Marine COLREGs Collision Avoidance Patents: Leaders & White Space 2026
- Filing peaked in 2018 at 10 records then fell away — the field's busiest year is already behind it, not ahead.
- One assignee holds 10 of the ranked filings against a field where the fifth-ranked holder has just 1, a steep drop rather than a gradual taper.
- G05D control claims cover 69.2% of the 13 records in scope while traffic-control classification (G08G) sits close behind at 61.5%, showing the two are usually claimed together.
What this landscape covers
This landscape tracks patent filings at the intersection of autonomous vessel navigation and compliance with the International Regulations for Preventing Collisions at Sea (COLREGs). The scope combines collision-avoidance methods, unmanned surface vessel control, and marine navigation systems, filtered to records that explicitly address COLREG compliance or autonomous-vessel avoidance logic. 13 published records sit in scope, spanning filings from 2015 through the 2026 cut-off.
Because publication typically lags filing by roughly 18 months, the most recent filing years in this set are undercounted — the 2026 figure of zero should be read as data still arriving, not as a drop in activity.
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
Filing trend and technology composition
Two views of the same 13 records: how filings moved year over year, and which control and navigation classes carry the claims.
Filing trend, 2017–2026
Activity in this scope is concentrated in a short window rather than spread evenly. Filing rose to a peak of 10 records in 2018 and has since thinned out, with 2026 showing no published records yet — consistent with the publication lag rather than a genuine stop in filing.
IPC subclass composition
G05D (control of non-electric variables) and G08G (traffic control systems) dominate, appearing in 69.2% and 61.5% of the 13 records respectively — most filings claim a control method alongside a traffic or collision-classification layer. G06N (AI-based computing) appears in 23.1% of records, and G01C (navigation and gyroscopes) in just 7.7%, marking it as the thinnest-claimed adjacent class in this set.
Shares are the percentage of the 13 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Marine Autonomy & Navigation — COLREGs-Compliant Collision Avoidance Patent Landscape with Eureka
This page is one run against one query. Ask Eureka your own question about marine autonomy & navigation — colregs-compliant collision avoidance patent landscape and every answer comes back with the patent numbers behind it.
Try EurekaMost-cited records in this landscape
Dynamic collision avoidance method for unmanned surface vessels based on trajectory replanning (WO2020253028A1)
The method builds a collision cone between an unmanned surface vessel and neighbouring vessels using onboard sensor data, applies a soft constraint to account for uncertainty in observing neighbouring vessel motion, and bounds the candidate velocity set by the vessel's own speed and heading limits. A cost function then selects the optimal avoidance velocity, which is simulated forward to produce replanned trajectory points. The output is framed explicitly as trajectory replanning constrained to satisfy the International Regulations for Preventing Collisions at Sea, while remaining compatible with the vessel's own manoeuvring control.Filed by South China University of Technology, published 2020-12-24.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | WO2019126755A1 | Generating and classifying training data for machine learning functions | 104 |
| 2 | US20210125502A1 | A collision avoidance method and system for marine vessels | 47 |
| 3 | WO2020253028A1 | 一种基于航迹重规划的水面无人艇动态避碰方法 | 33 |
| 4 | WO2019121237A1 | A collision avoidance method and system for marine vessels | 17 |
| 5 | US11915594B2 | Collision avoidance method and system for marine vessels | 4 |
| 6 | EP3729407B1 | A collision avoidance method and system for marine vessels | 3 |
| 7 | CN116430853A | 一种基于改进速度障碍策略的无人艇动态避障方法 | 1 |
Citation counts reward earlier filings within a searched corpus and should be read as a signal of influence, not of present-day 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. Publication numbers are shown where the record carries one (7 of 7 rows); 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 filing strategy
Three read-outs from the trend, the class composition and the citation table, translated into what they imply for anyone deciding where to file next.
The busiest filing window has already passed
Filing rose to 10 records in 2018 and has fallen away since, with no published 2026 record yet. Given the roughly 18-month lag between filing and publication, some of that recent thinning is an artefact of the data cut-off rather than a genuine slowdown — but the shape of the curve still shows this is not a field in an early filing rush.
Control and traffic-classification claims are usually paired
Nine of the 13 records carry a G05D control claim and eight carry a G08G traffic-control claim; the overlap suggests most filers are claiming the control loop and the collision-classification logic together rather than as separate filings. A standalone control-only or classification-only claim is the less-crowded position.
One holder dominates a short ranked list
The assignee ranking returned by this dataset covers five companies, with the leading holder at 10 ranked filings and the fifth-ranked holder at just 1. That is a steep drop-off rather than a gradual taper, and it means most of the ranked activity in this scope sits with a single filer.
