MarkyMark2016
Posts: 3
Joined: Mon Oct 03, 2016 8:12 pm

Disappointing sense hat results

Mon Oct 03, 2016 11:30 pm

Hi all,

I'm hoping that someone might be able to help with issues I'm having with the sense hat (before I sent them back)

Bit of back ground - I build a single axis (yaw) gyro system some years back (high quality gyro with dedicated amtel MPU for sampling. Sample rate 100Hz, averaging @ 10Hz). Repeat-ability approx 0.5 degrees or less, drift 1 deg/min.Reason I liked the idea of the hat was because of the IMU and thus cancelling out the drift.

Results so far have been very disappointing. Just testing the gyro a 360 yaw turn will yield about a 3-7 degree variance
Gyro drift is close to 1 degree/sec

Using the IMU (i.e. compass, accelerometer and gyro) repeat ability is improved but the lag is high to get a steady result.

I've tried with the stock RTIMULin.ini file as well a a number of calibrated ones.

I also found this link - but to no avail: viewtopic.php?f=104&t=110594

Are my expectations for this device simply to high? The application is robotics. Perhaps I should look else where?

I have checked all the common suspects (power etc). This is running jessie on a RPi 2

Thanks in advance for any advice

Mark

ShiftPlusOne
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 6043
Joined: Fri Jul 29, 2011 5:36 pm
Location: The unfashionable end of the western spiral arm of the Galaxy

Re: Disappointing sense hat results

Tue Oct 04, 2016 7:54 am

Have you tried doing your own full calibration?

MarkyMark2016
Posts: 3
Joined: Mon Oct 03, 2016 8:12 pm

Re: Disappointing sense hat results

Tue Oct 04, 2016 9:17 am

Hi ShiftPlusOne,

Yes - perhaps my comment "as well a a number of calibrated ones." wasn't clear - these were ones I'd done myself

Mark

ShiftPlusOne
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 6043
Joined: Fri Jul 29, 2011 5:36 pm
Location: The unfashionable end of the western spiral arm of the Galaxy

Re: Disappointing sense hat results

Tue Oct 04, 2016 9:40 am

Ah, alright. Of course there will be tradeoff between latency and accuracy, but you may be able to play with the filter's parameters. It may very well be the case that it's just not suitable for your application.

Return to “Astro Pi”