We can have the off diagonal terms of a covariance matrix negative also, right. Then in Unscented Kalman Filter algorithm, while calculating sigma points, we need to take the square root of the covariance matrix. This obviously results in complex numbers. How to proceed further?
The condition that is required is that the covariance matrix is positive definite. Square root of a matrix can be found using Cholesky factorisation. It can be performed using chol command in Matlab. If the matrix is positive definite, chol results in a solution; Else, it throws an error.
Dear Nisha A Sathyamoorthy,
For more information about your question i suggest you to see links and attached files in subject.
- Kalman Filter, Extended Kalman Filter, Unscented Kalman Filter
https://medium.com/.../kalman-filter-extended-kalman-filter-unsce...
-Robot Mapping Unscented Kalman Filter
ais.informatik.uni-freiburg.de/teaching/ws12/.../slam05-ukf.pdf
- Unscented Kalman Filtering for Single Camera Based Motion and ...
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3231712/
Best regards
Dear Nisha A Sathyamoorthy, For more information about your question i suggest you to see link and attached files in subject.
http://ieeexplore.ieee.org/abstract/document/1271397/?reload=true
good luck
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix