Fastest
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 →https://www.patsnap.com/resources/blog/rd-blog/quadruped-robot-learning-and-control-patent-landscape/ · Patsnap · data cut-off 2026-07-31 · downloaded from the live page
Robotics & Automation · Patent LandscapeQuadruped Robot Learning and Control Patents: Filing Trends and Open Claim Space
- Filing peaked in 2024 at 10 records, then eased off — a signal that the field's most obvious control-loop claims may already be staked out.
- China accounts for 20 of 35 receiving-office filings, versus 7 in the US and single digits elsewhere, so freedom-to-operate work has to start with CNIPA prior art.
- No assignee shows positive year-over-year momentum in the latest year, with several leading filers down to zero or -100% YoY — the leaderboard is more historical than active right now.
35
Published Records
57%
Top-5 Share of All Records
+900%
Filing Growth 2021→2024
CN
Leading Jurisdiction
Filing growth compares 2021 (1 records) with 2024 (10) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 35 records in scope (CR5), not by the ranked leaders only.
Published byPatsnap Research··6 min readSourced from Patsnap Eureka
Overview
What this landscape covers
This review tracks patent families at the intersection of quadrupedal and legged robot hardware and the control and learning methods that drive them — reinforcement-learning locomotion, sim-to-real transfer, adaptive control and model predictive control. The search combines mechanical-platform classifications with control-system and AI-computing IPC codes, so it captures both the physical robot claims and the software that governs how it walks.
Across the 2015-2026 window the dataset holds 35 published families, concentrated heavily in the last four years. Because publication typically lags filing by around eighteen months, the 2025 and 2026 counts in any trend chart understate real filing activity for those years.
Filing activity by year, 2017-2026
Source: Patsnap Eureka. Assignee ranking and totals. Derived from a Patsnap search on Quadruped Robot Learning and Control covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
10,000 free credits to start
Data
Filing trend and technology composition
The two views below show when families were filed and which technical subclasses they sit in — together they indicate whether the field is still opening up or has settled into a fixed set of claim shapes.
A short, uneven filing curve
Filings moved from zero in 2017 to a peak of 10 in 2024, passing through 6 at the 2022 midpoint. That shape is closer to a burst than a steady climb, consistent with a niche technical area reacting to a small number of triggering publications or platform launches rather than broad, sustained industry investment.
Mechanical platform claims dominate over pure software claims
B62D (motor vehicles and steering, covering legged-locomotion mechanics) accounts for 27 of the classified records, well ahead of B25J (manipulators and robots, 14) and the control-system codes G05B and G05D (13 and 12). Pure AI-computing classification G06N appears only once, which suggests most applicants are still claiming locomotion control as a mechanical or control-systems invention rather than as a standalone machine-learning method — a distinction that matters for how narrowly a claim can be designed around.
Shares are the percentage of the 35 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Source: Patsnap Eureka. Filing trend and technology composition. Derived from a Patsnap search on Quadruped Robot Learning and Control covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.
Go deeper on Quadruped Robot Learning and Control with Eureka
This page is one run against one query. Ask Eureka your own question about quadruped robot learning and control and every answer comes back with the patent numbers behind it.
Try EurekaKey Patents
The records other filings build on
Most-cited records in this dataset
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | CN114609918A | 一种四足机器人运动控制方法、系统、存储介质及设备 | 19 |
| 2 | US6961640B2 | Motion control for a legged robot | 18 |
| 3 | WO2025102514A1 | 足式机器人规划与控制方法、装置、机器人及存储介质 | 9 |
| 4 | US20240269846A1 | Method, apparatus, and device for controlling legged robot, legged robot, computer-readable storage medium, a… | 8 |
| 5 | CN117944061A | 基于模型预测控制和全身力控的足臂机器人末端跟踪方法 | 8 |
| 6 | CN112859851A | 多足机器人控制系统及多足机器人 | 8 |
| 7 | CN121232603A | 一种基于深度强化学习的四足机器人鲁棒运动控制方法 | 7 |
| 8 | CN116142349A | 多自由度脊柱关节四足机器人及其刚度自适应控制方法 | 7 |
| 9 | CN116985113A | 控制足式机器人的方法和装置及足式机器人 | 7 |
| 10 | CN118818968A | 一种基于深度强化学习的四足机器人运动控制方法 | 6 |
Citation counts reflect influence within the searched corpus and skew toward older filings; treat them as a map of what other applicants had to design around, not as a ranking of current importance.
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.
Source: Patsnap Eureka. Citation counts and representative records. Derived from a Patsnap search on Quadruped Robot Learning and Control covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Run it yourself
Put your own technology through the same analysis
Where to run it
For builders
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 →Insights
What the numbers mean for a filing decision
Three patterns stand out once family counts, geography and citations are read together.
Geography
20 of 35 filings via China
receiving office share
CNIPA is the primary prior-art pool
With China responsible for 20 of the 35 tracked filings against 7 in the US and single digits across WIPO, EPO, Hong Kong and India, any freedom-to-operate search that skips Chinese-language filings is working from an incomplete map. Several of the most-cited records in this dataset are Chinese-origin.
Receiving office data
Claim shape
27 records in B62D
mechanical/steering classification
Claims lean mechanical, not algorithmic
The dominant classification is B62D — motor vehicles and steering — rather than G06N's AI-model codes, which appear in only one record. Locomotion control is mostly being claimed as a mechanical or control-systems invention, which narrows how a pure reinforcement-learning method claim would need to be drafted to avoid overlap.
