $15 grenade to

39:28

the bottom of a $500 drone no your

39:31

future hypothetical drone optimized

39:33

payload effector Army future or dope AF

39:37

might add a zero or two to the price of

39:39

your average Ukrainian option but we

39:41

probably have to model in the fact that

39:42

they might be much more lethal for it as

39:44

drones have evolved they've flown

39:46

further found their targets more

39:48

effectively and done more damage when

39:49

they get there which is all of course

39:51

horrifically terrifying and logically

39:53

brings us to the question of how to stop

39:55

them and in Ukraine just as the uas have

39:58

continued to evolve so too are the

40:00

counter measures being relied on to

40:01

defeat them during the years of fighting

40:03

in the dbass pre full scale Invasion and

40:06

during the early stages of the fullscale

40:08

invasion you saw a number of reports of

40:10

drone operators claiming to suffer

40:12

because of countermeasure options that

40:13

were provided by the Drone manufacturers

40:16

themselves you can think of this

40:17

essentially as a software-based

40:18

countermeasure system where a

40:20

manufacturer has decided they don't want

40:22

their drones to be used for Warfare or

40:24

other such purposes might for example

40:26

offer a system like the old DJI

40:28

aeroscope which was marketed to law

40:30

enforcement and enabled an operator to

40:32

see both nearby DJI drones and also the

40:34

location of their operators the

40:36

Ukrainian vice prime minister would then

40:38

complain about the Russians using the

40:39

aeroscope system to hunt down Ukrainian

40:42

operators DJI would then later

40:44

discontinue aeroscope but that's

40:46

certainly not the end for manufacturer

40:48

built-in countermeasures in the US for

40:50

example the FAA has steadily moved

40:52

towards a world where the manufacturers

40:54

of many drones are required to build in

40:56

a remote identification broadcast system

40:58

relying on the software and Hardware of

41:00

the Drone itself to provide one of the

41:02

counter measures against potential

41:04

misuse however in 2024 you won't really

41:07

hear anything about any of these systems

41:09

or options in Ukraine based on the

41:12

evidence we have they don't really seem

41:13

to be a problem for drone operations now

41:15

although for hopefully obvious reasons I

41:17

won't go into any detail about how

41:19

that's been done you could probably

41:21

argue that this sort of evolution

41:23

provides a little bit of a warning to

41:24

governments and security agencies that

41:26

are relying on this sort of approach to

41:28

answer the potential security threats

41:30

drones might pose domestically and it

41:32

helps highlight the difficulty the

41:34

manufacturer of something like a drone

41:35

might face controlling their product

41:37

after it's sold as opposed to a service

41:40

like starlink with a system like

41:41

starlink the customer buys a terminal

41:44

but the terminal can't magically provide

41:46

internet it can only do that by

41:47

communicating with satellites and the

41:49

satellites are still controlled by

41:51

starlink and what that means is that if

41:53

Elon Musk and SpaceX decide to shut down

41:55

starlink service in a particular area

41:57

area you probably can't get around that

41:59

problem by messing with your terminal

42:00

you'd have to mess with the satellites

42:02

and those are significantly harder to

42:04

physically reach and tamper with a drone

42:07

by contrast is going to be intrinsically

42:09

mechanically functional even if you

42:11

write a line in its codes saying it

42:12

shouldn't be operable in particular

42:14

areas say around sensitive military and

42:16

civilian sites the Drone is physically

42:18

capable of getting there it's just a

42:20

line of software telling it not to and

42:22

if hypothetically for some reason that

42:24

block of code was ever to change or

42:26

disappear then the system would likely

42:28

be physically capable of doing a couple

42:30

of loopy Loops over places it really

42:32

shouldn't be that's a concern with

42:34

potentially Global implications but for

42:36

now let's focus on the countermeasure

42:38

picture in Ukraine in the absence of

42:40

manufacturer provided options the most

42:42

common answer to drones in Ukraine has

42:44

been electronic warfare you could say

42:46

that over the last 2 years the frequency

42:48

of electronic warfare in Ukraine has

42:50

constantly changed but I'm not sure we

42:52

could handle a dad joke that bad Jammers

42:54

have proliferated to the point where

42:55

we've seen them strung along fly lines

42:57

mounted on armored vehicles or even kept

42:59

in trench systems at the extreme end

43:02

recently we even got images of this

43:03

monster here where reportedly what the

43:06

Russians did was put a pallet on top of

43:08

a tank and strapped that pallet a

43:10

variety of Jammers and electronic

43:12

warfare equipment battery and Diesel

43:14

Generator while the reporting is purely

43:16

anecdotal it suggested the thing did

43:18

manage to bring down more than a few

43:19

fpvs but in the end as the image

43:22

suggests the boom box Jammer approach

43:24

ultimately wasn't enough to save the

43:26

vehicle and and while the balance

43:27

especially at the local level can swing

43:29

continuously between the Jammer and the

43:31

jam we have seen a variety of different

43:34

approaches used by drone and loitering

