Humanoid Robot Control System Patents: Who Leads, Gaps 2026
- Filing peaked in 2025 at 39 records after a decade of slow, single-digit growth from 2017 — the field only recently became a filing priority, not a mature one.
- B25J dominates but B62D is close behind 184 records sit in manipulators/robots against 83 in motor-vehicle steering classes, showing how much bipedal control IP still borrows from legged-vehicle mechanics.
- The strongest co-filing pair appears 28 times far ahead of any other assignee pairing, pointing to one research-lab-to-commercial-arm pipeline as the dataset's clearest technology transfer route.
What this landscape covers
This landscape tracks patent families where whole-body control, balance control, model predictive control or compliance control are claimed against humanoid, bipedal or legged robot bodies, filtered to the core mechanical and control IPC codes for manipulators, legged vehicles and non-electric control systems. It spans filings from 2015 through the 2026 cutoff, covering 210 published families across six major receiving offices.
Because publication trails filing by roughly 18 months, the 2025 and 2026 counts in the trend chart are the least complete and will revise upward as later filings publish. The receiving-office split — led by the United States and China, with the EPO and PCT route some distance behind — indicates where applicants are currently prioritising enforceable rights rather than where invention is happening.
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
Filing trends and technology composition
Two views of the same 210-family dataset: how filing volume has moved year over year, and which IPC subclasses the claims actually sit in.
A late, steep ramp rather than steady growth
Annual filings moved from 2 in 2017 to a midpoint of 10 in 2022, then climbed to a peak of 39 in 2025 — the growth curve is backloaded, not linear, which means most of the claim landscape here is under five years old.
Manipulator and vehicle-steering codes dominate
B25J (manipulators & robots) covers 184 of 210 records and B62D (motor vehicles & steering) covers 83, with G05D and G05B control-systems codes trailing well behind — AI-specific codes like G06N and G06T appear in only a handful of records, suggesting most claims still anchor on mechanical/control architecture rather than learned policies.
Shares are the percentage of the 210 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Humanoid Robot Control System with Eureka
This page is one run against one query. Ask Eureka your own question about humanoid robot control system and every answer comes back with the patent numbers behind it.
Try EurekaThe records other filings build on
Planning and control method for legged robot, apparatus, robot, and storage medium
Provided are a planning and control method for a legged robot, an apparatus, a robot, and a storage medium. The method includes determining a foot-end reference state for each leg from movement state and swing parameters, planning an impact-aware swing leg trajectory using a foot-end dynamic model and a discrete collision model, and performing impact-aware whole-body control based on a whole-body dynamic model and the discrete collision model.Filed by Tsinghua University, published 2026-04-09 — one of the most recent records in the dataset and a useful marker of where whole-body control claims are heading.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US5644204A | Anti-slip control for a legged robot and realisitc simulation of a legged creature | 95 |
| 2 | US20120245734A1 | Humanoid robot push recovery on level and non-level ground | 63 |
| 3 | US7136722B2 | Method for generating a motion of a human type link system | 62 |
| 4 | EP1393866A1 | Apparatus walking with two legs; walking control apparatus; and walking control method thereof | 60 |
| 5 | US20030220714A1 | Method for generating a motion of a human type link system | 57 |
| 6 | US20040051493A1 | Apparatus walking with two legs, walking control apparatus, and walking control method thereof | 55 |
| 7 | CN106078752A | 一种基于Kinect的仿人机器人人体行为模仿方法 | 54 |
| 8 | US20210387346A1 | Humanoid robot for performing maneuvers like humans | 53 |
| 9 | US6583595B1 | Robot and joint device for the same | 51 |
| 10 | CN108858208A | 一种复杂地形仿人机器人自适应平衡控制方法、装置和系统 | 47 |
Citation counts accumulate over time, so older records such as the 1990s anti-slip control patent and early-2000s walking-control filings lead the table by volume — that reflects age and influence on later drafting, not current relevance to new filings.
Patent titles are shown in the language they were filed in, not translated, so that each record stays verifiable against the original filing — a translated title will not match in Eureka or in any national register. Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the numbers mean for a filing decision
Three read-outs from the trend, technology split and citation data that matter more for strategy than the raw counts alone.
The field's IP activity is recent and still accelerating
A move from 2 filings in 2017 to 39 in the peak year, with the midpoint year at only 10, shows that most of the claim space was staked out in the last three to four years. Treat pre-2020 records as foundational prior art rather than as the current competitive set.
Vehicle-steering mechanics still shape bipedal control claims
B62D (motor vehicles & steering) sitting at 83 of 210 records, second only to B25J, means a meaningful share of legged-robot balance and compliance control claims are drafted by or against parties with automotive steering-control backgrounds. Search this subclass explicitly, not just robotics classes.
