Git Product home page Git Product logo

Comments (2)

marcusabate avatar marcusabate commented on June 27, 2024

We haven't tried Kimera on iPhone data but it should work. Your associations between headers is probably right. A large difference in magnitude isn't necessarily concerning since the iPhone may be moving at much slower accelerations than the Euroc drone, however Kimera does require that the gravity readings are present in the imu data so you should see one vector (not necessarily axis-aligned) with values in the 8-9 m/s^2 range.

If you're able to upload your dataset I can take a look. It's possible that the iPhone preprocesses out the gravity vector before publishing. I'd recommend researching if anyone else has tried to use IMU data from the iPhone and if so, what they had to do to make it usable.

from kimera-vio.

HtutLynn avatar HtutLynn commented on June 27, 2024

@marcusabate , Thanks for the reply! Yes, I am trying to make Kimera works with sensor data from iphone and still doesn't work due to error from #218 (comment). For your information, I uploaded the sensor + visual data collected from iphone through Core Motion Framework and ARKit in this GDrive.

Here is the sensor data format from iOS (iphone),

  1. ax, ay, az are accelerometer sensor values.
  2. rx, ry, rz are gyroscope sensor values.
  3. mx, my, mz are magnetometer sensor values.
  4. gx, gy, gz are gravitometer sensor values.

But the values that EUROC dataset provides are,

  1. Linear Acceleration
  2. angular velocity

Here is the transformation process for sensor values from iOS (iphone) to EUROC format,

linear_acceleration_x = (ax+gx) * -9.81
linear_acceleration_y = (ay+gy) * -9.81
linear_acceleration_z = (az+gz) * -9.81

angular_velocity_x = rx
angular_velocity_y = ry
angular_velocity_z = rz

imu_msg = [nano_timestamp,
           angular_velocity_x,
           angular_velocity_y,
           angular_velocity_z,
           linear_acceleration_x,
           linear_acceleration_y,
           linear_acceleration_z
          ]

pose.txt file contains the pose generated from Apple ARKIt not from CoreMotion framework. So for visual data and camera intrinsic, I extract it from the pose.txt. But since the camera is rotated, I have to rotate the extracted frames from video to 90' clockwise to get correct orientation.

But the problem with using data from iphone is that I don't know where I can get the intrinsics of imu sensor from iphone. IMU noise values such as Gyrometer Noise, Accelerometer Noise, GyroWalk and AccWalk. Also I am also lost on how to get transformation matrix from camera to body-frame.

Thanks for the help in advance!

from kimera-vio.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.