I'm trying to use a accelerometer/gyroscope sensor to figure out the displacement of a device I'm making. The device will only be moved vertically (up/down). Given that information, is there a better way to double integrate the sensor values? I've heard that getting the displacement using acceleration can be wildly inaccurate. Any help would be greatly appreciated.
You have heard correctly. This page explains just one of the major reasons why the double integration approach doesn't work with consumer grade IMUs.
I do understand that, but could you provide more information on how to make it as accurate as possible? The device will be moving up and down at most 5-6cm, so it won't be moving a whole lot. Will using this sensor be easier? It says it provides the acceleration vector without the gravity. Once I get the acceleration data, how do I make sure the measurements are being made at a consistent rate? Thank you.
The double integration question is posted about once or twice a week on the forum, and to my knowledge, no one has ever returned to report success. Good luck with your project.
Alright. Thanks for your advice. Would using a potentiometer or an encoder for this purpose be a sensible idea?
Consider attaching calipers to the device and read the digital output of the calipers to get the movement.
There are plenty of noncontact optical distance sensors to choose from.
Alright. Thanks everyone for the advice!
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.