Product documentation — installation, licensing, and integration guides.
SolvFilter
Integration

SolvFilter: Integration

Python API

import solvfilter
import numpy as np
 
cfg = solvfilter.Config(ndim=7, n_obs=3)
x_pred, P_pred = solvfilter.predict(cfg, x, P, dt, f_jacobian)
 
# With custom mean propagator (e.g. SolvSRK):
x_pred, P_pred = solvfilter.predict_with(cfg, x, P, x_propagated)
 
# Update step
x_upd, P_upd = solvfilter.update(cfg, x_pred, P_pred, z, H, R)
 
nees_val = solvfilter.nees(x_upd, P_upd, x_true)
nis_val = solvfilter.nis(innovation, S)

C API

#include "solvfilter.h"
 
solvfilter_predict(&cfg, x, P, dt, &x_pred, P_pred);
solvfilter_update(&cfg, x_pred, P_pred, z, H, R, &x_upd, P_upd);

SolvSRK composition

Use SolvSRK for mean propagation on stiff/noisy dynamics, then pass the propagated state to solvfilter_predict_with() for covariance update.

Related products

Underlying integrator: SolvSRK.