43:35

munition manufacturers to try and get

43:37

around at least some of the ew problem

43:40

one very strange solution we've seen is

43:42

actually a very low tech one this was a

43:44

Russian drone which was reportedly

43:46

captured which seems to have followed

43:47

the basic line of thought that if ew

43:49

systems inhibit your wirer signals how

43:52

about you just don't use wireless

43:53

signals Instead This drone had a reel of

43:56

fiber optic cable connected to it and

43:58

seems to have been intended to just

44:00

literally spool out a line as it flew in

44:02

order to maintain a connection to the

44:04

operator well that may sound promising

44:06

because no one's found a way to jam a

44:08

wire yet short of you know cutting it

44:10

and wire guided weapons very much are a

44:12

reality with some other systems like

44:14

atgms and Torpedoes there are some

44:16

pretty obvious drawbacks to try to

44:18

operate fpvs this way we've only seen

44:20

one example so far and it failed so we

44:23

may not see other producers raced to

44:24

emulate A system that shares

44:26

characteristics with kite and has a 0%

44:29

success rate instead what we've seen

44:31

more of is a constant shift in the

44:32

frequency that drone operators use

44:34

coupled with methods of guidance and

44:36

targeting that might be more Jam

44:38

resistant frequency selection

44:40

coordination and jamming can be major

44:42

factors in drone operations if you jam a

44:44

frequency your own guys are using you're

44:46

going to bring down your own Birds which

44:48

reportedly has often been a major

44:49

problem in Ukraine but is also one that

44:51

I think other militaries if given the

44:53

time and resources might find some

44:55

answers to you can also have cases where

44:57

systems might be redesigned to operate

44:59

on frequencies that are less frequently

45:00

jammed Ukrainian media for example

45:03

claims that the country has produced a

45:04

small counterd drone system that can

45:06

operate in the 850 to 940 MHz range

45:09

which is reportedly a pretty standard

45:11

range for a lot of fpvs but that same

45:13

article points out that there are

45:15

Russian kamakazi drones that can operate

45:17

on either lower or higher frequencies

45:19

those aren't really going to care about

45:21

this sort of system and only be

45:22

vulnerable to those that can hit the

45:24

wider range meanwhile if you want to

45:26

look at longer range oneway attack

45:27

systems that don't require constant user

45:29

control but instead use GPS so a global

45:32

navigation system of some kind we've

45:34

still seen a number of approaches being

45:36

used to protect those systems from just

45:38

being neutralized by jamming the Iranian

45:40

Shahed 136 for example presumably to

45:43

decrease its vulnerability to things

45:45

like GPS spoofing doesn't have one gns

45:48

receiver it has multiples and so if two

45:50

receivers say that it's approaching its

45:52

Target while one is convinced it's 100 m

45:54

off course the majority is going to

45:55

shout down the problem China fa then try

45:57

and continue to Target if gns is denied

46:00

entirely most longrange attack systems

46:02

include inertial navigation backup which

46:04

is usually going to be much less

46:05

accurate than gns but will function even

46:08

if the system is being jammed we're also

46:10

seeing some reporting now that things

46:11

like long range oneway attack drones

46:13

might be using other classical methods

46:15

of navigation these include terrain

46:17

recognition based approaches where

46:19

you're not reliant on receiving a

46:20

navigation signal instead you're using

46:22

your onboard sensors to look at the

46:24

terrain that you're flying over

46:26

comparing that to the database and

46:27

Mission plan that's been pre-loaded into

46:29

the system and using that to find out

46:31

where you currently are tomahawk

46:33

missiles from the Cold War era had this

46:35

sort of technology and obviously even

46:37

then you didn't need a GPS connection to

46:39

recognize a river a crossroad or a

46:41

mountain range while there are a lot of

46:43

old approaches to navigating in a

46:45

spectrum denied environment however the

46:47

thing we're seeing change now is the

46:49

degree to which new technology makes

46:50

these approaches cheaper and more

46:52

accessible the availability of

46:54

commercial satellite imagery for example

46:56

makes it much more practical to get a

46:57

picture of the terrain all the way

46:59

between your launch site and the Target

47:01

and memory computing power and good

47:02

cameras are all much cheaper more

47:04

miniaturized and more accessible than

47:05

they were in the 1980s this and other

47:08

alternative navigation options obviously

47:10

aren't perfect but they do point to the

47:12

risk of systems like oneway attack

47:14

drones becoming more and more resistance

47:16

to attempts to counter them using

47:17

electronic warfare that of course does

47:19

still leave the hard kill option of just

47:21

shooting down the incoming drones but

47:23

then you might run into issues like the

47:24

difficulty getting sufficient

47:25

geographical coverage or shot exchange

47:28

problems if you do it using longer range

47:29

missiles so in terms of the big picture

47:31

for the current war in Ukraine trying to

47:33

assess where the status quo is in the

47:35

race between countermeasures and

47:36

