Patent Publication Analysis - autonomous parking-related patent documents, 2010-2018

By Hindsights Analytics | Contact us | Main site Last Updated: 2019-08-26

Introduction & Table of Contents

We present the below report of the patent and technical landscape in context of the proposed concept of: 'automatic car parking for car sharing service to park and retrieve cars to its fleet'.

Broadly, the analysis looks at a data set of 441 patent documents published between 2010 and 2018 to summarise recent developments in autonomous parking-related technologies.

Where applicable, we look more specifically at the proposed concept of: 'automatic car parking for car sharing service to park and retrieve cars to its fleet', including identifying a subset of data which may be considered more closely related to the concept.

With this goal and background in mind, the report includes the following sections:

Data overview

Patenting / R&D trends over time

The "publication trends" chart below plots publication numbers across time for the entire data set of 441 patent documents.

Patent activity over time is generally considered to correspond to level of R&D activity or expenditure, and therefore commercial interest in a particular field. As a result, any increases or decreases in the chart below indicates not only a rate of innovation over time, but also general trends in commercial interest by those involved in the patenting activity.


This figure shows an unmistakable trend of increases in the number of publications, firstly around 2014 and then more dramatically around 2017. While some increases may be expected due to a general increase in patenting rates, this type of significant increase is a strong signal of increased (and still increasing) R&D activity and commercial interest in this field.

Back to top

Key patent owners

Much as a review of levels of patenting activities can reveal general levels of commercial interest in the technical field, a review of the entities involved in the patenting activities can reveal much information about how each entity perceives the field in question.

By overall portfolio size

An owner entity's overall portfolio represents a sum of their R&D outputs over a period which they considered worth seeking patent protection for. Thus, a review of the patent owners and size of their portfolios reveal entities with significant historical interests and corresponding investments in a particular technical area. Here, key patent owner entities are shown below, by plotting the number of patent documents associated with each significant entity in the entire data set.

These figures indicate sizes of each owner entity's overall portfolios in this particular subject matter area. In other words, the graph indicates a total sum of outputs of their R&D activities over the period between 2010 and 2018 in this area.
** Applications by individual inventors is collected and listed below under a "[blank]" entity name.


The owner entities listed here include expected automotive giants such as Ford, GM and Volkwagen. Interestingly, Bosch appears to own the largest portion of the patent documents discovered in this search, indicating their focussed R&D efforts in this area, and technology companies such as IBM and Here are also found in this list.

Notably an entity called "CloudParc" shows up prominently in this report, which appears to be a New York-based startup, focussing on parking systems.

By annual filing rates

The below chart also presents portfolio sizes for key owners; however, in this chart the patent publication data is broken down across time, as to present rates of patenting activity.

So, this chart aims to reveal activity levels across time for each entity, revealing waxing and waning of each entity's level of interest.

Each bar in the grouped bar graph shows the number of publications by period. Accordingly, we look to show through this chart two key insights, which are:
1) trends in activity levels of each owner over time, and
2) relative patent activity levels for each company against each other over time.


As the area is so new, most of the entities' publications have been in the past two years or so. Bosch, while having the largest overall portfolio here, only appears to really have ramped up their efforts in the last couple of years, possibly indicating a relatively recent change in strategic initiatives.

Back to top

Patent landscape - subject matter clusters overview

As further detailed in the below appendix, we categorise the data set into a hierarchical set of 'clusters', or groups, according to their subject matter using our in-house algorithms. By doing so, we provide an efficient, unbiased and effective breakdown of data to help in understanding underlying trends in various niche areas within the particular field of interest.

In this section, we present outputs of our analysis of the very top-level subject matter categories. Further details for each category are available by scrolling down, or clicking through the relevant links (e.g. in the word cloud).

Top-level categories - topics

7 high level categories were identified in the data set. Each category is represented below by a word cloud, where key concepts of each cluster are shown, and the size of each concept indicates its significance in the category.

Category No. Title Word Cloud
1 Systems/payments
2 Navigation
3 Guidance
4 Surrounding structures
5 Maneouvering
6 Mechanical
7 Communication/electronics

Top-level categories - trends

The chart below shows the publication trends for each top-level category, further split into grouped bars by their relevant period. This chart indicates relative sizes and trends of R&D activity and innovation in each categories of technologies at a high level.

By comparing and contrasting the data in each category, and between categories, insights can be quickly gained regarding trends across time and subject matters.


Again, most of the increases appear to have occurred in the last few years. As for categories 6 and 7, their numbers are relatively steady, which is consistent with the fact that they mostly appear to relate to filings that are only peripherally related to autonomous driving or parking.

The image below show cumulative sums of patent documents in each category. So, this chart shows relative growth over time of each category, where acceleration/deceleration may have occurred in each category as interest waxed and waned, as well as their relative total sizes at each point in time.

To further investigate each category, such as to review their constituent/typical patent documents, leading assignees, filing trends over time, or subject matters of their sub-categories, please scroll down to the 'cluster details' section, or click here.

Back to top

Patent landscape - cluster details

In this section, we delve into each subject matter cluster in detail, including for each category:

  1. The key topics captured in the category (as a word cloud),
  2. the overall filing trends over time, indicating interest and R&D activity/innovation in the field
  3. key assignees present in the category, and their own filing trends, and
  4. subcategories including their sizes, trends and subject matter.

