Hi. I have read in several articles that it is possible to improve the performance of the Kalman Filter using neural networks to dynamically change the filter model, thus adapting it to the non-modeled variations of the system with greater efficiency.
I have the Kalman filter working correctly, but I don't know how to incorporate the information provided by the neural networks, nor with what parameters to train them.
I am a beginner in the subject and would appreciate some advice on how to implement this at a simulation level in matlab.
Hi, could you share the information about some of the references you have read to get an idea about what kind of NN - Kalman filter structure you have in mind? Then the answers could be more precise and based on your previous knowledge.
Dear Alberto,
I suggest you to see links and attached files on topic.
Article Dynamic modeling of neural responses in fMRI using cubature ...
Article Control Loop Sensor Calibration Using Neural Networks for Ro...
Article Augmented Kalman filter and map matching for 3D RISS/GPS int...
https://www.frontiersin.org/articles/10.3389/fams.2018.00046/full
https://www.seas.harvard.edu/courses/cs281/papers/unscented.pdf
Article Statistical modelling of cell movement
Best regards
Thanks for
your help Jorge Humberto Urrea-Quintero and Mohamed-Mourad Lafifi.
I am trying to duplicate the results of this article:
Neural network aided adaptive unscented kalman filter for nonlinear state estimation
https://www.google.com/url?q=https://ieeexplore.ieee.org/document/1642720&sa=U&ved=2ahUKEwjakMLXuf7nAhUQknIEHV6iD6oQFjADegQIBxAB&usg=AOvVaw1aj-Q3YXtLjmNSMrWtlnpF
I have some theoretical knowledge about the multilayer perceptron
As i said before, I am new to this field, and I appreciate the opinion of the more experienced researchers.
Dear Alberto, have you ever seen this?
https://www.mathworks.com/matlabcentral/fileexchange/18355-neural-network-training-using-the-unscented-kalman-filter
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Control Theory
- Control Systems