Developer

ArduIMU V3 is finally here!

3689431212?profile=original

Long time since my last post!

I'm glad to announce the new ArduIMU V3 that will take the place of the very successful ArduIMU v2 (flat version). We called it flat version because ArduIMU V2 was created in a time where 3 axis and even 2 axis gyros were not so common and we had to place some of the gyros vertically. Good times... 

 

ArduIMU v2 was the first low cost IMU that had gyros that allowed me to keep it flat, but they were bulky and expensive. I had to use two gyro chips for the XYZ axis, another for the three axis acceleration and you had to place the magnetometer externally. Another "disadvantage" was that all the sensors (but the mag) were analog, by default limited to 10-bits and required all the analog inputs in order to read them and you were also limited by the default filtering configuration and other noisy issues. 

 

All the limitation we had with ArduIMU V2 are all gone with the V3. ArduIMU V3 features the new MPU-6000, that includes 3 axis gyros & accells built-in and process everything internally that is some kind of dual core power (24-bits ADC sampling and sensor fusion is handled inside the MPU), the latest 3 axis I2C magnetometer HMC-5883L and the classic but very robust Arduino Atmega328 running at 16Mhz... 

3689431240?profile=original

 

 

MPU-6000 uses SPI for max performance. Did I say user configurable low pass filtering and scale range? Hell yeah... and here are some of the other specs:

  • Digital-output of 6 or 9-axis MotionFusion data in rotation matrix, quaternion, Euler Angle, or raw data format
  • Tri-Axis angular rate sensor (gyro) with a sensitivity up to 131 LSBs/dps and a full-scale range of ±250, ±500, ±1000, and ±2000dps
  • Tri-Axis accelerometer with a programmable full scale range of ±2g, ±4g, ±8g and ±16g
  • Reduced settling effects and sensor drift by elimination of board-level cross-axis alignment errors between accelerometers and gyroscopes
  • Digital Motion Processing™ (DMP™) engine offloads complex MotionFusion, sensor timing synchronization and gesture detection
  • MotionApps™ Platform support for Android, Linux, and Windows, with Windows Phone platforms on the roadmap
  • Digital-output temperature sensor
  • Digital input on FSYNC pin to support video Electronic Image Stabilization and GPS
  • Programmable interrupt supports gesture recognition, panning, zooming, scrolling, free fall interrupt, high-G interrupt, zero-motion detection, tap detection, and shake detection
  • Full Chip Idle Mode Supply Current: 5µA
  • On-chip timing generator with ±1% frequency variation over full temperature range
  • User self test
  • 10,000g shock tolerant

3689431141?profile=original

Other ArduIMU V3 features:

  • Pin compatible with ArduIMU V2.
  • Breadboard compatible (ArduIMU V2 was not!).
  • Cheaper price (around $60 $78dlls, not confirmed).. 
  • Smaller size (1.5" x 1.0").
  • Atmega CPU has more % available for other tasks.  
  • The 6 analog pins are now available!
  • Arduino compatible and open source.
  • 3 status LED's (RGB).
  • I2c port with 3.3V translation.   
  • GPS port with FTDI autoswitch. 
  • This device is suitable for ANY application from rockets to simple movement detection.

 

3689431253?profile=original

 

The release date is estimated to be in two-three weeks. We are waiting for the final big batch and finish some code. We will take pre-orders soon.   

 

We also cooking BIG new goodies that will be announced also this month... =) 

 

[UPDATE 11/15/2011]: You can buy it now check this post: http://diydrones.com/profiles/blogs/arduimu-v3-ready-for-pre-order-and-check-our-new-color

 

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • 3D Robotics

    Fab, there's no way we can support 328s anymore--it's way too much work to strip out most of the ArduPlane and ArduCopter features, document the strip-outs, modify the Mission Planner and all the other GCSs so they can deal with the strip-out, handle the tech support from people who are confused by the different versions, etc.

    It would be many man-months of work, halting development for the current boards and their thousands of users. And for what? By this time next year, I expect ARM-powered APM boards with huge power and memory will cost little more than the new ArduIMU. It makes no sense for the dev team to waste its time looking backwards when there's so much more to be gained by looking forward. 

     

    But hey, this is an open source project. If you want to put in the time to do it yourself and support it, that would be awesome. It just doesn't make sense for the dev team.

  • @Doug it seems the program is just going to get bigger and bigger with so many variations. I think the new arduino 1.0 supports command line switches for compilation. So maybe start shipping semi-compiled files, maybe some source, and produce hex files on the fly. Program size might then fit smaller chips.

  • Developer

    Scott, the feature set on ArduPlane has so far overgrown the original ArduPilot board (328) that it is a lost cause.  ArduPlane's current development trunk is now at the point where you have to turn some features off just to get it on a 1280....

  • @Jordi - Any idea on timeline for software? ArduPlane support for an original ArduPilot board?

  • Moderator

    Sweet!   "10,000g shock tolerant"... do you have a video of you testing this? :-)

  • Isn't this board enough for a full APM-lite autopilot? No airspeed sensor nor baro altitude, but still plenty (along with a separate GPS) to provide the standard APM functionality. I like the price, size, and weight over the MegaAPM. Of course, development of the software as a branch off of APM would require dev team agreement.

  • @ThomasB

    I agree with you, it should not be a problem :P. I only highlighting that the chip is "factory calibrated" over temperature xD, and the datasheet claims that is not needed any extra calibration.

    But the linear acceleration could be a problem for aerobatics for instance. The chip is new and there is not enough performance curves o gyros by now :P. But for 49$ / unit (6DoF + attitude algorithms), I will try it out.

  • Oh, I have to make a correction about MotionFusion HMC5883L support.

    Just found this an Invensense site:

    "We do support HMC5883L in our MotionApps software (for handsets and tablets) that runs on Application processor. We don’t have a target date for supporting HMC5883L in the Embedded MotionApps software release."

    So it seems they will support it  (some day :)

  • @ I do not know how well works the attitude algorithm embedded

    As I read in the Invensense developer forum the "motion fusion" software doesn't support the HMC5883L.

  • The "ZRO Variation Over Temperature" with +-20º/sec should be no big problem,

    there is a temperature sensor on it, sampling automatically the temp in some registers parallel to acc and gyro data, so you can correct this 'drift' per software :)

     

This reply was deleted.