To drill down further into each category, please click through the word cloud, or the "details" link to explore further details of each category. These pages are designed to allow for further detailed investigations of specific data groups. For further backgrounds on our categorisation algorthms, please see the appendix below.

Notably, it is interesting to find that often, key owner entities for each category do not correspond to the key owners for the overall data set. By reviewing each entity's share of the overall data set as opposed to its share of a particular category, it is possible to gain an understanding of the particular category to the entity's portfolio, and thus an insights into its overall strategy.

Category 1 details

  • Category size: 126 documents; rank 1 out of 8 (1 being the largest)
  • Top owners: [blank]; International Business Machines Corp; Cloudparc Inc

Category 2 details

  • Category size: 26 documents; rank 6 out of 8 (1 being the largest)
  • Top owners: Here Global BV; Ford Global Technologies LLC; [blank]

Category 3 details

  • Category size: 79 documents; rank 3 out of 8 (1 being the largest)
  • Top owners: [blank]; Ford Global Technologies LLC; Lyft Inc

Category 4 details

  • Category size: 103 documents; rank 2 out of 8 (1 being the largest)
  • Top owners: Robert Bosch GmbH; Ford Global Technologies LLC; Dura Operating LLC

Category 5 details

  • Category size: 50 documents; rank 4 out of 8 (1 being the largest)
  • Top owners: Valeo Schalter und Sensoren GmbH; Ford Global Technologies LLC; LG Electronics Inc

Category 6 details

  • Category size: 25 documents; rank 7 out of 8 (1 being the largest)
  • Top owners: Ford Global Technologies LLC; ZF Friedrichshafen AG; Robert Bosch GmbH

Category 7 details

  • Category size: 32 documents; rank 5 out of 8 (1 being the largest)
  • Top owners: [blank]; Borealis Technical Ltd; UNITRONICS AUTOMATED SOLUTIONS Ltd

Back to top

Related patent documents

This section of our report analyses a particular sub-section of the data. More specifically, here we analyse patent documents identified by our algorithms as being potentially more related to the proposed concept ("automatic car parking for car sharing service to park and retrieve cars to its fleet") than the remainder of the data set.

By showing the publication trends, key owners and sample patent documents for the "related" data set, we aim to assist in understanding this particular segment; but also how it compares to the analysed field (entire dataset) in general.

The data set here is collected by our proprietary "relevance" analysis to the proposed conept. Accordingly, we note that the data set in this section here may originate from multiple "categories" above.

Related patent documents - overview

Our algorithms identified 32 patent documents out of 441 as being related to the proposed concept. Some of the key patent owners identified in this data set include:

  • Robert Bosch GmbH: 14 patent documents
  • GM Global Technology Operations LLC: 3 patent documents
  • [blank]: 2 patent documents

Examples of related patent documents

The below list shows examples of the patent documents potentially related to the proposed concept. We also list our "relevance factor", which is an approximate indication of each patent document's relevance to the proposed concept.

Year Patent Publication Assignee Relevance factor
2018 US20180004218 METHOD FOR OPERATING A VEHICLE Robert Bosch GmbH 0.79
2017 US20170329346 VEHICLE AUTONOMOUS PARKING SYSTEM Magna Electronics Inc 0.75
2015 US20150346727 Parking Autonomous Vehicles [blank] 0.74
2018 US20180335777 Autonomous Vehicle Parking System Meter Feeder Inc 0.73
2018 US20180336427 Vehicle Parking Enforcement System Meter Feeder Inc 0.73

Publication trends - related patent documents

This "publication trends" chart below shows publication rates, only for the patent documents found to be related to the proposed concept. Accordingly, we use this chart to show levels of R&D activity and rate of innovation over time for those patent documents identified to be related to the proposed concept.

Related patent documents - comments

Interestingly, although the same size is relatively small, the majority of patent publications have occurred in this area starting in 2016, with a significant increase in 2018. This indicates that in comparison to autonomous parking in general, this area may be an even more of a "hot topic".

Back to top

Appendices & notes

Patent data categorisation

Typically, categorisation of patent documents rely on official patent 'classifications', or manual catagorisations, which can be very time-consuming. Our AI-based analysis algorithms automatically analyse each document's subject matter, and group documents together by similarity to any level of detail, after which the key subject matters for the group of documents are displayed.

Illustration of hierachical categorisation
As shown in the figure above, the dataset is grouped at multiple 'levels', such that at each level, m groups are used to categorise the n (patent) documents in the dataset. As a result, each document belong to a category at each level.

Each category is identified by their 'level' and 'ID'. At the top level, the category with ID 2 is referred to as '1-2', and category ID 12 at the third level is referred to as '3-12'.

By reviewing these categories, you can quickly gain insights into a make-up of an individual portfolio, compare multiple portfolios. Or, by looking at trends of a category or sub-category, valuable insights into industry trends of company directions can be instantly gained.


With this demonstration, we aim to show the effectiveness of our algorithms, as well as the types of insights that can be quickly gained with the outputs.

* In determining 'applicants' associated with patent documents, we attempt to include patent applications filed under associated organisations or separate corporate entities within one group.

About us

Hindsights is a patent analytics specialist startup, focussed on extracting valuable, unique insights from patent data.

© hindsights 2019
Back to top