IPC composition
Momentum
0% or negative YoY for tracked leaders
latest-year filing change
The leaderboard is not currently active
None of the assignees with the largest historical family counts show positive year-over-year growth in the latest tracked year; several sit at -100% YoY. Combined with the 2024 peak and subsequent softening, this points to a field where early movers have slowed rather than one where a dominant player is still accelerating.
Recent-year momentum by assignee
Eureka AI Agent
Looking for what nobody has claimed yet?
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to quadruped robot learning and control, with the prior art for and against each one.
Source: Patsnap Eureka. Co-assignee relationships and derived observations. Derived from a Patsnap search on Quadruped Robot Learning and Control covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Players
Who is filing, and where the openings sit
Filing activity is split between Chinese universities and research institutes, a small number of large technology companies, and isolated corporate filers — with no single organisation showing sustained recent momentum.
Academic filers
6+ university/institute assignees tracked
share of named assignees
Universities hold a meaningful share of the family count
Institutions such as Zhejiang University, Tsinghua University, Guangdong University of Technology and Shandong University appear among the named assignees, several with co-assignee filings alongside nuclear-power research entities — an unusual pairing that suggests joint projects rather than pure robotics R&D.
Assignee list
Corporate filers
Sony, Tencent, UBTECH among named assignees
corporate presence
Large technology companies file selectively
Tencent, Sony Group and UBTECH each appear in the dataset, but none shows active recent-year filing momentum, indicating past exploratory work rather than an ongoing program at scale.
Recent-year momentum
Concentration
35 families, 3 co-assignee pairs
collaboration density
Collaboration is rare and narrow
Only three co-assignee pairs appear across the full dataset, each linking a single university to a single partner institute. There is no dense collaboration cluster — most families are filed by a single named assignee acting alone.
Co-assignee pairs
Recent-year filing momentum by assignee
| Assignee | Recent year | YoY |
|---|---|---|
| Zhejiang University of Technology | 1 | 0% |
| Tencent Technology (Shenzhen) Co., Ltd. | 0 | — |
| Tsinghua University | 0 | -100% |
| Zhejiang University | 0 | — |
| Guangdong University of Technology | 0 | -100% |
| Shandong University | 0 | — |
| Mitsubishi Electric Research Laboratories, Inc. | 0 | — |
| Qilu University of Technology (Shandong Academy of Sciences) | 0 | — |
Source: Patsnap Eureka. Assignee-level momentum. Derived from a Patsnap search on Quadruped Robot Learning and Control covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
What's Next
Where to take this next
The dataset points to specific follow-up work depending on whether the goal is freedom-to-operate, whitespace identification, or competitive tracking.
Run a full-text FTO check against the Chinese-origin leaders
With 20 of 35 filings routed through China and the highest-cited record originating there, a freedom-to-operate opinion built only on English-language search will miss the densest part of the claim landscape.
Explore the assignee tableTest draft claims against the under-claimed branches
Sim-to-real transfer methods and terrain-adaptive gait logic show thin filing density relative to core locomotion-control claims, which may leave room for a narrowly drafted method claim.
Review the IPC breakdownMonitor for renewed filing activity
Every tracked leader shows flat or negative year-over-year filing in the latest period; a return to positive momentum from any one of them would be an early signal worth flagging.
Check momentum by assigneeSource: Patsnap Eureka. Forward-looking reading of the same dataset. Derived from a Patsnap search on Quadruped Robot Learning and Control covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
FAQ
Common questions about this landscape
Within this dataset the named assignees include Chinese universities such as Zhejiang University and Tsinghua University alongside corporate filers like Tencent, Sony Group and UBTECH, but none currently shows positive year-over-year filing growth. The field does not have a single dominant, actively-filing leader right now; the largest family counts are historical rather than reflecting an ongoing filing push. Anyone assessing competitive risk should weight recent-year activity, not just cumulative family count.
Mostly hardware-adjacent. The IPC composition shows 27 of 35 classified records in B62D, the motor-vehicles-and-steering classification that covers legged-locomotion mechanics, compared with just one record classified under G06N, the AI-computing code. This means most applicants are claiming locomotion control as a mechanical or control-systems invention rather than as a pure machine-learning method, which affects how narrowly a competing algorithmic claim needs to be drafted.
China is the largest receiving office in this dataset by a wide margin, with 20 of 35 tracked filings, compared with 7 in the United States and low single digits across WIPO, EPO, Hong Kong and India. Any search or filing strategy that excludes Chinese-language patent literature will miss the majority of the documented activity in this space, including some of the most-cited records.
Filing activity rose from zero in 2017 to a peak of 10 families in 2024, but the trend has since flattened, and recent-year momentum figures show flat or negative year-over-year change for every tracked assignee. Because publication lags filing by roughly eighteen months, the very latest years will always look artificially quiet, but the pattern through 2024 already suggests a burst of activity rather than a sustained climb.
The highest-cited record in this dataset, CN114609918A, covers a quadruped robot motion control method and system and has been cited 19 times within the searched corpus; other frequently cited records include US6961640B2 on legged-robot motion control and WO2025102514A1 on legged-robot planning and control. High citation counts mark records that later filers had to design around, but they also skew toward older filings simply because they have had more time to accumulate citations, so they should be read as an influence signal rather than a current-relevance ranking.
Answers are grounded in the same dataset. Derived from a Patsnap search on Quadruped Robot Learning and Control covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Research Quadruped Robot Learning and Control in depth with Eureka
Go past this page: query the whole quadruped robot learning and control 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.