countermeasure resistance is harder than

47:38

in some of the other areas we've looked

47:40

at we know that approaches and

47:42

Technologies used are constantly

47:43

shifting and we have some examples of

47:45

new Jammers succeeding or failing but

47:47

getting good at least semi-complete

47:49

information on success rates is very

47:51

difficult and it would be quickly

47:52

outdated even if it was available but

47:54

one Trend that does stand out because it

47:56

might have wider implications is the

47:58

reportedly greater application even in

48:00

the case of relatively cheap one-way

48:02

attack drones of navigation methods that

48:05

might not have an easy electronic

48:06

warfare based answer a Defender isn't

48:08

going to be able to move mountains

48:10

Crossroads and river systems just to fo

48:12

a navigation system and so the trend in

48:14

2024 and Beyond might be towards more

48:16

one-way attack drones with greater

48:18

resistance to existing ew

48:20

countermeasures and greater pressures

48:21

for countries to reconsider how their

48:23

air defense systems might be configured

48:25

to deal with exhaustion or saturation

48:27

based threats from affordable longrange

48:30

attack options and touching on the

48:32

possibility of saturation or exhaustion

48:34

based tactics brings us to another

48:36

section a key element to what is making

48:38

drones so dangerous in Ukraine which has

48:40

nothing to do with the capability of the

48:42

drones themselves and while this may be

48:44

my bias talking one of those key factors

48:47

is production ultimately sketches CAD

48:49

files and project plans don't destroy

48:51

tanks physical weapons do and so if you

48:54

have a system or technology that's well

48:55

suited for quickly moving from that

48:57

design to production to the turret

48:59

popping phase that's potentially going

49:01

to add to its utility as a weapon system

49:03

when you're a country in the midst of a

49:04

war and from that perspective drones

49:06

have massive advantages and have

49:08

benefited from changing design and

49:09

production processes in 2022 and early

49:12

20123 for example for a lot of drone

49:14

operators on both sides of the war in

49:16

Ukraine the most common approach for

49:18

getting your hands on a lot of small U

49:20

quickly was just to go and find a

49:22

commercially available civilian system

49:24

buy it in significant numbers and if

49:26

necessary Tinker with and modify it as

49:28

you'll see on screen there I've

49:29

tentatively called this the adapted cots

49:31

phase cot stands for commercial offthe

49:34

shelf and this approach was behind the

49:36

massive surge of commercially available

49:38

quad rotor drones in particular that we

49:39

saw move into Ukraine and later Russia

49:42

the production and scaling model is

49:43

really simple because you're not doing

49:45

much production domestically instead you

49:47

have some Avid drone enthusiasts in

49:50

Estonia or Poland hypothetically calling

49:52

up retailers wholesalers or producers

49:54

and saying that their drone Club really

49:56

needs you know a couple of thousand

49:58

extra mavic 3s the bit where the local

50:01

production comes in is in the adaptation

50:03

part and these are changes to the fully

50:05

assembled drone that are necessary to

50:07

make it more suitable for Battlefield

50:09

use in a basic observation role fewer

50:11

modifications are likely to be necessary

50:14

but if you're talking about making the

50:15

Drone suitable as a weapon system in its

50:17

own right more modifications might be

50:19

required for example for some reason the

50:21

DJI mavic 3 doesn't come with a bomb

50:24

carriage and release mechanism as an

50:25

optional extra the advantage of this

50:27

model is you might be able to get a lot

50:29

of drones relatively cheaply have small

50:32

organizations or field workshops make

50:34

the necessary field modifications and be

50:36

relatively confident the Russians aren't

50:38

exactly going to bomb the factories in

50:39

China the disadvantages however are that

50:41

you are reliant on a foreign supply

50:43

chain and also on a civilian design

50:46

civilian quad rotors have been used in

50:48

war but they weren't designed for it and

50:50

so particularly when you're talking

50:51

about fpvs and also to an extent the

50:53

heavier quad and octocopter bomber

50:55

drones we've seen increased Reliance

50:58

placed on a new model this is where we

51:00

are building drones that are purpose

51:01

built for military environment but we're

51:03

doing it mostly using commercial

51:05

off-the-shelf components metaphorically

51:07

we are no longer going to Toys R Us

51:09

buying a complete drone and figuring out

51:10

a way to duct tape and munition to it

51:12

instead you're going and buying critical

51:14

components batteries Motors flight

51:16

controllers and now often in a slightly

51:18

larger Workshop what you are doing is

51:20

assembling those components into a

51:22

design that you have come up with

51:23

yourself potentially with some

51:25

custombuilt local components

51:27

the frame of the Drone for example might

51:29

be locally manufactured to make sure you

51:31

can fit your chosen bit of ordinance to

51:32

it the advantage of this model is you're

51:34

still able to leverage someone else

51:36

often China's productive capacity and

51:38

cost base but now you're not as limited

51:40

by the design of the underlying drone in

