Industrial Robot AI/ML Patent Landscape 2026
The industrial robot AI/ML patent field is in a growth phase, with a broad base of applicants spanning pure-play robotics startups, diversified electronics majors, and academic institutions — yet the top five filers hold a modest share, signaling a still-fragmented competitive structure. Dexterity Inc leads by patent family count, but no single player dominates, and momentum data show multiple new entrants accelerating rapidly.
Fragmented field with no runaway leader — Dexterity Inc holds the top position
Dexterity Inc ranks first with 59 patent families, ahead of Ocado Innovation Ltd (45), ABB (Schweiz) AG (40), Mitsubishi Electric Corp (36), and Mitsubishi Electric Research Laboratories Inc (30). The top five account for 18% of the combined output of the hundred largest filers, confirming a fragmented competitive landscape rather than a duopoly or oligopoly.
The gap between first and fifth place spans only 29 patent families, a relatively narrow tier differential that leaves the ranking susceptible to reordering as new entrants accumulate filings. Samsung Electronics, Amazon Tech, and several Indian academic institutions appear in the top twenty, underlining the diversity of participant types.
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 1 | Dexterity Inc | 59 | |
| 2 | Ocado Innovation Ltd | 45 | |
| 3 | ABB (Schweiz) AG | 40 | |
| 4 | Mitsubishi Electric Corp | 36 | |
| 5 | Mitsubishi Electric Research Laboratories Inc | 30 | |
| 6 | Samsung Electronics Co. Ltd | 28 | |
| 7 | Amazon Technologies Inc | 25 | |
| 8 | Mazor Robotics Ltd | 24 | |
| 9 | South China University of Technology | 24 | |
| 10 | Amgen Inc | 23 |
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 11 | LG Electronics Inc | 23 | |
| 12 | Fanuc Ltd | 19 | |
| 13 | Nimble Robotics Inc | 18 | |
| 14 | Canvas Construction Inc | 18 | |
| 15 | SATHYABAMA INST OF SCI & TECH DEEMED TO BE UNIV | 17 | |
| 16 | Intelligrated Headquarters LLC | 17 | |
| 17 | Sony Group Corp | 17 | |
| 18 | GDM Holding LLC | 16 | |
| 19 | ABB Research Ltd | 16 | |
| 20 | Toyota Motor Corporation | 16 |
The leaders’ positions imply that early-mover advantages are still being established: Dexterity and Ocado are both classified as new entrants in momentum terms, meaning their current rank reflects recent concentrated filing rather than a long-accumulated portfolio — a pattern that can indicate either genuine R&D acceleration or strategic portfolio building ahead of commercialization.
Filing counts for 2025 and 2026 are understated due to typical patent publication lag of 18–24 months; the apparent recent totals should be treated as lower bounds rather than final figures. Longer-window growth, applicant concentration, and technology-route coverage are therefore more reliable signals than the latest-year bar alone.
Sustained filing growth and a manipulator-dominated technology mix with expanding AI adjacencies
Annual filing volume and the IPC class breakdown together reveal both the pace at which the field is scaling and the specific technical sub-domains attracting investment. Reading both charts side by side helps identify where growth is broad-based versus concentrated in a single branch.
Annual filing trend
Filings grew from 58 in 2017 to a recent-window high, with a 53% increase recorded over the most recent comparable period — consistent with the lifecycle assessment of active growth. The 2025 and 2026 bars are materially understated by publication lag and should not be read as plateaus or declines; final counts for those years will be higher once pending applications publish.
↗ Hover for values · click a bar to ask EurekaTechnology composition
B25J (Manipulators & robots) dominates the corpus by a wide margin, reflecting the core mechanical subject matter. G06N (Computing based on AI models) is the second-largest branch, confirming that AI/ML methods are now a substantial and distinct layer of the IP stack rather than a peripheral annotation. G06T (Image data processing), G05B (Control & regulating systems), and A61B (Diagnosis & surgery) follow, indicating that perception, closed-loop control, and surgical robotics are the three leading application vectors beyond the core manipulator class.
↗ 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.
System and method for building machine learning or…
This application relates to a method and a system for building machine learning or deep learning data sets for automatically recognizing labels on items. The system may include an optical scanner configured to capture an item including one or more labels provided thereon, the item captured a plurality of times at different positions with respect to the… (excerpt from the patent abstract)
| # | Patent | Citations |
|---|---|---|
| 1 | Real-Time Determination of Object Metrics for Traj… | 248 |
| 2 | Sensing arrangements for robot-assisted surgical p… | 235 |
| 3 | Robotic grasping of items in inventory system | 233 |
| 4 | Networked robotic manipulators | 209 |
| 5 | Method and system for programming a robot | 174 |
| 6 | Method and a system for programming an industrial … | 163 |
| 7 | Mobile robot manipulator | 158 |
| 8 | Input controls for robotic surgery | 157 |
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 positioning
Four structural dimensions — lifecycle stage, applicant concentration, collaborative activity, and geographic spread — each carry distinct implications for where and how to invest in industrial robot AI/ML IP.
