Humanoid Robots Patent Landscape 2026
Humanoid Robots Patent Landscape in 2026
The humanoid robot field is in an active growth stage, with 8,105 patent families on record and a 19% gain over the recent window — a corpus still expanding as publication lag understates the most recent years. Sony Group and Honda Motor hold the top two positions in a moderately concentrated field where the five largest filers account for 39% of the hundred largest filers’ combined output.
Sony and Honda lead a two-tier field with a clear gap to challengers
Sony Group Corp and Honda Motor Co Ltd occupy the top two positions with 826 and 807 patent families respectively, separated by fewer than 20 families — making the leadership effectively co-dominant. Boston Dynamics (595) and SoftBank Robotics. Europe (508) form a second tier, with UBTECH Robotics (333) anchoring a third.
The top five filers together hold 39% of the hundred largest filers’ combined total, indicating moderate concentration: no single entity commands an overwhelming share, yet the gap between tier one and the broader field is substantial enough to create meaningful barriers to catch-up on volume alone.
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 1 | Sony Group Corporation | 826 | |
| 2 | Honda Motor Co., Ltd. | 807 | |
| 3 | Boston Dynamics, Inc. | 595 | |
| 4 | SoftBank Robotics Europe | 508 | |
| 5 | UBTECH Robotics Corp Ltd | 333 | |
| 6 | Sanctuary Cognitive Systems Corp | 229 | |
| 7 | Toyota Motor Corporation | 225 | |
| 8 | Samsung Electronics Co., Ltd. | 186 | |
| 9 | Beijing Institute of Technology | 152 | |
| 10 | Tencent Technology (Shenzhen) Co., Ltd. | 146 |
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 11 | Aldebaran Robotics SA | 117 | |
| 12 | Shanghai Fourier Intelligence Co., Ltd. | 109 | |
| 13 | Hangzhou Yushu Technology Co., Ltd. | 94 | |
| 14 | Zhejiang University | 94 | |
| 15 | Sarcos Corp | 94 | |
| 16 | Beijing Xiaomi Robot Technology Co., Ltd. | 92 | |
| 17 | Harbin Institute of Technology | 92 | |
| 18 | Tsinghua University | 88 | |
| 19 | Kawasaki Heavy Industries, Ltd. | 86 | |
| 20 | Shenzhen Zhuji Power Technology Co., Ltd. | 85 |
The near-parity between Sony and Honda at the top suggests two distinct strategic paths — Sony emphasising manipulation and sensor-heavy robotics, Honda carrying deep walking-robot heritage — coexist at the frontier. New entrants such as Sanctuary Cognitive Systems and Beijing Xiaomi Robot Tech signal that the competitive set is still forming.
Filings from 2025 and 2026 are materially under-counted due to standard patent publication lag and should not be read as reflecting true recent activity levels. Longer-window growth, applicant concentration, and technology-route coverage are therefore more reliable signals than the latest-year bar alone.
Sustained growth with manipulator mechanics at the core and AI control emerging at the margin
The annual filing trend and the IPC technology composition together reveal a field that is broadening in scope while remaining anchored in mechanical foundations. The trend chart tracks annual filing volumes; the composition chart shows how patent activity distributes across technology branches.
Annual filing trend
Annual filings rose from 602 in 2017 to 921 in 2023, with a marked step-up to 1,122 in 2024. The 2025 and 2026 bars are severely truncated by publication lag and should not be treated as evidence of slowing; the lifecycle evidence confirms the field remains in a growth stage. The 19% recent-window growth figure reinforces this read.
↗ Hover for values · click a bar to ask EurekaTechnology composition
B25J (Manipulators and robots) is by far the dominant IPC class, reflecting the primacy of mechanical arm and body structure patents. B62D (legged-robot locomotion) is a strong second, consistent with bipedal walking as a defining challenge. Smaller but notable branches — G06N (AI models), G05D (autonomous control), and G05B (regulating systems) — show the field beginning to absorb machine-learning and software-defined control methods, though these remain well behind the mechanical core.
↗ Hover for values · click a bar to ask EurekaHighly cited patent families surfaced by the query
Citation-heavy patent families returned by the query. Use this section as citation context, not as a curated list of the most topic-specific patents.
Method for gait generation and tracking control fo…
A control method that executes a numerical non-linear optimization procedure with forward simulation of the full dynamics of the bipedal robot (including ground collisions) to compute a foothold, which produces the desired center of mass velocity during the next step of the walking cycle. The controller includes a gait generator that outputs desired joint… (excerpt from the patent abstract)
| # | Patent | Citations |
|---|---|---|
| 1 | Obstacle recognition method for autonomous robots | 728 |
| 2 | Charging system for mobile robot, method for searc… | 647 |
| 3 | Robot remote-control system and software distribut… | 447 |
| 4 | Methods and systems for food preparation in a robo… | 446 |
| 5 | Obstacle recognition method for autonomous robots | 444 |
| 6 | Robotic manipulation methods and systems for execu… | 444 |
| 7 | Interactive systems employing robotic companions | 417 |
| 8 | Systems and methods for operating a robotic system… | 365 |
Ranked by total forward citations. Citation counts favour older and broadly cited patent families, and broad or adjacent patents may appear when they match the search scope. Treat this section as citation context, not as a curated list of the most topic-specific patents. Some patent titles may be shown in their original, non-English language where an accurate translation could not be guaranteed.
