search

CathIAS / TLIO

Tight Learned Inertial Odometry
Other
289 stars 70 forks source link

Measurement update #23

Closed ccwss1232 closed 3 years ago

ccwss1232 commented 3 years ago

Thank you for your excellent paper. After reading your paper, I have two questions: 1. How does the updated measurement function H (x) come from? Can I interpret it as a pseudo-measurement? 2. How was the partial derivative of Formula (15) calculated? Could you give the specific derivation steps?

best wishes

CathIAS commented 3 years ago

Hi! Thanks for your questions.

  1. The measurement matrix H comes from the linear derivative of the measurement against the state, and from equation (14), there are only non-zero values in the matrix in three blocks corresponding to the rotation and position in i and j. It is not a pseudo measurement and the actual measurement is the network output displacement between state i and state j.
  2. I attached an appendix file with the detailed derivations to this question. tlio-appendix.pdf
ccwss1232 commented 3 years ago

Hello, author, thank you for your reply. I still have some details to ask you. Do you have QQ and can you add me as a friend? 1102165192