One lab-to-commercial pairing dominates joint filing
Of only 6 co-assignee pairs in the dataset, the strongest single pairing appears 28 times — an order of magnitude ahead of the next pairing at 9. That concentration marks a specific research-to-product pipeline worth tracking for licensing or freedom-to-operate purposes.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to humanoid robot control system, with the prior art for and against each one.
Who is filing, and who has slowed down
Recent-year momentum tells a different story than cumulative counts: several assignees with historically strong positions show flat or negative filing growth in the latest year.
Several established filers went to zero in the latest year
Assignees with meaningful prior filing histories show zero or sharply negative year-over-year activity in the most recent period, which is consistent with publication lag but also suggests some programmes have paused new filing rather than simply waiting on grants.
United States and China lead filing volume closely
The gap between the top two receiving offices is narrow — 65 versus 59 — with Europe and the PCT route well behind. Any competitive monitoring program needs both jurisdictions covered, not just one.
Joint filing is rare and concentrated
Only six co-assignee pairs appear across 210 families, and filing partnerships that do exist are dominated by one research-institute-to-commercial-robotics relationship. Most assignees in this space file solo.
| Assignee | Recent year | YoY |
|---|---|---|
| Digital Artificial Intelligence Corporation | 1 | -95% |
| Beijing Institute of Technology | 1 | -50% |
| French National Institute for Research in Digital Science and Technology (INRIA) | 0 | — |
| Tencent Technology (Shenzhen) Co., Ltd. | 0 | -100% |
| SoftBank Robotics Europe | 0 | — |
| UBTECH Robotics Corp Ltd | 0 | — |
| The University of Tokyo | 0 | — |
| Aldebaran Robotics | 0 | — |
Where to take this analysis
The trend and technology split point to a fast-moving, recently active field with narrow but real white space around AI-conditioned control and terrain-adaptive mechanics.
Watch the 2025-2026 filings as they mature
The most recent two years are the least complete in this dataset due to publication lag. Re-run the trend query periodically to see whether the 2025 peak holds or is revised upward further.
Track filing trends in EurekaSearch the under-claimed branches directly
Learned whole-body policies and vision-conditioned balance recovery sit outside the dominant B25J/B62D mechanical claims. A targeted search here surfaces filings that a standard humanoid-robot query would miss.
Explore white space in EurekaMap the strongest co-filing pipeline
One assignee pairing accounts for 28 of the dataset's co-filed families. Understanding that relationship's scope and licensing posture matters more than surveying all six pairs equally.
Review assignee relationships in EurekaCommon questions about humanoid robot control system patents
The dataset's assignee ranking (rendered separately on this page) shows filing concentrated among a small group of research institutes and robotics manufacturers, with a long tail of single-filing entrants. Cumulative counts favour organisations that have been filing since the mid-2010s, but recent-year momentum data shows several of these same assignees filing zero or fewer new families in the latest year — so the leaderboard by total count and the leaderboard by current activity are not the same list. Anyone doing competitive tracking should check both views rather than relying on lifetime totals alone.
The core codes are B25J (manipulators and robots), which covers 184 of the 210 records in this dataset, and B62D (motor vehicles and steering), which covers 83 — a substantial secondary cluster that reflects shared mechanics between legged robots and vehicle steering-control systems. G05D and G05B cover general control-system claims, while G06N and G06T appear only in a small number of records tied to AI-based motion planning and vision processing. A thorough freedom-to-operate search needs to cover all of these, not just the robotics-specific codes.
Yes, and recently so: annual filings moved from 2 in 2017 to a midpoint of 10 in 2022 and peaked at 39 in 2025, an accelerating rather than steady trend. Because publication lags filing by roughly 18 months, the 2025 and 2026 figures will likely revise upward as more filings publish, meaning the true peak may be even higher than currently visible. This is a young, fast-moving filing landscape rather than a mature one with declining activity.
US20260097488A1, filed by Tsinghua University and published April 2026, claims a planning and control method that ties foot-end trajectory planning to whole-body dynamic control using a discrete collision model for impact-aware movement. Its recency means its actual claim scope and prosecution outcome are not yet settled, but its approach — linking swing-leg trajectory planning directly to whole-body control through a shared collision model — sits close to the terrain-adaptive control work other filers in this dataset are also pursuing. Teams working on uneven-terrain locomotion should read the full claim set rather than the abstract alone before designing similar control architectures.
Relative to the dense B25J and B62D mechanical claim base, AI-conditioned control approaches are thin: G06N-tagged claims appear in only 12 of 210 records and G06T-tagged vision-processing claims in 11, despite the broader industry's interest in learned locomotion policies. Toy-scale bipedal mechanisms under A63H also show minimal filing activity at just 3 records. These are the branches where claim space is least occupied, though thin filing can also mean thin commercial interest, so any white-space claim should be paired with a market read before drafting.
Research Humanoid Robot Control System in depth with Eureka
Go past this page: query the whole humanoid robot control system corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
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.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.