Robotics Bottleneck Research机器人瓶颈研究

Main主页 / companies / public / doosan-robotics

company

Doosan RoboticsDoosan Robotics 454910

Country国家 Korea韩国

Systems系统 Muscle肌肉 · Integration集成

Bottlenecks瓶颈 P13 Dexterous manipulation灵巧操作 · P17 Integration, uptime, service集成、在线率与服务

Market cap市值 Live, via TradingView.实时数据,来自 TradingView。

Collaborative robot arms and North American end-of-line automation via the 2025 ONExia acquisition. Q2 2026 revenue +290% YoY to KRW 17.67bn; still deeply loss-making, with the stock down roughly 55-60% from its June 2026 peak.

协作机器人手臂,以及通过 2025 年收购 ONExia 获得的北美产线末端(EOL)自动化业务。2026 年二季度收入同比 +290% 至 176.7 亿韩元;仍处于严重亏损状态,股价较 2026 年 6 月高点回落约 55-60%。

What it doesCollaborative robot arms and North American end-of-line automation via the 2025 ONExia acquisition. Q2 2026 revenue +290% YoY to KRW 17.67bn; still deeply loss-making, with the stock down roughly 55-60% from its June 2026 peak.

在干什么协作机器人手臂,以及通过 2025 年收购 ONExia 获得的北美产线末端(EOL)自动化业务。2026 年二季度收入同比 +290% 至 176.7 亿韩元;仍处于严重亏损状态,股价较 2026 年 6 月高点回落约 55-60%。

Name sheetDoosan Robotics

Why it mattersKorea's largest listed pure-play collaborative-robot maker and Universal Robots' (inside Teradyne) most direct listed competitor -- the cleanest available read on whether the cobot economics that have hurt Teradyne's Robotics segment are a UR-specific problem or a category-wide one. The 2025 acquisition of US integrator ONExia layers a North American end-of-line (EOL) packaging and palletizing automation business onto the core arm, making this a P13/P17 hybrid rather than a pure component play.

为何重要韩国最大的上市纯协作机器人厂商,也是 Teradyne 旗下 Universal Robots 最直接的上市竞争对手——是判断“拖累 Teradyne 机器人部门的协作机器人经济学”究竟是 UR 自身问题还是整个品类问题的最干净读数。2025 年收购美国集成商 ONExia,在核心手臂业务之上叠加了北美产线末端(EOL)包装与码垛自动化业务,使其成为 P13/P17 的混合体,而非单纯的部件业务。

Caveat

For Q2 2026 (three months to June 2026), disclosed as provisional consolidated results on 2026-07-24, revenue was KRW 17.67bn, up 290.0% year over year from KRW 4.53bn in Q2 2025, driven substantially by North America, which accounted for roughly 53% of H1 2026 revenue. Operating loss narrowed 7.9% year over year to KRW 14.42bn (from KRW 15.65bn) and net loss narrowed 22.8% to KRW 12.79bn.

H1 2026 cumulative: revenue KRW 32.97bn (+236.0% YoY from KRW 9.81bn), operating loss KRW 26.49bn, net loss KRW 21.96bn. Growth is real but the loss has not shrunk nearly as fast as revenue has grown, and a meaningful share of the increase looks inorganic: ONExia's own EOL packaging/palletizing segment posted full-year 2025 revenue of only USD 9.4m (~KRW 13.6bn), up just USD 1.9m year over year, which means most of the H1 2026 revenue jump is still core cobot-arm and North American solutions sales rather than the acquired integration business scaling on its own.

FY2025 full-year results showed consolidated revenue of KRW 33.0bn, down 29.6% year over year, and an operating loss of KRW 59.5bn, up 44.3% -- a year in which revenue fell and losses widened, attributed to soft developed-market demand plus one-off ONExia acquisition costs and R&D-centre buildout. The stock has been extremely volatile: it hit an all-time high of KRW 170,000 on 2026-06-02 and traded around KRW 64,700-73,000 in early August 2026, down roughly 55-60% from that peak, on a market capitalisation of approximately KRW 4.73tn against a company that has not reported a profitable quarter since its October 2023 IPO.

No humanoid-specific revenue line is disclosed; Doosan Robotics' near-term commercial reality is collaborative arms and EOL integration, not humanoid platforms, despite market commentary that frequently frames it as a humanoid name.

