IMU options, path forward?

Has there been any talk of looking at the current IMU options for sale, and how they relate to current AOG requirements?

My thoughts, non of the currently accepted v5 IMUs are available, anywhere, which means many like me are in a situation of either run v4, or run without an IMU.

I know the MMAs and the 55s arent as good as the higher end models, but are they worse then no IMU?

Would it be possible to write into the v5 code the ability to use the older IMUs? Allowing those of us who are later adopters to use the most up to date version with the equipment we can get our hands on?

If we could get some boards made, how many are you looking for?

Ino for Bno055 it’s available, but for the 055 better throw it out the window :wink: :wink:

IMU_BNO055.ino (7,5 KB)

I did get some pretty good results from AOGv3 and v4 with the BNO055, but always on very flat ground. Many here extoll the virtues of the 08x since v5 and perhaps this is because more people are using it on undulating ground, but the fact remains that a lot of users were getting reliable results with the 055 for a couple of years.

I am playing with a small vehicle (1m wheelbase) on pretty flat ground and I see no real performance difference between a well configured V4 / cheap BNO055 and a v5 / suddenly expensive 08x.

However, my tests have been crude and I have never measured accuracy, is the 08x much more precise? Is that the reason for the change?

Did you try v4 or v5 without any IMU on flat ground?

Has the FSM300 proven to be unacceptable? BNO080 on board, hillcrestlabs etc.
1100 in stock currently.

Sorry 1100 in stock Digikey currently.

Anyone tried the QMC5883L??

Not that I recall. Think I always used a BNO055.

The FSM300 has a slightly different format with halfholes, but other than that there is no reason why it would not work like the other BNO080 modules. It has the same specifications and connections, and it is designed by Ceva themselves. One user (DvD001) confirmed it worked, but had not done field testing.

I don’t know yet if it’s a good option, but I will soon receive a witmotion wt61p (a wt901 without magnetometer). I will test it, but price level it seems interesting, and especially available!

1 Like

Hi guys, i do not own a good imu atm to test myself. But can any one show me what the correct data from the imu Ago needs? In the imu-usb ino i see 127,0,0,0,0 why is that. The hmc5883L has adress 0x1E when i change the cmps 14 adress to 0x1E AGO works and i can connect. But it only show 150 degrees and while moving the imu its just stuck.

I need a short movie or picture from.serial reader so i can see what ago expects.

Hope you guys can help.

Example code here

Changing the address is just one step. your imu certainly has a communication protocol different from cpms or bno. you must refer to your imu datasheet to modify or write the corresponding code.

Thank you all, well after a hour copy paste coding. It works well on the simulator it responds very well. Heading and roll. This week wil test on machine. Will keep you posted on results.

What was the reason for going away from the BNO55? I have read many different reasons on different threads. I have implemented one into my wireless boards and it seemed to perform flawlessly on the bench but once I got it onto a shaking piece of machinery it would often give some unexpected readings every so often. Is this what other users experienced as well?

I had systems running both with cmps14 and was really great about it. But they stole my complete sets and now building new. But due no stock of cmp14 and bno085 i tried this to see results.

Like i say in simulator mode it works well. Next week i try i machine.

This is what i bought.

I think it could work with a good filtering algorithm to eliminate the noise. When I said noise, I meant jumps and false readings because it is a cheaper sensor.

If you need help with algorithms feel free to contact me.

Thank you, i will give shout when needed. But its pretty stable i must say. I will make recording when in MF.