51:42

question you can mix and match

51:44

components and Technical features design

51:46

with a purpose in mind and still

51:47

hopefully keep your system pretty cheap

51:49

also because in industrial terms a lot

51:51

of the work here is relatively basic

51:54

it's still something that a relatively

51:55

small team in a volunteer organization

51:57

or small company might be able to do inh

51:59

house if you want an imperfect analogy

52:02

it takes a lot less skill in Capital

52:03

Equipment to modify a car than it does

52:05

to build one from scratch down to its

52:08

base components the next potential

52:10

evolutionary step that we've seen some

52:12

of an Ukraine and almost inevitably will

52:14

see basically everywhere on the global

52:16

market is various drone manufacturers

52:18

moving from just assembling various

52:20

commercial off the-shelf components to

52:22

designing and arranging the production

52:23

of their own in some cases this might be

52:26

just out of a desire to vertically

52:27

integrate remove Reliance on Supply

52:29

chains you don't control and potentially

52:31

bring down cost but in other cases it's

52:33

because the underlying civilian product

52:35

may not align with what you want the

52:37

military system to be able to do to give

52:39

one example a lot of the commercially

52:41

available Motors you might use to power

52:43

an fpv or a larger loitering munition

52:45

have a significant problem they're too

52:47

good a drone hobbyist probably doesn't

52:49

want their new toy to give up the ghost

52:51

after 10 hours of service and if you're

52:53

a military or civilian organization

52:54

using a larger ISR drone you might want

52:56

to get thousands upon thousands of

52:58

flight hours out of that thing but if

53:00

you're talking about a one-way attack

53:01

drone that thing has a Destiny the

53:03

moment it takes off and the destiny in

53:05

question is very likely to either be

53:07

detonation the dirt or both so in that

53:10

context it makes no bloody sense to fit

53:13

the thing with motors that are rated for

53:14

hundreds or thousands of hours instead

53:17

what you care about is short run

53:18

performance and price there are plenty

53:20

of other components that also benefit

53:22

from being custom built for a military

53:24

context but hopefully you get the

53:25

picture there is is a difference in some

53:27

cases and the introduction of these more

53:29

customized systems alongside continuing

53:31

purchases it must be said of those

53:33

commercial off-the-shelf drones from

53:34

before is part of the reason why we've

53:36

been able to see such a growth in scale

53:38

on one hand but also the development of

53:40

new capabilities on the other reflecting

53:42

that it has to be said that the

53:44

production methods used by Russia and

53:45

Ukraine for small uas have changed over

53:48

time in Russia a lot of the effort has

53:50

been in true command economy style a top

53:52

down effort which has the advantage of

53:54

being able to throw a lot of resources

53:56

at at a Consolidated number of systems

53:58

and designs and get them produced in

53:59

quantity Russia for example has sunk a

54:02

lot of effort into scaling Shahed 136

54:04

production lanet production and some

54:07

lighter systems as well the weakness of

54:09

that system that's sometimes complained

54:11

about in Russian language sources is it

54:13

can be a little bit slow to adapt to

54:14

changing Battlefield demands which is a

54:17

situation where you can see volunteer

54:19

organizations small workshops and field

54:21

modifications still really proving their

54:23

worth the Ukrainian model by contrast is

54:26

more fragmented orbe it with some

54:28

unifying efforts you have a mixture of

54:30

all sorts of startups foreign suppliers

54:32

volunteer organizations small field

54:34

workshops right down to individuals who

54:37

just happen to own a 3D printer and

54:39

produce drone accessories Ukraine to an

54:41

extent has been able to standardize

54:43

things like pricing for systems like

54:44

fpvs but they haven't yet chosen to

54:47

standardize around particular producers

54:49

or models instead you have a lot of

54:51

producers making a lot of designs in a

54:53

fairly artisanal fashion and as some

54:55

drop out a business or can't keep up

54:57

they and all their workforces are often

54:59

Consolidated into those firms that

55:01

remain as going concerns in raw economic

55:03

terms this model has the potential to be

55:05

less efficient than a Russian one where

55:07

a lot of physical capital is thrown at a

55:09

particular problem but it does allow for

55:11

some pretty freewheeling Innovation and

55:14

provides a sort of natural push to a

55:15

more distributed production Network that

55:18

all else being equal is probably going

55:19

to be more resistant to missile attack

55:21

than concentrated production in one or

55:23

two megafactories which is of course a

55:25

relevant factor for you Ukraine and

55:27

maybe even for Russia because ultimately

55:29

this is 2024 and you never know when

55:31

someone's going to slam a remote

55:32

controlled Cessna strapped with a bunch

55:34

of explosives into your production

55:35

facility but one way or the other in

55:37

both Russia and Ukraine a lot of

55:39

resources have been invested in ramping

55:40

up drone production and the results has

55:42

been a massive increase in the scale of

55:44

just how many of these systems both

