IMU320
 

I saw this and I thought it could be really useful in auto:

http://www.euroinvestor.co.uk/news/s...px?id=10263804

PS Does anyone know the exact cost for one of these?

- Alex

Re: IMU320
 

Never used that particular one, but I have used some of Microstrain's units. (If I remember correctly, they were worth a couple thousand dollars about 5 years ago.)

They're similar internally to the sensor boards we use in FRC, with the exception that they have an on-board processor providing significant built-in filtering and communications (via RS-232, and USB on the newer ones). You're paying for the filtering and calibration when you get one of these. In an application like a UAV, the extra cost is justified, because your AHRS sensor makes your autopilot work—you can't afford to have spurious readings.

Re: IMU320
 

Do you have to integrate all the values into angle and position values yourself??

Re: IMU320
 

It depends on the unit (I assume you are curious about IMU's in general, not just the one you linked). Most cheap units are essentially just all of the sensors on one board, some just simply put the sensors on one board then pull out the data connections for you. The next level offer all of the sensors on one board and stream the data through one communication standards. More expensive units handle the integration and filtering and allow you to poll for information like velocity and position.

Check out SparkFun, they have some imu's available at various price points. http://code.google.com/p/ardu-imu/wiki/Theory might be worth reading, it is a pretty good solution at a relatively low price.

Re: IMU320
 

Can the on you linked work without a gps?

Re: IMU320
 

It can work but you will see significant drift, in order to correct for drift you are going to need speed measurements. The link I posted is for drones so the GPS can provide X,Y and Z location (and therefore speed) measurements. You could get simmilar results with encoders on free rolling wheels on the robot in both the x and y direction (to do it right you would also need a magnetometer to establish direction but for the length of autonomous period you could get accurate enough results without it).