3D Robotics

New all-in-one gyro/accel/mag chip from Invensense

"The MPU-6000 is the world’s first integrated 6-axis motion processor that eliminates gyro/accel cross-axis misalignment and offers dramatic package size reduction compared with multi-chip solutions. The MPU-6000 has an embedded 3-axis gyroscope, 3-axis accelerometer, and Digital Motion Processor™ (DMP™) hardware accelerator engine with a secondary I2C port that interfaces to third party accelerometer, compass, or other sensors to deliver a complete 9-axis sensor fusion output to its primary I2C port as a single data stream for the application.

The InvenSense Motion Processing Library abstracts motion-based complexities, offloads sensor management from the operating system and provides a structured set of APIs for application development.

For precision tracking of both fast and slow motions, the part features a user-programmable gyro full-scale range of ±250, ±500, ±1000, and ±2000°/sec (dps) and a user-programmable accelerometer full-scale range of ±2g, ±4g, ±8g, and ±16g. Communication of the device is performed using either I2C at up to 400kHz or SPI at up to 20MHz."

I'm not sure about the pricing, but given that it's targeted at game controllers and tablets, it should be pretty cheap.

E-mail me when people leave their comments –

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

Join diydrones


  • For anyone who wants to use an MPU-60X0 chip with a Propeller processor, see my post here for code:


    The code does not use the DMP feature of the MPU-60X0 chips - it just reads the gyro and accel data.  It is compatible with the DCM code that Jason Dorie presents earlier in that same thread.

  • Chris

    In the quest for the ultimate stabilization aerial platform. the gyro plays an important part. Our multi-rotor craft are used for high aerobatic performance that need 300 deg sec. gyros but then we also need the ultimate in slow moving highly stabilized camera platforms. Horses for courses. Invensense make gyros for many applications and in particular a 20 deg/sec gyros that outputs 50 mv/deg./sec as opposed to a 300 deg/sec gyro that outputs only 2 mv./deg/sec. So how about considering the production of an alternative board for just that purpose. Anyone who has tried to stabilize a camera with a 300 deg/sec gyro such as the AD 610 will know that noise oscillation sets-in long before you get anywhere near proper performance. Lets tighten up the copter stabilization as a first priority!!!

  • @^^Johnny5 where did you hear the end of Feb?
  • The chip presently retails around 15$ and should be available end of February 2011.

    Evaluation board from Invensense retails around 70$.

  • Their brochure page implies that this new chip handles the rotation matrix/quaternions onboard the chip, which would take a buttload of math cycles off a main processor. It doesn't say what the update rate is, though. Invensense seems to be good at making their own chips obsolete.
  • Can;t wait to get a sample. Anyone know the expected release date? I would love to scrap my ITG-3200 for this on the next IMU.
  • @James re mag at cm, their target market is cellphones and game controllers - there is no motor, and the baseline they could acheive is only a few centimeters anyway. I fully expect an all-in-one eventually. We already have 3-acc+3-gyro and (separately) 3-acc+3-mag, so it's only a matter of time.
  • Chris, I've had this thought for some time when it comes to programming max rates on a gyro. In the past, specs show the accuracy (fine) drops off as you put in a higher rate. Would it be practical to have two setups where one is Course has a higher rate but lower accuracy and Fine would have higher accuracy but lower rate. Should be able to do this programatically by switching between the two as fine max is approached but not sure that would be worth while. Reason I would like this is I'm just all uptight about exact data and proper control. Not sure it's even needed.
  • Developer
    This chip loops amazing by the way! I like how they solve the mag problem by having an external i2c bus then fusing the signal onboard with hardware. Great idea!
This reply was deleted.