What the competitive structure means for R&D investment decisions
Four structural dimensions — lifecycle stage, concentration, collaboration patterns, and geography — each carry distinct implications for teams deciding where to allocate IP effort.
Growth stage: rising volume, mechanical core established
The lifecycle evidence places humanoid robots firmly in the Growth stage, with annual filings still rising and the multi-year corpus up 19%. The mechanical-manipulation and legged-locomotion branches are mature enough that incremental patents face high prior-art density; differentiation increasingly requires novel combinations with AI control, perception, or actuation subsystems. Teams entering now need a clear sub-domain thesis rather than a broad mechanical platform play.
Growth stageModerate concentration with an open mid-field
The top five filers account for 39% of the hundred largest filers’ combined output, and the leadership is split almost evenly between Sony and Honda. Below the top tier, the gap to rank-five (UBTECH, 333 families) and beyond is large enough that a focused entrant in a specific sub-domain — say, dexterous manipulation or AI-driven locomotion control — could establish a credible position without matching the leaders’ absolute volume. The mid-field (ranks 6–20) is diverse and includes both corporates and universities.
Moderate concentrationIndustry-academic partnerships concentrated around Sony and SoftBank
The most active co-filing pair is Sony Group Corp with individual inventor Jinichi Yamaguchi (70 joint families), reflecting Sony’s deep individual-expert model. SoftBank Robotics Europe co-files extensively with INRIA (44 families) and CNRS (14 families), embedding French public-research IP into a commercial platform. Toyota Motor co-files with Toyota Central R&D Labs (14 families) and the University of Tokyo (4 families). Honda collaborates with Tokyo Institute of Technology (6 families) and the University of Tokyo (4 families). These partnerships signal that access to elite academic robotics groups is a recurring competitive lever.
Industry-academicChina leads filings; US is the primary validation market
China accounts for the largest share of patent records filed in this space, with the United States second and Europe (EPO) third. Japan and South Korea follow. This distribution reflects both China’s aggressive domestic filing strategy and the US’s role as a key commercialisation and enforcement jurisdiction. Teams seeking global freedom-to-operate should treat the China, US, and EPO offices as the minimum coverage set; Germany and South Korea represent secondary validation markets worth monitoring.
China · US · EPOGo beyond the landscape: Eureka’s TRIZ Solution agent breaks down an R&D problem and returns patented concept solutions, each with a technical approach and cited patent & literature evidence.
| Applicant | Collaborator | Co-filings |
|---|---|---|
| Sony Group Corporation | Jinichi Yamaguchi | 70 |
| SoftBank Robotics Europe | INRIA (French National Institute for Research in Digital Science and Technology) | 44 |
| Sony Group Corporation | ATR International (Advanced Telecommunications Research Institute International) | 14 |
| SoftBank Robotics Europe | CNRS (French National Centre for Scientific Research) | 14 |
| Toyota Motor Corporation | Toyota Central R&D Labs, Inc. | 14 |
| Aldebaran Robotics SA | INRIA (French National Institute for Research in Digital Science and Technology) | 12 |
| Honda Motor Co., Ltd. | Tokyo Institute of Technology | 6 |
| Honda Motor Co., Ltd. | The University of Tokyo | 4 |
| Toyota Motor Corporation | The University of Tokyo | 4 |
| Sony Group Corporation | Sony Electronics Inc. | 3 |
Co-filing pairs, ranked by the number of jointly-filed patent families.
Sony and Honda co-lead; Boston Dynamics and Sanctuary show the strongest recent momentum
The top two applicants are separated by fewer than 20 patent families, yet their technical emphasis differs meaningfully. Below them, Boston Dynamics and Sanctuary Cognitive Systems are the momentum stories of the recent filing window.
Sony Group Corp
Sony leads with 826 patent families, concentrated heavily in B25J manipulator and robot mechanics across multiple sub-classes (B25J5, B25J13, B25J9). Despite its cumulative scale, Sony’s recent three-year filing trend is down 73% from its prior period — suggesting a strategic consolidation or portfolio optimisation phase rather than continued volume expansion. The close co-filing relationship with individual expert Jinichi Yamaguchi (70 joint families) is a distinctive structural feature of Sony’s IP model.
families: 826Boston Dynamics Inc
Boston Dynamics ranks third with 595 patent families and is the strongest momentum story among established players, with recent filings up 12% versus its prior three-year period. Its technology emphasis spans B25J9 (robot configuration), B62D57 (legged and walking robots), and G05D1 (autonomous motion control) — a profile that uniquely bridges mechanical structure with software-defined autonomy. Sanctuary Cognitive Systems, entering as a new entrant with 229 families and 197 recent-window families, is the fastest-rising challenger in the field.
families: 595| Applicant | Recent (3 yrs) | Trend |
|---|---|---|
| Sony Group Corporation | 25 | ▼ -73% |
| Honda Motor Co., Ltd. | 38 | ▲ +52% |
| Boston Dynamics, Inc. | 217 | ▲ +12% |
| UBTECH Robotics Corp Ltd | 113 | ▼ -36% |
| Sanctuary Cognitive Systems Corp | 197 | ▲ new entrant |
| Toyota Motor Corporation | 9 | ▼ -31% |
| Samsung Electronics Co., Ltd. | 1 | ▲ new entrant |
| Google LLC | 31 | ▼ -16% |
AI control and digital data processing are under-served relative to the mechanical core
The IPC branches below identify technology areas that are adjacent to the dominant manipulator and locomotion classes but carry disproportionately fewer patent records. These are observations of relative sparsity; technical and commercial value must be assessed independently before treating them as entry points.