Growth stage: annual filings still rising
The field is classified as Growth, with annual filing volume continuing to rise and the most recent years understated by publication lag. A 53% growth figure over the recent window confirms the field has not yet passed its peak. Entrants filing now are building portfolios in a window before the market consolidates, but rising volume also means freedom-to-operate searches are becoming more complex each year.
Growth stageFragmented top tier — 18% share across five leaders
The top five filers account for 18% of the hundred largest filers’ combined output, a low concentration figure that reflects genuine diversity of approach rather than a settled hierarchy. The narrow family-count gap between first and fifth place means the ranking can shift with a single sustained filing campaign. New entrants classified in momentum data — Dexterity, Ocado, Samsung Electronics, and Amazon Tech — are already inside the top ten, compressing the timeline for further reshuffling.
Low concentrationSouth China University of Technology is the most active co-filer
South China University of Technology appears in three distinct collaboration pairs: with Guangdong Zhike Intelligent Technology Co. (5 co-filed families), with Foshan Huashu Robot Co. (1 family), and with the Zhongshan Modern Industrial Technology Research Institute (1 family). Ocado Innovation Ltd co-filed with Kindred Systems Inc across two recorded pairings (4 and 2 families respectively). Collaboration density is currently low overall, suggesting the ecosystem has not yet consolidated around a small set of anchor partnerships.
Emerging ecosystemUS leads filings; China and India are significant secondary offices
The United States is the leading filing jurisdiction, followed by China and India as the next two largest destinations. WIPO PCT and the European Patent Office are close in volume, indicating that applicants with cross-border commercialization intent are actively pursuing international protection. South Korea, Canada, Australia, and the United Kingdom form a smaller third tier. Japan, historically a major robotics jurisdiction, appears below several emerging markets in raw filing volume, which may reflect applicant strategy choices or portfolio concentration differences.
US-led, globally distributedGo 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 |
|---|---|---|
| South China University of Technology | Guangdong Zhike Intelligent Technology Co. Ltd | 5 |
| Ocado Innovation Ltd | KINDRED SYSTEMS INC(US) | 4 |
| Ocado Innovation Ltd | Kindred Systems Inc | 2 |
| South China University of Technology | Foshan Huashu Robot Co. Ltd | 1 |
| South China University of Technology | Zhongshan Modern Industrial Technology Research Institute of South China University of Technology | 1 |
Co-filing pairs, ranked by the number of jointly-filed patent families.
Dexterity Inc and Ocado Innovation Ltd lead, both as new entrants with rapidly built portfolios
The top two filers entered the ranking as new entrants — meaning their current positions reflect concentrated recent filing rather than decades-long portfolio accumulation. Their technology emphases differ subtly, with Dexterity more weighted toward material handling and Ocado more focused on core manipulator mechanics.
Dexterity Inc
Dexterity Inc holds 59 patent families, the largest count in the corpus, and is classified as a new entrant with 44 families filed in the recent window — indicating that nearly all of its portfolio is freshly minted. Its primary technical focus is B25J 9 (manipulator programs and control), supplemented by B65G 61 (conveying and material handling) and B25J 19 (safety and sensing), a combination that maps tightly onto AI-driven warehouse and logistics automation.
families: 59Ocado Innovation Ltd
Ocado Innovation Ltd holds 45 patent families and is also classified as a new entrant, with 30 families in the recent period. Its portfolio concentrates on B25J 9 (manipulator control), B25J 15 (grippers and end-effectors), and B25J 19 (safety), reflecting a gripper-centric, fulfillment-oriented strategy consistent with Ocado’s automated warehouse business. The collaboration with Kindred Systems Inc (6 co-filed families across two pairings) suggests an open-innovation posture alongside its internal filing program.
families: 45| Applicant | Recent (3 yrs) | Trend |
|---|---|---|
| Dexterity Inc | 44 | ▲ new entrant |
| Ocado Innovation Ltd | 30 | ▲ new entrant |
| ABB (Schweiz) AG | 13 | ▼ -32% |
| Mitsubishi Electric Corp | 18 | ▲ +80% |
| Canvas Construction Inc | 5 | ▲ new entrant |
| Mitsubishi Electric Research Laboratories Inc | 20 | ▲ +100% |
| Samsung Electronics Co. Ltd | 18 | ▲ new entrant |
| Amazon Technologies Inc | 2 | ▲ new entrant |
Under-served branches adjacent to the AI/manipulator core
Several IPC classes appear at notably lower shares relative to the dominant B25J class, yet have plausible technical and commercial connections to the core industrial robot AI/ML stack. These are observations of relative sparsity and are not validated market opportunities; each would require deeper prior-art and freedom-to-operate analysis before investment.
