robotics

[SteveBiomed]

Hello,

I wish to use LabView to host algorithms for determining 3D position from a 3-axis accelerometer and 3-axis gyroscope. The problem is to write the algorithms for real time position determination, given the 6 sensor inputs, and initial conditions, which may be either I2C or (more usefully) strings of numbers acquired by DDE. Could somebody kindly point me to the LV module which contains the vis which would do this without having to write the code. Thank you.

[S_Kamath]

I am not aware of any VIs that do the entire job for you. However LabVIEW Robotics Module has drivers for many sensors and may be able to get data from your device. LabVIEW Control Design and Simulation Module has continuous and discrete Kalman Filters and Extended Kalman Filters. Kalman filters are usually the method used for sensor fusion and state estimation, but you will have to do some work to apply them to your situation.

[Ben]

Yes, you do have to wrie the code yourself.

It can be done but has to be developed to do what you have described.

Ben

[SteveBiomed]

Thanks; the drivers appear to be for physically larger sensors than the MEMS ones I am proposing to use, so I will try to write my own algorithms; perhaps this is fairly straightforward using equation blocks and shift registers. Another post I noticed on a similar problem had the suggestion of using the LabView RT (real time?) module and point-by-point toolset; would this simplify anything? I also looked at the inverse kinematic vi but this does not appear to help.

[S_Kamath]

I would separate the problem into 2 parts:

1) Getting data from the devices. If there is no existing driver for you device then try to modify an existing driver for a similar device. There are drivers for I2C devices as well as IMU devices that may be suitable templates.

2) Getting the 3D position and orientation from the two sensors: The common practice is to use a Kalman Filter with an appropriate model. Control Design and Simulation Module contains the Kalman Filter function.

If you prefer you can use the point by point math functions to implement your own Kalman filter. This may be helpful if you have the full KF equations and don't want to rederive the model separately. The Inverse Kinematics in the Robotics module is meant for robotic arms and is not related to your task.