G06N · AI models applied to humanoid control
G06N (Computing based on AI models) accounts for 609 patent records — roughly 6% of the B25J volume — despite the well-documented centrality of machine learning to next-generation humanoid behaviour. The sparsity is plausible given that many AI methods are filed as software patents under G06F rather than G06N, but the gap still indicates that explicit AI-model claims tied to humanoid embodiment remain relatively open. Teams with strengths in reinforcement learning for dexterous manipulation or whole-body motion planning could find less prior-art congestion here than in the mechanical branches. Entry paths include filing on training pipelines, sim-to-real transfer methods, and learned locomotion policies specific to bipedal or humanoid form factors.
Search this in Eureka →G05B · Control and regulating systems for humanoid platforms
G05B (Control and regulating systems) shows 659 patent records against a B25J base of over 10,000, a ratio that points to under-representation of closed-loop control architectures explicitly scoped to humanoid robots. Physical therapy and rehabilitation adjacencies (A61H, 152 records) and force/pressure measurement (G01L, 130 records) sit nearby and are similarly sparse. A team working on impedance control, compliant joint regulation, or safe human-robot interaction control laws could address a genuine technical gap while occupying IP space that the volume leaders have not densely covered. The G05B branch also intersects naturally with industrial safety standards — a regulatory-driven entry path.
Search this in Eureka →Leaders diverge on locomotion versus manipulation emphasis and AI integration
Strength of each leader across the main technology routes.
| Player | B25J 9 · Manipulators & robots | B62D 57 · Legged & walking robots | B25J 19 · Manipulators & robots | B25J 13 · Manipulators & robots | B25J 5 · Manipulators & robots |
|---|---|---|---|---|---|
| Sony Group Corporation | Moderate · 267 | Moderate · 195 | Moderate · 163 | Strong · 458 | Strong · 558 |
| Honda Motor Co., Ltd. | Moderate · 209 | Strong · 355 | Moderate · 188 | Strong · 317 | Strong · 448 |
| Boston Dynamics, Inc. | Strong · 429 | Strong · 263 | Emerging · 64 | Emerging · 65 | Emerging · 48 |
| UBTECH Robotics Corp Ltd | Strong · 240 | Strong · 144 | Emerging · 37 | Moderate · 74 | Absent |
| SoftBank Robotics Europe | Strong · 232 | Absent | Strong · 138 | Absent | Moderate · 54 |
| Toyota Motor Corporation | Absent | Moderate · 61 | Absent | Strong · 84 | Strong · 149 |
| Sanctuary Cognitive Systems Corp | Strong · 187 | Absent | Moderate · 51 | Moderate · 42 | Absent |
Frequently asked questions
The evidence covers 8,105 patent families in scope globally. This total reflects families filed and published as of the data extraction date; filings from the most recent 18–24 months are under-counted due to standard patent publication lag.
Sony Group Corp leads with 826 patent families, followed closely by Honda Motor Co Ltd with 807. Boston Dynamics (595) and SoftBank Robotics Europe (508) form a second tier. Together, the top five filers account for 39% of the hundred largest filers’ combined output.
China is the leading filing jurisdiction by patent record count, followed by the United States and Europe (EPO). Japan and South Korea are the next most active jurisdictions. For global freedom-to-operate, the China, US, and EPO offices represent the minimum coverage set.
The evidence places the field in the Growth stage. Annual filings have risen consistently from 602 in 2017 to over 1,100 in 2024, and the recent-window growth rate is 19%. The 2025 and 2026 data are truncated by publication lag and do not indicate any slowdown.
B25J (Manipulators and robots) is the dominant branch by a wide margin, followed by B62D (legged-robot locomotion). AI-related branches — G06N (AI models) and G05B (control systems) — are present but carry far fewer records relative to the mechanical core, indicating earlier-stage IP density.
Among established players, Honda Motor Co Ltd shows the strongest positive momentum at +52% in its recent three-year period versus the prior period. Boston Dynamics is up 12%. Sanctuary Cognitive Systems and Samsung Electronics are identified as new entrants in the recent window, with Sanctuary accumulating 197 recent-window families from a standing start.
Ready to map your own humanoid robot patent landscape?
Join 18,000+ innovators using PatSnap Eureka to map any technology landscape: search 2B+ patents and papers, surface key assignees, and generate a report like this in minutes.
Disclaimer. This page is generated from PatSnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. PatSnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.