55:46

sides have available in May 2023 Russy

55:49

estimated that Ukraine was expending

55:50

about 10,000 uavs per month that was

55:53

already a very high number and probably

55:55

meant that even at that Point small

55:56

drones were the most commonly used

55:58

Precision weapon system on the Ukrainian

56:00

Battlefield by the end of 2023 however I

56:03

was suggesting that number was probably

56:05

outdated and potentially several times

56:07

too low and now in 2024 we've had

56:10

multiple reports from Ukraine the

56:11

country is aiming to manufacture at

56:13

least a million drones domestically this

56:15

year is on track to do so and aims to

56:18

import potentially a million more

56:20

obviously the starting point for any

56:22

claim like this is to take it with a

56:23

grain of salt but it may give a good

56:25

indication of the sort of numbers that

56:27

are now being aimed for even taking a

56:29

lower end of those two combined figures

56:31

at 1.2 million per year that's not

56:33

10,000 drones a month it's 100,000

56:36

essentially marking the transition from

56:38

Ukraine using a metric [ __ ] ton of

56:40

drones to a metric Decat ton only a year

56:43

later and while a lot of that volume and

56:45

the Russian equivalent is likely to be

56:47

in the smallest and cheapest systems the

56:48

fpvs some of the stated goals we've seen

56:51

for long-range strike systems are also

56:53

incredibly ambitious just one

56:55

manufacturer of of one Ukrainian

56:57

longrange strike drone for example has

56:59

publicly said that they can go from

57:00

producing 100 a month to 4 or 500 a

57:03

month in relatively short order provided

57:05

one assumes contracts and resources are

57:07

forthcoming and amidst all these figures

57:10

I think there might be a key takeaway

57:11

for Ukraine and to a lesser extent also

57:14

for Russia the availability of systems

57:16

and Munitions is a challenge air defense

57:18

systems like Patriot and the missiles to

57:20

fire from it are in very short supply as

57:23

is artillery ammunition artillery

57:25

barrels armored vehicles and just about

57:26

anything else you care to name small

57:28

drones are about the only thing that

57:30

both Ukraine and Russia have a lot more

57:32

of now than they did a year ago and

57:34

against that backdrop against that

57:35

context of a shortage in a range of

57:37

other systems and Munitions you can see

57:40

why there might be more and more

57:41

Reliance placed on drones to fill

57:42

certain Battlefield roles in the end you

57:45

fight with the systems you have which

57:47

brings us to the end of the discussion

57:49

on the technical changes and gets us to

57:51

the final question based on all of these

57:53

changes we've seen and expect to

57:55

continue seeing how impactful are drones

57:57

in Ukraine today and how influential are

57:59

they likely to be going forward if you

58:01

just track through a lot of the videos

58:03

interviews and reports that come out of

58:04

Ukraine it's probably fairly easy to

58:06

argue that drones have helped make the

58:08

battlefield incredibly transparent and

58:10

extremely dangerous moving or

58:12

concentrating forces is going to be

58:13

difficult when both sides have so many

58:15

eyes in the sky the already dangerous

58:17

artillery systems on both sides have

58:19

probably benefited from having drones to

58:21

spot targets and correct fire and the

58:23

threat posed by things like fpvs and

58:25

drop drones is such that if you are in a

58:27

defensive position that doesn't have

58:28

adequate overhead cover or concealment

58:30

nor adequate ew cover somewhere where

58:32

your opponent has skilled drone

58:34

operators present then the threat is

58:36

probably constant and growing but at the

58:38

same time there are still arguments

58:39

playing out over just how impactful

58:41

drones have actually been recently for

58:43

example there was an article in the

58:44

conversation by Paul lenko an assistant

58:46

professor and director of Special

58:48

Operations at the US Army war college

58:50

and he essentially argues among other

58:52

things that while drones have delivered

58:53

some tactical and operational success

58:55

for both Ukraine and Russia

58:56

they've been strategically ineffective

58:58

to quote directly from the article

59:00

drones have not and are not likely to

59:02

shape the outcome of the war in Ukraine

59:04

they have not allowed Ukraine to break

59:05

its stalemate with Russia nor have they

59:07

encouraged Russia to end its occupation

59:09

of Ukraine he also says quote the lesson

59:12

from Ukraine is that while drones have

59:13

some value at the Tactical and

59:14

operational levels of War they are

59:16

strategically inconsequential they are

59:18

not a Magic Bullet offering a

59:20

gamechanging capability to decide the

59:21

fate of Nations instead countries must

59:24

rely on time tested combined arms

59:26

maneuver wherein they integrate

59:27

personnel and weapon systems at a

59:28

particular time and place to achieve a

59:30

particular goal against an adversaries

59:32

end quote one wonders it that means that

59:34

almost no weapon system can be

59:36

considered strategically consequential

59:38

because they have to be used together in

59:39

Combined arms for maximum effect and

59:41

even an extreme case like a nuclear

59:43

warhead realistically requires a

59:45

delivery system it also I think arguably

59:47

