Indoor navigation with machine learning

GPS lets you determine the position of an object down to single-centimeter accuracy – as long as the object is outside. If the object is inside, the task is a bit more complicated. Satellite navigation doesn't work well through doors and rooftops, and even if you could replace the satellite signal with equivalent transmissions from locally placed beacons or WLAN access points, the presence of interior walls and furniture muddles up the results of classical analytical techniques such as those used with GPS. What is more, when someone is inside a building, the question is not so much about "What are his coordinates." What you really want to know is "What room is he in?" Such a problem is better addressed through the tools of machine learning.

Of course, creating a complete machine learning solution to find someone in a small house might seem like overkill, but this article is intended as an exercise to show these machine learning tools and techniques in a simple situation – a kind of machine learning "Hello, World" application. One could imagine scenarios where these techniques could find broader utility, such as tracking down an executive in a large office complex or even finding a lost set of car keys.

In this example, Tom, the protagonist, has lost his way. Fortunately, his smartphone shows the signal strength of seven hotspots in his vicinity (Figure 1). Because Tom often gets lost, I have mapped the four rooms as a precaution (the blue crosses in Figure 2), and I have a machine learning dataset I can use to train a program to find Tom.

[...]

Use Express-Checkout link below to read the full article (PDF).

Buy this article as PDF

Express-Checkout as PDF
Price $2.95
(incl. VAT)

Buy Linux Magazine

SINGLE ISSUES
Print Issues
Digital Issues
 
SUBSCRIPTIONS
Print Subs
Digisubs
 
TABLET & SMARTPHONE APPS
Get it on Google Play

US / Canada

Get it on Google Play

UK / Australia

Related content

  • Data Science Methods

    Data science is all about gaining insights from mountains of data. We tour some important tools for the trade.

    more »
  • Perl – k-means Clusters

    A human observer can register clusters in a two-dimensional set of points at a glance. Artificial intelligence has a harder time getting it done; however, the relatively simple k-means method delivers usable results.

    more »
  • KNIME

    They say data is "the new oil," but all that data you collect is only valuable if it leads to new insights. An open source analysis tool called KNIME lets you analyze data through graphical workflows – without the need for programming or complex spreadsheet manipulation.

    more »
  • Unsupervised Learning

    The most tedious part of supervised machine learning is providing sufficient supervision. However, if the samples come from a restricted sample space, unsupervised learning might be fine for the task.

    more »
  • Treasure Hunt

    A geolocation guessing game based on the popular Wordle evaluates a player's guesses based on the distance from and direction to the target location. Mike Schilli turns this concept into a desktop game in Go using the photos from his private collection.

    more »
comments powered by Disqus
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Subscribe to our ADMIN Newsletters

Support Our Work

Linux Magazine content is made possible with support from readers like you. Please consider contributing when you’ve found an article to be beneficial.

Learn More

News

Tag Cloud

Administration Community Events Hardware Linux Linux Pro Magazine Mobile Programming Security Software Ubuntu Web Development Windows free software open source