Mapless Vehicle Localization Patents: Who Leads, Where Gaps Are 2026
Top-5 share is the combined record count of the five largest assignees divided by all 11 records in scope (CR5), not by the ranked leaders only.
What this landscape covers
This review tracks patent filings that describe mapless or map-free vehicle localization methods paired with autonomous or self-driving vehicle contexts — approaches that place a vehicle in space using onboard sensing and inertial data rather than a pre-built high-definition map. The scope spans visual-inertial localization, online vehicle localization, and related techniques where the localization pipeline is designed to function without a stored map layer.
The dataset in scope is small — 11 published records — which makes this a concentrated, early-stage corner of autonomous driving IP rather than a mature, crowded field. That size matters for how the figures below should be read: a single filer's activity can swing the yearly trend, and the technology composition reflects a handful of applicants' claim choices rather than an industry-wide consensus.
Filing trend and technology composition
Publication lags filing by roughly 18 months, so the most recent year in the trend below is understated — treat the tail end as a floor, not a ceiling.
Filing activity, 2017–2026
Filings across the period are sparse and uneven, rising to a peak of three records in 2022 before falling back. With fewer than four complete years of stable data once publication lag is accounted for, no growth rate can be reliably stated from this trend.
IPC subclass composition
Positioning (G01S) and image processing (G06T) each appear in 63.6% of the 11 records in scope, and control of non-electric variables (G05D) appears in 54.5% — together they describe a field built on sensor fusion and downstream vehicle control. Navigation/gyroscope classing (G01C) and AI-model classing (G06N) each sit at only 18.2%, and data-recognition, digital-processing and image-recognition classes each cover a single record at 9.1%. Because records can carry multiple classes, these shares add up to more than 100% of the 11 records and should not be summed.
Shares are the percentage of the 11 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Autonomous Driving — Mapless Vehicle Localization Patent Landscape with Eureka
This page is one run against one query. Ask Eureka your own question about autonomous driving — mapless vehicle localization patent landscape and every answer comes back with the patent numbers behind it.
Try EurekaMost-cited records and a representative filing
Data processing method, data processing apparatus, electronic device and storage medium
The filing describes a computing device that inputs a reference image and a captured image into a feature extraction model, derives a set of reference descriptors, builds multiple sets of training descriptors, and predicts a vehicle pose by feeding training poses and similarity scores into a pose prediction model, with both models trained jointly.Filed by Beijing Baidu Netcom Science and Technology Co., Ltd., published 2022-05-26.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20200333466A1 | Ground intensity lidar localizer | 54 |
| 2 | US20220164603A1 | Data processing method, data processing apparatus, electronic device and storage medium | 13 |
| 3 | CN115014346A | 一种面向视觉惯性定位的基于地图的一致高效滤波算法 | 4 |
| 4 | US20230071784A1 | Ground intensity lidar localizer | 2 |
| 5 | US11852729B2 | Ground intensity LIDAR localizer | 1 |
| 6 | US11493635B2 | Ground intensity LIDAR localizer | 1 |
Citation counts favour older records inside this searched corpus; treat them as a signal of influence within the field, not as a measure of current technical 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 (6 of 6 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 a filing decision
With only 11 records in scope, this field rewards a close read of who filed what rather than broad trend-fitting.
The entire ranked field sits with six companies
The top five assignees already account for 90.9% of the 11 records in scope, and the sixth closes the set to 100%. There is no long tail of single-filing entrants here — whoever is not among these six has not yet filed a matching record.
Volume peaked early and has not sustained
Filing rose to three records in the peak year and has not repeated that level since, with the most recent year showing no new filings on record. Given the roughly 18-month publication lag, the last one to two years understate true filing activity.
Sensor fusion and image processing anchor the claim space
Radar/positioning (G01S) and image data processing (G06T) each appear on 7 of the 11 records, with control classing (G05D) close behind at 54.5%. AI-model classing (G06N) covers only 18.2%, suggesting learned-model claims are a smaller share of the drafted claim space than sensor and control claims.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to autonomous driving — mapless vehicle localization patent landscape, with the prior art for and against each one.
Where to take this analysis
The dataset points to a concentrated field with specific technical gaps rather than a broad competitive landscape.
Map the leader's claim boundaries
With one assignee holding four of the 11 records, understanding exactly what its granted claims cover is the fastest way to see what remains open for a new filing.
Explore claim scope in EurekaCheck the under-claimed branches
Learned pose-prediction and gyroscope-fused approaches show thin IPC coverage against the core positioning and imaging classes, which may indicate open claim space rather than a solved problem.
Run a white space search in EurekaTrack the next filing wave
Given the 18-month publication lag, filings from the last two years are likely undercounted; revisiting this trend in the next data cut will show whether 2022's peak was a one-off.
Set a monitoring alert in EurekaCommon questions on mapless vehicle localization patents
The ranked field consists of six assignees, which together account for all 11 records in scope for this dataset. The leading assignee holds four records, and the count falls off quickly after that, down to a single record by fifth place. This is not a top-50 or top-100 ranking — it is the complete list the data endpoint returns for this search, so there is no long tail of additional filers hidden outside it.
The trend data does not support a stated growth rate. Filing activity rose to a peak of three records in 2022 but has not sustained that level, and there are fewer than four complete years of stable filing data once the roughly 18-month publication lag is accounted for. The most recent year should be read as understated rather than as a genuine slowdown.
Radar and positioning methods (IPC class G01S) and image data processing (G06T) each appear on 63.6% of the 11 records in scope, with control-of-non-electric-variables claims (G05D) close behind at 54.5%. AI-model classing (G06N) and navigation/gyroscope classing (G01C) are each present on only 18.2% of records, which points to sensor fusion and vehicle control as the dominant claim strategy rather than learned-model architectures.
The thinnest IPC coverage sits in AI-model classing, gyroscope-fused dead reckoning, and multi-sensor descriptor fusion for pose estimation — each covering a small share of the 11 records against the dominant positioning and imaging classes. A first claim in these areas would need to specify a concrete sensor combination or model architecture rather than a general localization method, since the core positioning and imaging claim space is already occupied by the ranked leaders.
Ten of the 11 records in scope were filed through the United States receiving office, with only one routed through China. Any freedom-to-operate review for mapless vehicle localization technology should therefore prioritise US prosecution history and granted claims, since that is where almost all of the documented activity sits.
Research Autonomous Driving — Mapless Vehicle Localization Patent Landscape in depth with Eureka
Go past this page: query the whole autonomous driving — mapless vehicle localization 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.