The most-cited record sits outside the core collision-avoidance claim
The top-cited filing in this set addresses generating and classifying training data for machine learning functions rather than collision avoidance directly, followed by two related filings on collision avoidance methods for marine vessels. Citation counts favour older records in a searched corpus, so treat this as a marker of influence on later filings, not current commercial weight.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to marine autonomy & navigation — colregs-compliant collision avoidance patent landscape, with the prior art for and against each one.
Assignee landscape and collaboration patterns
The ranking returned by this dataset covers five companies. Rolls-Royce holds the largest share of ranked filings, South China University of Technology and Fugro NV follow, and two individual inventors, Andrew Martin Dickie and Chase John Gaudet, appear together across all three recorded co-assignee pairs.
Rolls-Royce leads the ranked list by a wide margin
At 10 ranked filings against a fifth-place holder at just 1, Rolls-Royce accounts for the bulk of the assignee ranking in this scope, making its published claims the first reference point for anyone assessing freedom to operate here.
South China University of Technology anchors the academic side
The university's trajectory-replanning filing, WO2020253028A1, is the representative record for this landscape and frames avoidance as an explicit COLREGs-constrained replanning problem rather than a generic control loop.
A small collaborative cluster sits outside the corporate leaders
Andrew Martin Dickie, Chase John Gaudet and Fugro NV appear together across all three recorded co-assignee pairs in this dataset, suggesting a distinct collaborative filing group separate from the larger corporate and academic holders.
| Assignee | Recent year | YoY |
|---|---|---|
| Rolls-Royce | 0 | — |
| South China University of Technology | 0 | — |
| MARTIN DICKIE ANDREW | 0 | — |
| GAUDET CHASE JOHN | 0 | — |
| FUGRO NV | 0 | — |
Where to take this next
The dataset points to a concentrated field with a dominant holder and a short ranked list — the next steps depend on whether the goal is freedom-to-operate or finding open claim space.
Map the leader's claim scope in detail
With one holder at 10 ranked filings, a clause-by-clause read of its collision-avoidance claims is the fastest way to know what a new filing would need to design around.
Explore assignee claims in EurekaTrack the under-claimed IPC branches
G01C-classed navigation and gyroscope filings are the thinnest branch in this set at 7.7% of records, worth watching as filing activity in the field resumes.
Run a white-space search in EurekaWatch the co-assignee cluster
The Martin Dickie, Gaudet and Fugro NV grouping is small but distinct from the corporate and academic leaders, and worth monitoring for follow-on filings.
Set an alert in EurekaFrequently asked questions
Within this dataset's ranked assignee list, Rolls-Royce holds the largest share of filings at 10 ranked records, well ahead of the other four ranked holders, the smallest of which holds just 1. South China University of Technology and Fugro NV also appear on the ranking. This is a five-company ranking drawn from a 13-record scope, not a top-50 or top-100 list, so it should be read as the whole picture the dataset returns rather than a sample of a larger field.
Filing in this scope peaked in 2018 with 10 published records, and has declined in the years since. The 2026 figure shows zero published records, but that almost certainly reflects the roughly 18-month lag between filing and publication rather than an actual stop in activity. Anyone using this trend to time a filing decision should treat the most recent one to two years as incomplete rather than final.
WO2020253028A1, filed by South China University of Technology, claims a trajectory-replanning method that builds a collision cone between an unmanned surface vessel and nearby vessels, applies a soft uncertainty constraint, bounds candidate velocities by the vessel's speed and heading limits, and selects an optimal avoidance velocity via a cost function. It frames its output explicitly as satisfying COLREGs constraints. Anyone building a similar trajectory-replanning avoidance loop with an uncertainty-weighted collision cone should review this filing's specific claim language closely, though it is one filing among several similar collision-avoidance method claims in this set, including two US filings on the same general problem.
The IPC composition shows G01C, covering distance, navigation and gyroscope-related classes, appearing in only 7.7% of the 13 records in scope, making it the thinnest represented branch. G06N, AI-based computing methods applied to collision decisions, sits at 23.1%, also relatively open compared to the 69.2% and 61.5% held by the two dominant control and traffic-classification classes. These thinner branches are reasonable starting points for a novelty search before drafting a new claim.
Exact concentration cannot be stated from this evidence because the assignee ranking (counted in records) and the total record count are not directly comparable units in this dataset. What can be said is that the ranking itself is steep: the leading holder has 10 ranked filings against a fifth-place holder with just 1, indicating most of the ranked activity sits with a small number of filers rather than being spread evenly across the five ranked assignees.
Research Marine Autonomy & Navigation — COLREGs-Compliant Collision Avoidance Patent Landscape in depth with Eureka
Go past this page: query the whole marine autonomy & navigation — colregs-compliant collision avoidance patent landscape 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.