G05B · Control & Regulating Systems
G05B holds 266 records — roughly 5% of the corpus relative to B25J’s dominant share — despite closed-loop control being a foundational requirement for any AI-driven industrial robot. The gap between the scale of AI model filing (G06N at 587) and control-system filing suggests that applicants are claiming the intelligence layer more aggressively than the underlying control architecture. Entrants with expertise in adaptive control, model-predictive control, or safety-certified PLC programming may find comparatively less crowded ground here, particularly for real-time feedback loops integrating learned policies.
Search this in Eureka →G06V · Image/Video Recognition
G06V (image and video recognition) appears at 176 records — approximately 3% of the corpus — despite perception being a central bottleneck in robot dexterity and autonomous manipulation. Given the heavy citation activity around trajectory determination and sensing arrangements in the top-cited patents, the relatively sparse G06V branch suggests that vision-based perception for manipulation is either being claimed under G06T or remains an under-patented area. Teams working on real-time 3-D object recognition, pose estimation, or multi-modal sensor fusion for gripper guidance may find this branch less congested than the manipulator core.
Search this in Eureka →How leading filers differ by technology route
Route coverage across the main technology branches in the current evidence set.
| Player | B25J 9 · Manipulators & robots | G06N 3 · Computing based on AI models | B25J 13 · Manipulators & robots | B25J 19 · Manipulators & robots | B25J 15 · Manipulators & robots |
|---|---|---|---|---|---|
| Dexterity Inc | Strong · 57 | Absent | Emerging · 11 | Moderate · 12 | Absent |
| Nimble Robotics Inc | Absent | Absent | Strong · 18 | Strong · 18 | Strong · 18 |
| Canvas Construction Inc | Strong · 33 | Absent | Absent | Absent | Strong · 20 |
| Samsung Electronics Co. Ltd | Strong · 27 | Absent | Moderate · 10 | Strong · 14 | Absent |
| Ocado Innovation Ltd | Strong · 37 | Absent | Absent | Emerging · 6 | Emerging · 7 |
| Mitsubishi Electric Corp | Strong · 36 | Emerging · 6 | Absent | Absent | Absent |
| Mitsubishi Electric Research Laboratories Inc | Strong · 30 | Absent | Moderate · 9 | Absent | Absent |
Frequently asked questions
The analysis covers 2,323 patent families in scope. Filing counts for 2025 and 2026 are understated due to the 18–24 month publication lag inherent in the patent system and should be treated as lower bounds.
Dexterity Inc leads with 59 patent families, followed by Ocado Innovation Ltd (45), ABB (Schweiz) AG (40), Mitsubishi Electric Corp (36), and Mitsubishi Electric Research Laboratories Inc (30). Both Dexterity and Ocado are classified as new entrants whose portfolios are predominantly recent filings.
The top five filers account for 18% of the combined output of the hundred largest filers, which is a low concentration figure. The narrow gap between first and fifth place — just 29 patent families — means the ranking is susceptible to rapid reordering as new entrants scale their filing programs.
The United States is the leading filing jurisdiction, followed by China and India. WIPO PCT and the European Patent Office are the next largest destinations, indicating meaningful international protection activity. South Korea, Canada, Australia, and the United Kingdom form a smaller third tier, while Japan files at lower volume relative to its historical robotics prominence.
B25J (Manipulators & robots) is the overwhelmingly dominant IPC class. G06N (Computing based on AI models) is the second-largest branch, confirming AI/ML methods as a distinct and substantial IP layer. G06T (Image data processing), G05B (Control & regulating systems), and A61B (Diagnosis & surgery) follow as the leading application vectors.
South China University of Technology is the most active co-filer, with joint families recorded alongside Guangdong Zhike Intelligent Technology Co. (5 families), Foshan Huashu Robot Co. (1 family), and the Zhongshan Modern Industrial Technology Research Institute (1 family). Ocado Innovation Ltd co-filed with Kindred Systems Inc across two pairings totaling 6 families. Overall collaboration density remains low, suggesting the ecosystem has not yet consolidated around anchor partnerships.
Ready to map your own industrial robot AI/ML 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.