Is there a way to access the DMP?

[masasin]

I cannot find it in the docs.

[turbinenreiter]

Sadly no. There is no real documentation on how to use the DMP in the datasheet, so I couldn't implement it.

data sheet and register map
http://dlnmh9ip6v2uc.cloudfront.net/datasheets/Sensors/IMU/PS-MPU-9150A.pdf
http://dlnmh9ip6v2uc.cloudfront.net/datasheets/Sensors/IMU/RM-MPU-9150A-00.pdf

[masasin]

Yep, those are the docs I was referring to.

I have seen some implementations with C and C++, but I can't really figure
out their structure. Would you like to see them?

On Mon, Mar 16, 2015 at 8:31 PM Sebastian Plamauer notifications@github.com
wrote:

Sadly no. There is no real documentation on how to use the DMP in the
datasheet, so I couldn't implement it.

data sheet and register map
http://dlnmh9ip6v2uc.cloudfront.net/datasheets/Sensors/IMU/PS-MPU-9150A.pdf

http://dlnmh9ip6v2uc.cloudfront.net/datasheets/Sensors/IMU/RM-MPU-9150A-00.pdf

—
Reply to this email directly or view it on GitHub
#6 (comment)
.

[turbinenreiter]

Sure, but I can't promise anything. Spent a lot of time searching around, but never found anything good.

[masasin]

People from my lab got this working on mbed:
http://developer.mbed.org/users/syundo0730/code/MPU6050-DMP/

It is written in C++, and based on code written by Jeff Rowberg for the
MPU-6050.

On Mon, Mar 16, 2015 at 8:57 PM Sebastian Plamauer notifications@github.com
wrote:

Sure, but I can't promise anything. Spent a lot of time searching around,
but never found anything good.

—
Reply to this email directly or view it on GitHub
#6 (comment)
.

[peterhinch]

It's a can of worms as far as I can see. Read the comments section of the link below - particularly the most recent one by @SVFeingold - to summarise the chip can only do 6-axis sensor fusion, and even this is slow and uses a proprietary encrypted binary. The job is probably best done in Python in the driver. I'm sure @turbinenreiter would look favourably on such a contribution from anyone with a grasp of quaternions.
Sparkfun breakout board

[blmorris]

I have used the RTIMULib software discussed further down in the comments in the link from @peterhinch
My setup used the micropython pyboard to stream the sensor data from the mpu9150 to a host PC via USB; I then used the RTHostIMUGL (one of the programs provided at https://github.com/richards-tech/RTIMULib) to compute the sensor fusion and display the results.
Unfortunately when I tried to get it running a couple of weeks ago I found that I didn't really document where I left off; it will take me a few hours to resurrect the system when I get a chance.
The RTIMULib software does implement the quaternion sensor fusion algorithm in C++; it could potentially provide a model for a Python implementation, or perhaps even be ported as an optional C module to MicroPython.
I should add that RTIMULib is well supported and under active development; Richard Barnett was very responsive when I had questions about his code a few months ago.
I'd love to get back to working on this, but not sure when I will be able to squeeze in the time.

[peterhinch]

Much as this interests me I've too many things on to address it now. But the RTIMULib looks good and has Python bindings. I'd try cross compiling it: if it compiles OK MicroPython might interface to it without much trouble. It strikes me as a better solution than the code cited by @masasin with its opaque binary blob in MPU6050_6Axis_MotionApps20.h

[masasin]

In the end we used RTIMULib as well, and just communicated with the chip
via i2c.

On 2015年5月6日(水) 17:23 Peter Hinch notifications@github.com wrote:

Much as this interests me I've too many things on to address it now. But
the RTIMULib looks good and has Python bindings. I'd try cross compiling
it: if it compiles OK MicroPython might interface to it without much
trouble. It strikes me as a better solution than the code cited by
@masasin https://github.com/masasin with its opaque binary blob in
MPU6050_6Axis_MotionApps20.h

—
Reply to this email directly or view it on GitHub
#6 (comment)
.