thanks @luc, the solution is really helpful and practical!
our main concern is the velocity of the satellite
we want to assume the velocity is always 0 , but this seems lead to an error : cannot normalize a zero norm vector (a PV with a zero velocity will lead to this problem when running the tests)
(so our recent conclusion is to use a value which is close to 0, but not exactly 0, for example 0.0011, but no idea whether the result from this approach is acceptable in terms of accuracy.)