undere explores the potential impact of

59:49

drones both on how combined arms

59:51

operations are conducted and also how

59:53

viable they are the article acknowledges

59:55

for example the role of Ukrainian drone

59:57

operators in stopping the Russian

59:58

Advance on keev saying that Ukrainian

60:00

air reconnaissance units quote use

60:02

drones to interdict and block a massive

60:04

Russian Convoy traveling from Chernobyl

60:06

to keev a month after Russia's February

60:08

24 2022 invasion of Ukraine it did so by

60:11

destroying slow moving vehicles that

60:13

stretch nearly 50 mil causing Russia to

60:15

abandon its Advance end quote I don't

60:18

think drones deserve sole credit for

60:20

that but let's go with it for a moment I

60:22

think you could argue that halting

60:23

Russian columns moving towards ke were

60:26

exactly the kind of tactical actions

60:28

that could be aggregated together to

60:29

have operational and strategic

60:31

significance enough Russian tactical

60:33

successes that enabled an operation to

60:35

take keave successfully May potentially

60:38

have ended up having all sorts of

60:39

strategic significance taking a capital

60:42

is no small thing and it's difficult to

60:43

assess what the full ramifications that

60:45

would have been for the stability of the

60:47

Ukrainian government foreign support or

60:49

a range of other macro factors as I said

60:51

I'll link the article in the description

60:53

and it's probably not the only one of

60:54

its kind but for what it's worth from my

60:56

perspective when you have an argument

60:58

that a weapon system isn't strategically

61:00

decisive because the conflict it's being

61:02

used in is a balanced one without a

61:04

decisive result I think that's falling

61:05

into the Trap of looking at the outcome

61:07

of a conflict as opposed to the impact

61:09

of a system which among other things

61:11

ignores the fact that a stalemate might

61:13

in fact be a strategically significant

61:15

result if one side would have lost a

61:17

conflict but then because it deploys a

61:19

new system is able to score a draw I'd

61:21

argue that's still a pretty significant

61:23

impact and if both sides deploy a new

61:25

techn techology keeping them in a sort

61:27

of balance I don't think that takes away

61:29

from the impact and value that system

61:31

has this is one reason why I tentatively

61:33

suggest if you're trying to determine

61:35

just how impactful a new system or

61:36

technology is the best test might not be

61:39

to look at how the conflicts that system

61:40

was used and played out but instead what

61:43

would have happened if the decision to

61:45

invest in that technology or system

61:46

wasn't made by one or both sides for

61:49

example if you take a passage from the

61:50

article and swap around a few words you

61:52

could end up with something that could

61:53

have been written in 1916 the despite

61:56

these tactical effects and limited

61:57

operational gains artillery is

61:59

strategically ineffective artillery has

62:01

not and is not likely to shape the

62:03

outcome of the war in Europe they have

62:04

not allowed the onon to break its

62:06

stalemate with Germany nor has it

62:07

encouraged Germany to end its occupation

62:09

of Belgium to the extent artillery

62:11

batteries have been strategically

62:13

consequential the implications have been

62:15

psychological and you could make that

62:17

argument at the time because the Western

62:19

Front despite enormous quantities of

62:21

artillery being used was a stalemate

62:23

which would have ignored the fact that

62:24

artillery was part of the stalemate and

62:27

that if either the onon or Central

62:28

Powers had entirely failed to invest in

62:30

their artillery then I'd argue the war

62:32

probably would have swung massively and

62:34

potentially strategically decisively

62:36

against them so I'd suggest we then look

62:38

at drone deployment through that same

62:39

sort of butt for lens what would happen

62:42

if the investments in drone systems

62:43

hadn't been made if not for the

62:45

investments in drones the battlefield in

62:47

Ukraine would be so much less

62:48

transparent it would presumably be

62:50

easier to move and concentrate forces

62:52

and supplies spotting targets for

62:54

indirect fires would be more difficult

62:56

as would be correcting those fires and

62:58

in a situation like that where it might

62:59

be safer to concentrate forces and

63:01

defending fires might be less responsive

63:03

especially against Dynamic targets it

63:06

might be that the balance between

63:07

attacker and defender in that scenario

63:09

would look a little bit different from

63:10

the war in Ukraine that we see today but

63:12

for the range and accessible Precision

63:14

that drones and Laing Munitions provide

63:16

lot more of the Ukrainian battle space

63:18

would be under considerably less threat

63:19

there would be a lot of things out there

63:21

destroyed by systems like Lancet or

63:23

spotted by drones and then engaged by

63:24

longer range fire options that might

63:26

have survived artillery tactics probably

63:28

wouldn't have to be optimized against a

63:30

threat that didn't exist and the forces

63:32

might have been able to skip certain

63:33

tactical and operational adaptations

63:35

that do come with a cost in that sense

63:37

you might see parallels to the arguments

63:39

around whether or not high Mars was

63:40

