The prediction quantities were indeed zero, as they get discretized by multiplying with the sampling time. You're right that some discretized part of the update

The prediction quantities were indeed zero, as they get discretized by multiplying with the sampling time. You're right that some discretized part of the update quantities were also zero, resulting in an identity 'A' matrix. I'll correct it, thanks!