Harvesting Robot AI/ML Patent Snapshot 2026
The harvesting robot AI/ML patent space is a nascent, highly fragmented field of 9 patent families, led by academic institutions with Yamagata University holding the top position. Annual volume peaked around 2020–2021 and has eased since, signalling a field that has not yet attracted sustained commercial R&D investment.
Academic institutions lead a highly fragmented, early-stage field
Yamagata University holds the top position with 2 patent records, ahead of a group of eleven applicants each holding a single record — confirming that no single commercial player has established a visible position in harvesting robot AI/ML.
The top five filers account for 46% of the combined total of the ranked applicants visible in this query, a moderate concentration figure that, in a corpus of only 9 patent families, reflects the outsized weight of each individual filing rather than entrenched market leadership.
| # | Applicant | Patent records | Share |
|---|---|---|---|
| 1 | Yamagata University | 2 | |
| 2 | Nantong Institute of Technology | 1 | |
| 3 | Dr. A. Maria Jackson | 1 | |
| 4 | Chongqing University OF POSTS & TELECOMM | 1 | |
| 5 | Bengbu College | 1 | |
| 6 | R. S. Jasuvant | 1 |
| # | Applicant | Patent records | Share |
|---|---|---|---|
| 7 | P. G. Shaju | 1 | |
| 8 | CENTRE FOR INNOVATION FABRICATION ENGINEERING & RE… | 1 | |
| 9 | Dr. M. Umar | 1 | |
| 10 | Chongqing University of Technology | 1 | |
| 11 | Shandong Normal University | 1 | |
| 12 | Chongqing University | 1 |
The leading positions being held by universities — Yamagata University, Nantong Institute of Technology, Chongqing University of Posts and Telecommunications, Bengbu College — suggests the technology remains largely in the research phase, with limited evidence of deep industrial commercialisation to date.
The most recent filing years are subject to publication lag and may not yet fully reflect activity; the corpus should be interpreted as a minimum count for recent periods. Longer-window growth, applicant concentration, and technology-route coverage are therefore more reliable signals than the latest-year bar alone.
Filing activity peaked in 2020–2021; AI perception dominates the technology mix
The annual filing trend reveals a short burst of activity between 2019 and 2022, with the technology mix weighted toward AI inference and vision systems rather than mechanical harvesting hardware.
Annual filing trend
Filings first appeared in 2019, reached a modest peak across 2020 and 2021 (two records each year), then fell back in 2022 and recorded no filings in 2023 or 2024 in the current corpus. The 2025 figure of three records should be treated with caution given publication lag — it may partially reflect late-publishing applications rather than a confirmed resurgence.
↗ Hover for values · click a bar to ask EurekaTechnology composition
G06N (Computing based on AI models) is the most represented branch, followed closely by B25J (Manipulators and robots), G06K (Data recognition and presentation), and G06V (Image/video recognition). A01D (Harvesting and mowing) — the core agricultural application class — ranks fifth, indicating that patent claims in this space lean more toward the AI/perception stack than toward the physical harvesting mechanism itself.
↗ 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.
Fruit harvesting robot with hybrid vision weight r…
The present invention reveals an independent pomegranate harvesting robot combining multispectral computer vision (RGB+N1R imagery), real-time weight sensing, and reinforcement learning-based robotic control to realize historically unprecedented precision and delicacy of fruit picking. The system includes: (1) a bespoke vision module (101) with a quantized… (excerpt from the patent abstract)
Open this patent in Eureka →| # | Patent | Citations |
|---|---|---|
| 1 | 基于CNN的果实与障碍物的同步识别方法、系统与机器人 | 39 |
| 2 | 一种非结构化环境下基于深度学习的果实采摘机器人目标检测方法 | 24 |
| 3 | 一种杆状作物收割机器人控制方法 | 4 |
| 4 | 基于边缘细节的果实分割方法及系统 | 2 |
| 5 | Harvesting robot, control method and control progr… | 2 |
| 6 | 基于物联网的蔬果采摘机器人采摘目标建模方法和系统 | 1 |
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.
Assignee snapshot from the current evidence set
The applicants below are visible in this query result. Because the evidence set is relatively small, read this section as a directional snapshot rather than a full competitive ranking.
Yamagata University
Yamagata University holds 2 patent records — the largest share in this corpus — with a technology focus spanning A01D (Harvesting and mowing), B25J (Manipulators and robots), and G05D (Control of non-electric variables). This hardware-centric profile distinguishes it from the perception-heavy Chinese university filers. No momentum trend data is available for this applicant in the current evidence.
patent records: 2Shandong Normal University
Shandong Normal University holds 1 patent record and is identified as a new entrant in the momentum data, with a technology focus on G06K (Data recognition and presentation), G06N (Computing based on AI models), and G06V (Image/video recognition). Its recent entry — classified as a new entrant — places it at the frontier of the current filing wave.
patent records: 1Frequently asked questions
The current corpus contains 9 patent families in scope. This is a small field, and the low count reflects both the nascent stage of the technology and the narrow intersection of agricultural robotics with AI/ML patent claims.
Yamagata University is the top applicant with 2 patent records, ahead of eleven other applicants each holding a single record. No commercial OEM appears in the current ranking.
China leads with 6 patent records, followed by Japan with 2 and India with 1. Major agri-tech markets such as the United States and Europe do not appear in the current corpus.
G06N (Computing based on AI models) is the most represented class, followed by B25J (Manipulators and robots), G06K (Data recognition and presentation), and G06V (Image/video recognition). A01D (Harvesting and mowing) ranks fifth, indicating that patent activity leans toward the AI/perception stack rather than mechanical harvesting.
The lifecycle is assessed as Decline. Annual filings peaked in 2020 and 2021 and have eased since. The 2025 figure should be treated cautiously due to publication lag. The field has not demonstrated the sustained multi-year growth seen in more established robotics or agricultural technology domains.
Two adjacent branches show relatively sparse coverage: G06Q (Business, commerce and admin data processing) with 2 patent records and G05D (Control of non-electric variables) with 1 patent record. G05D in particular — covering autonomous navigation and closed-loop control — is technically central to deployed harvesting robots yet is the least-claimed branch in the corpus, representing a potential differentiation area for new applicants.
Map your own harvesting robot AI/ML Snapshot
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.