风险提示

2026 年二季度(截至 2026 年 6 月的三个月),2026-07-24 公布的合并口径临时业绩显示:收入 176.7 亿韩元,同比 +290.0%(2025 年二季度为 45.3 亿韩元),主要由北美市场拉动——北美占 2026 年上半年收入约 53%。营业亏损同比收窄 7.9% 至 144.2 亿韩元(上年同期 156.5 亿韩元),净亏损收窄 22.8% 至 127.9 亿韩元。

2026 年上半年累计:收入 329.7 亿韩元(同比 +236.0%,上年同期 98.1 亿韩元),营业亏损 264.9 亿韩元,净亏损 219.6 亿韩元。增长是真实的,但亏损收窄的速度远不及收入增长的速度,而且相当一部分增量似乎并非有机增长:ONExia 自身的 EOL 包装/码垛业务 2025 全年收入仅 940 万美元(约合 136 亿韩元),同比仅增加 190 万美元,这意味着 2026 年上半年收入跃升的主体仍是核心协作臂与北美解决方案销售,而非被收购的集成业务自身放量。

FY2025 全年业绩:合并收入 330 亿韩元,同比 -29.6%;营业亏损 595 亿韩元,同比扩大 44.3%——收入下滑、亏损扩大的一年,公司将其归因于发达市场需求疲软,叠加 ONExia 收购一次性费用与研发中心建设支出。股价波动极大:2026-06-02 创下 17 万韩元的历史新高,2026 年 8 月初交投于 6.47 万-7.3 万韩元区间,较高点回落约 55-60%,对应市值约 4.73 万亿韩元;公司自 2023 年 10 月上市以来从未录得盈利季度。

公司未披露任何人形机器人专项收入;尽管市场评论常将其视为人形概念股,但斗山机器人的近期商业现实仍是协作臂与 EOL 集成,而非人形平台。

ProductCollaborative robot arms, application modules, and (via ONExia) end-of-line packaging and palletizing automation

Value accrualDirect merchant competition with Universal Robots in cobot arms, plus North American systems-integration attach via the ONExia acquisition

Key riskH1 2026 revenue KRW 32.97bn (+236% YoY) against an operating loss of KRW 26.49bn; the company has not posted a profitable quarter since its October 2023 IPO, and FY2025 revenue fell 29.6% while the operating loss widened 44.3%. Stock down roughly 55-60% from its June 2026 all-time high; market cap ~KRW 4.73tn against a business still burning cash at scale.

Key risk2026 年上半年收入 329.7 亿韩元(同比 +236%),营业亏损 264.9 亿韩元;公司自 2023 年 10 月上市以来从未录得盈利季度,FY2025 收入下滑 29.6% 而营业亏损扩大 44.3%。股价较 2026 年 6 月历史高点回落约 55-60%;市值约 4.73 万亿韩元,而公司仍在大规模烧钱。

Bottlenecks瓶颈 P13 Dexterous manipulation灵巧操作 · P17 Integration, uptime, service集成、在线率与服务

Class
public
Country
KR
HQ
Suwon, South Korea
Layer
Robot OEM / Collaborative robots & integration
Systems
muscle · body
Caveat stance
avoid
Ticker
454910
Exchange
KOSPI
Valuation tier
3
Last verified
2026-08-26
Sources
S-PUB-030 · S-PUB-031
Translated fields
what_it_does_en · why_it_matters_en · caveat_en · key_risk_en
Chain position
midstream
COMPANY RESEARCH公司研究 Commercial evidence商业证据 Expand +展开 + Collapse −收起 −
Commercial evidence shows how far a company has moved from a product claim to repeatable customer value. We look for a defined customer task, paid deployment, operating results, the human help required, and economics that can support repeat use. Missing information stays unassessed; it does not count as evidence against the company.商业证据用于判断一家公司从产品宣称走到了可重复客户价值的哪一步。我们关注明确的客户任务、付费部署、运行结果、所需人工协助,以及能否支持重复采用的经济性。缺失信息保持未评估,不会被当作对公司的负面证据。 Each item below applies only to the cited document and period. It should be read with its evidence class, review date, and stated limits.以下每项内容仅适用于所引文件与期间,阅读时应同时考虑证据类别、核查日期与已说明的局限。
Customer and task
客户与任务
Not yet assessed in this structured record.本结构化记录尚未评估。
Deployment and payment
部署与付款
Not yet assessed in this structured record.本结构化记录尚未评估。
Task results and human assistance
任务结果与人工协助
Not yet assessed in this structured record.本结构化记录尚未评估。
Cost and repeat purchases
成本与复购
Not yet assessed in this structured record.本结构化记录尚未评估。
What the evidence cannot establish
证据的局限
Not yet assessed in this structured record.本结构化记录尚未评估。
How to evaluate commercial evidence如何评估商业证据