Measuring Displacements Using Accelerometers

[@garimasinghal5987]

Thank you Roboro for taking out time in explaining your work. The results are very helpful.

[@greatvbn]

Really excellent work.Even I have been trying to get the displacement using navigation grade accelerometer but I could not compromise on the accuracy and accumulated error .can you please share the code of computation displacement from acceleration

[@ientucamranh9270]

How to get position from imu?

[@infocentrousmajac]

Thank you Roboro for sharing the code. I am in the process of getting a set of AD accelerometers and I will try something similar. I'll let you know what I find since the accels are different from the ones you used. Cheers

[@kasiviswanadhsripada]

Hi, I'm not able to understand the averaging part to remove constants. can you explain me please?

[@Roboro]

I wrote about the different methods I tried on my blog, it also explains how I removed the constants roboroblog.wordpress.com/2016/08/01/measuring-displacement-using-accelerometers-part-1-calculating-gravity-vectors/

[@shamimahamed6939]

Thanks for sharing the code

[@aligasco7206]

reference height arduino I want to know my room floor level how much height from other room in which video solution

[@infocentrousmajac]

It is almost a year since you posted this video. Excellent work. I was wondering what is your experience with slow movements (let's say 1 inch every three seconds) ? I would expect more difficulty during the sampling and overall with accuracy for displacement measurement, do you agree? Also, Would it be possible to get the code to perform the double integration and plot the displacement graph?

[@Roboro]

Infocentro Usmajac Thanks for enjoying the video, and yes I would agree with it being more difficult to get an accurate displacement for a slow movement using the methods I am. you could try other methods for isolating the acceleration due to the movement such as filters if you don't want to use the gyroscope. And I will try to edit the code and post on my blog by maybe the end of next week, I'm very busy in work at the moment and won't get the chance this week. I will send you a message once it's up

[@infocentrousmajac]

Thank you for the reply. Just as a side note, I am looking at several measurement methods but the inertial part is still of interest to me. I'll be looking forward to hear from you. Thank you

[@jhonathanwpc9626]

Congratulations, very good job

[@shengkaizhang2369]

How did you get the displacement.? I use the Imu to calculate the trajectory,but the result was so bad. Do you know how to improve that? I have accel and gyro sensors.

[@Roboro]

Sorry, missed this comment. I ran into alot of problems with that as well. I write about how I went about it on my blog but you could also look into sensor fusion roboroblog.wordpress.com/2016/08/01/measuring-displacement-using-accelerometers-part-1-calculating-gravity-vectors/

[@shengkaizhang2369]

@@Roboro thanks ~~I will read that careful. and discuss it with you later

[@greatvbn]

Can you please share the code for displacement calculation,hoping to hear from you soon.

[@Roboro]

Hi Jyothi, the code can be found on my github github.com/RonanB96/Displacement-from-accelerometer A description of how I made it can be found on my blog roboroblog.wordpress.com/2016/08/01/measuring-displacement-using-accelerometers-part-1-calculating-gravity-vectors/

[@avsingh9250]

@Roboro I am using your github code but I am getting an Error "fatal error: Energia.h: No such file or directory" this type please help me

[@vedantkandoi7537]

I need to continuously measure distance(every few milliseconds) traveled and display it(on a lcd panel maybe, no need to store it). How can i do that?

[@Roboro]

Hey Kandoi, You can just double integrate the acceleration over a short period (apply any filtering if needed) and add the latest displacement to the old value and display this on the lcd

[@chiky1180070177]

Nice work

[@bholster8073]

Thank you for sharing ...i have a problem with tracking slow movement followed by rapid movement slow is 0 .1m/s and rapid is 5 m/s. Seems like these devices are good to do that.

[@Roboro]

Yeah this device should be able to do that, you can see how I made it and the code if you follow the link in the description. There is room for improvement both on the code and signal processing side though

[@mohmmedalmosawi8780]

hi thank you how to get on parts

[@Roboro]

Eng mohammed Ali Almosawi you can find the circuit diagram on my blog roboroblog.wordpress.com/2016/08/17/measuring-displacements-using-accelerometers-part-3-testing-and-video/ I would've ordered the parts from eBay/AliExpress/Farnell. The parts shouldn't be hard to find in any electronics shop

[@mohmmedalmosawi8780]

Dear Roboro, Can you help me to get of names of device part or commercial name and help me connect together parts. Best regards. Eng. M.Ali

[@abdulghoni2739]

thanks a lot

[@stanislavhristakiev1881]

This is fake! Were is library need to run code? Where class Timer_Handle? Where Error.h>BIOS.h> anf a.e....

[@zw7453]

In his post, he said that he subtracted the average of all data that is in the user selected range from each data in the data set, hoping to cancel the sensor error offset. However, this method only works if you return the whatever object you are moving to its original starting point, which makes this algorithm almost useless. WASTE OF TIME! I'm actually worried about the research group you were working for since this brainless design may cause some catastrophic error in the future.

[@zw7453]

Instead, you could let the sensor sit for a couple of seconds to collect its steady error. Then subtract that from future data