0

I want to determine how far a phone has been displaced. However, I want to determine this information to an semi accurate scale (couple of centimetres). I've already toyed around with the idea of GPS, however it's clear that GPS is not quite accurate enough to satisfy my thirst to determine a phone's displacement. The application: Build a mobile (obviously) app that can track how big an object is - essentially a phone ruler.

I have also looked into many pre-existing things and struggles with this problem, so I would like to introduce my idea and would like to know what you guys think about it, and how it can be implemented.

Some of the things that I have researched: 1) Taking the integral of accelerometer data - problem: Accelerometer drift (https://dsp.stackexchange.com/questions/34463/removing-drift-from-integration-of-accelerometer-data). 2) Taking the integral of accelerometer data + using gyroscopic data -> problem: still dealing with the drifting problem. 3) (Already mentioned) GPS - too inaccurate

What I am thinking: I was wondering if it possible to combine the second method that I have described along with AR (Kalman Filter from OpenCV (open to you guys' ideas aswell)) to determine how far the phone has travelled. I realize that this would require that the phone be able to see stationary objects (not just black screen), so let's assume in this its not laying flat on a table.

What I am asking you guys:

  1. Is this an achievable goal -> from the other forums that I've read, just taking method #2 that I described it's essentially pointless
  2. How would I go about implementing this Any help is seriously appreciated.

Relevant Links: Android sensor data displacement https://www.hindawi.com/journals/mpe/2019/3470604/ Android accelerometer accuracy (Inertial navigation)

Christoph Rackwitz
  • 11,317
  • 4
  • 27
  • 36
Frankvp11
  • 1
  • 1
    Welcome to Stack Overflow. Please read the information guides in the **help center** (https://stackoverflow.com/help), in particular, "How to create a Minimal, Reproducible Example" (https://stackoverflow.com/help/minimal-reproducible-example). – fmw42 Aug 18 '23 at 20:27
  • this isn't really something where you'd have a minimal, reproducible example. Its mostly asking for opinions and advice... – Frankvp11 Aug 18 '23 at 20:29
  • I am not sure advice is the objective of this forum. – fmw42 Aug 18 '23 at 20:35
  • Why can't it be? It's a perfectly good question - it has lots of room to be answered. For example, the first question - is it worthwhile to try... second question... what might be some noteworthy implementation details.... Perhaps, has this been done before.. What algorithms exist to fuse this data.. .etc – Frankvp11 Aug 18 '23 at 20:43
  • 1
    I did not mark your question down. I was just warning you that others may tell you it is off-topic possibly seen as from the help section as "Questions asking us to recommend or find a book, tool, software library, tutorial or other off-site resource are off-topic for Stack Overflow as they tend to attract opinionated answers and spam. Instead, describe the problem and what has been done so far to solve it." – fmw42 Aug 18 '23 at 21:11

0 Answers0