actually effective You could argue High

63:42

Mars and its gimler missiles weren't

63:44

particularly decisive because Russian

63:45

forces were able to disperse their

63:47

ammunition storages or pull back their

63:48

Depots but reframe that you could argue

63:51

that gimler was impactful because it

63:53

forced Russia to disperse its storages

63:55

and P back its Depots the costs imposed

63:58

on an enemy can be holistic they don't

64:00

have to take the form of a burning piece

64:01

of equipment and to that I'd add if

64:03

we're trying to predict potential impact

64:04

on future conflicts we need to keep the

64:06

scale in mind here when high Mars first

64:08

reached Ukraine it was a handful of

64:10

launchers some NATO countries are

64:12

looking at building up arsenals of

64:13

hundreds when you're talking about drone

64:15

systems in Ukraine a lot of these

64:17

Technologies were relatively embryonic

64:18

in 2022 and the numbers we're seeing now

64:21

are still in many cases driven by

64:23

relatively basic production chains so so

64:25

when you are trying to assess the

64:27

potential impact on future battlefields

64:28

it might be worth asking not just how

64:30

many drones can Ukraine produce in 2023

64:33

or 2024 but rather how many could

64:36

countries like China the United States

64:38

or the various EU members produce if

64:40

they really focused on it but even in

64:42

the limited numbers available to Russia

64:43

and Ukraine I'd argue that things like

64:45

one-way attack drones have had

64:47

operational and strategic impacts and

64:49

not merely psychological ones if not for

64:51

systems like Shahed 136 and Ukraine's

64:54

various longrange attack drones neither

64:56

side would need to dedicate Manpower

64:58

equipment and Munitions to forming

65:00

things like drone Hunter and drone

65:01

defense teams behind the lines to defend

65:03

critical infrastructure government

65:05

facilities Depots air bases ports

65:07

industrial facilities the list goes on

65:10

those resources could instead be

65:11

dedicated to getting more effect at the

65:13

front line if not Fisher head 136

65:15

Ukraine's arsenal of air defense

65:17

interceptors would not be under the

65:19

stress it is and it might be safe to

65:21

assume that more launchers and Munitions

65:22

could then be dedicated for jobs like

65:24

being pushed close to the front line to

65:26

hunt Russian Glide bombers and as a side

65:28

note while that might not in itself make

65:30

Ukrainian combined arms operations

65:31

viable I would argue it would dim

65:33

further the prospect of Russia ever

65:35

doing the same because Russia's

65:37

prospects of ever achieving the sort of

65:38

air superiority that are usually

65:40

associated with successful combined

65:42

operations or at least successful

65:44

combined arms operations on a larger

65:46

scale is intrinsically tied to the

65:48

capabilities and endurance of the

65:50

Ukrainian air defense system the one

65:52

that shahad 136 is so critical and

65:54

pressuring meanwhile if not for

65:56

longrange Ukrainian attack drones Russia

65:59

would not be having to distribute

66:00

electronic warfare and air defense

66:02

assets against thousands of square kilm

66:04

of Russian territory because even

66:06

relatively small numbers of simple

66:08

attack drones were capable of

66:09

threatening targets like air bases and

66:11

oil refineries with significant damage

66:14

without investment in drones is's also a

66:15

question of what could possibly have

66:17

replaced them because one of the most

66:19

significant aspects of a lot of these

66:21

systems particularly things like fpvs is

66:23

just how simple they are to access

66:26

manufacture and deploy It's relatively

66:28

simple to say that both Ukraine and

66:30

Russia might benefit from having all

66:31

sorts of additional capabilities that

66:33

help enable combined arms maneuver the

66:35

Russian Air Force could have been much

66:36

better equipped and trained for

66:37

suppressing or destroying enemy air

66:39

defenses Ukraine might really want

66:41

additional air power of its own things

66:43

like F-16 or gun either in a ground

66:45

attack role or to contest the Russian

66:47

Air Force but as we have seen even with

66:49

a lot of resources pledged that can be

66:52

very difficult to do at scale and

66:54

quickly over the time scale it has taken

66:56

Ukraine to go from flying 0 f-16s to 0

66:59

f16s fpvs have gone from being a rare

67:02

novelty to the most common Precision

67:04

weapon on the battlefield and implicit

67:06

in that is that in a lot of times in a

67:08

lot of places where Ukrainian or Russian

67:10

forces are under immense pressure and

67:12

Supply constraints while drones may not

67:14

have been a perfect answer to the

67:16

challenges they faced they were often an

67:18

answer the battlefield drones of the

67:20

future even the fpv analoges may not be

67:23

as cheap as the ones we've seen in

67:24

Ukraine so far there is some Rousy

67:26

research out there for example that

67:27

indicates that in order to have a good

67:29

chance of knocking out an armored

67:30

vehicle an fpv probably needs some

67:32

Advanced features like some of the

67:34

things we've covered here Jam resistant

67:35

Communications the ability to carry a

67:37

higher payload infrared or night vision

67:39

sensors but you're still talking about

67:41

systems that are going to be

67:42

considerably cheaper and more accessible

67:44

than things like atgms or even some of

67:46

the relatively fancy unguided artillery

67:48

shells out there in Ukraine I'd argue

67:51

that's probably mattered if for no other

67:53

reason then it's pretty difficult for

67:54

hobby work shops and volunteer

67:56

organizations to manufacture Javelin

67:58

missiles if you brwn out the arguments

68:00

to include sea drones the arguments

68:01

could probably go further there you

68:03

could argue that for a relatively small

68:05

investment Ukrainian Naval drones have

68:07

been able to fundamentally change how

68:08

the Russian Black Sea Fleet can operate

68:10

and I've explored some of the details of

68:12

that theater and the role Naval drones

68:14

have played in it in a previous episode

68:16

but by now hopefully you get the

68:17

argument yes the war in Ukraine is

68:19

relatively static and attritional no

68:22

drones haven't won the war by themselves

68:24

and combining arms together including

68:26

infantry armor artillery and air defense

68:28

assets just to name a few obviously

68:31

remains absolutely Central but I don't

68:33

think that should concealed just how

68:34

impactful drones have been and the

68:36

disruptive potential they have for both

68:39

sides drones represent a small minority

68:41

of the total resources invested and with

68:43

them both investing there's a sort of

68:45

symmetry although the scale of advantage

68:47

to one side of the other does change

68:49

somewhat over time but if tomorrow you

68:51

Thanos snapped all of Ukraine or all of

68:53

Russia's drones out of existence then

68:55

and while it's obviously difficult to

68:57

prove a hypothetical I think the

68:58

evidence suggests that would have a

69:00

pretty Major Impact on the war for

69:01

hardpressed Ukrainian Defenders it would

69:03

strip them of situational awareness and

69:05

a critical weapon system while for

69:07

Russia it would significantly reduce

69:09

their ability to leverage their

69:10

advantage in longrange Firepower drones

69:13

may not have won the war in Ukraine for

69:14

either side but it's hard to come up

69:16

with many systems that have been as

69:17

costeffective in shaping it and going

69:19

forward in the short and medium term I

69:21

think the technology and its Battlefield

69:23

potential may only grow tanks and arery

69:25

are fairly mature Technologies

69:27

militaries have had a long time to

69:28

figure out how to improve them and use

69:30

them but the technology and tactics of

69:32

these sort of Battlefield drone systems

69:34

is in many ways evolving in real time

69:36

we've seen hints of the potential pre

69:38

2022 but the war in Ukraine has put the

69:40

pace of development into overdrive and

69:42

so however scary and impactful you think

69:44

these systems might be now when it comes

69:46

to their future capabilities we may not

69:48

have seen anything yet and that fact may

69:51

put militaries around the world in a

69:53

difficult position if you stock up on

69:54

systems now they may be quickly

69:56

overtaken by Future countermeasures and

69:58

new more capable systems but if you wait

70:00

too long you run the risk of an opponent

70:02

gaining a potentially significant

70:04

Advantage for a relatively modest

70:05

investment the same sort of timing

70:07

imperatives apply when you're talking

70:09

about the imperative of Defending

70:11

against these systems do you wait for

70:13

example for directed Energy Systems to

70:15

mature or do you go all in on the

70:17

technologies that are available now

70:19

whatever decisions are made however I

70:20

think the reality is this technology

70:22

can't be put back in Pandora's Box the

70:24

lessons being learned can't be unlearned

70:27

and the supply chains and production

70:28

facilities that are springing up to

70:30

sustain the war in Ukraine aren't just

70:32

going to suddenly disappear even if the

70:33

war ends indeed you may even see a

70:36

scenario where countries like Ukraine

70:37

become major suppliers of these systems

70:40

because ultimately they and the Russians

70:42

are probably going to be the most

70:43

experienced experts with some of the

70:45

hottest production lines but for now

70:47

it's probably enough to observe just how

70:49

important these systems have become in

70:51

the war in Ukraine to note just how

70:53

quickly they've evolved and to wonder

70:55

just what sort of changes we're going to

70:56

see in

70:58

2024 and all right brief Channel update

71:00

to close out firstly I know this topic

71:02

want to poll a couple of weeks back but

71:04

thank you for giving me just a few more

71:05

weeks to massage the content a bit one

71:07

of the challenges making a video like

71:09

this is just how expansive the

71:10

underlying topic can be but hopefully

71:13

you're happy with what I ultimately

71:14

chose to cover going forward the patron

71:16

topic poll is starting to wind down so

71:18

I'll start to finalize the results from

71:20

that in the coming week and once that's

71:22

done I'll start releasing some of the

71:23

results in future Channel updates

71:25

finally for those of you who are

71:27

interested in the per and gaming side of

71:28

things I am told the new PC has now been

71:30

shipped so hopefully we see some return

71:33

to that sort of content in the coming

71:34

week or two thank you as always for your

71:36

ongoing engagement interest and support

71:39

and I hope to see you all again next

71:41

week