Method for determining the navigation speed of a carrier and hybridization device

1. Method for determining a navigation speed of a carrier by a hybridization device ( 1 ) including a Kalman filter ( 3 ) developing a hybrid navigation speed (V ref ) from inertial measurements calculated by a virtual platform ( 2 ) estimating a speed of the carrier ({circumflex over (V)}) and frequency measurements outputted by a Doppler radar ( 4 ) having a multi-lobe antenna built therein such that a Doppler frequency (F 1 , F 2 , F 3 , F 4 ) is associated with each of the lobes, wherein said method includes the following steps:

estimating, from the estimated speed ({circumflex over (V)}), frequencies ({circumflex over (F)} 1 , {circumflex over (F)} 2 , {circumflex over (F)} 3 , {circumflex over (F)} 4 ) corresponding to each of the Doppler frequencies (F 1 , F 2 , F 3 , F 4 ) outputted by the radar,

determining a group of observations where each observation represents an estimation deviation on the Doppler frequency associated with a lobe,

calculating, by the Kalman filter ( 3 ), a speed correction (ΔV) to be applied to the estimated speed ({circumflex over (V)}) on the basis of the group of observations.

2. Method according to claim 1 , in which each observation is determined by subtraction of the Doppler frequency (F 1 , F 2 , F 3 , F 4 ) associated with the lobe outputted by the Doppler radar and the estimated frequency ({circumflex over (F)} 1 , {circumflex over (F)} 2 , {circumflex over (F)} 3 , {circumflex over (F)} 4 ) corresponding to the lobe.

3. Method according to claim 1 or claim 2 , moreover including a step of pre-filtering of the group of observations before the step of calculating a speed correction by the Kalman filter ( 3 ).

4. Method according to claim 1 or claim 2 , moreover including a step of calculating, by the Kalman filter ( 3 ), a model correction (ΔK) to be applied to a physical model linking the estimated speed ({circumflex over (V)}) to the estimated frequencies ({circumflex over (F)} 1 , {circumflex over (F)} 2 , {circumflex over (F)} 3 , {circumflex over (F)} 4 ).

5. Method according to claim 1 or claim 2 , moreover including calculating, by the Kalman filter ( 3 ), a correction to be applied to an error model of the inertial unit.

6. Method according to claim 1 or claim 2 , in which the radar outputs successively the Doppler frequencies (F 1 , F 2 , F 3 , F 4 ) associated with each of the lobes according to a calculation period (T), the observations of the group of observations then being determined during successive calculation cycles.

7. Hybridization device ( 1 ) including a virtual platform ( 2 ) outputting inertial measurements, a Doppler radar ( 4 ) having a multi-lobe antenna built therein and outputting a Doppler frequency (F 1 , F 2 , F 3 , F 4 ) associated with each of the lobes, wherein said device includes:

an estimation module ( 5 ) configured to estimate, from a speed estimated ({circumflex over (V)}) by the virtual platform ( 2 ), frequency estimations (P 1 , P 2 , P 3 , P 4 ) corresponding to each of the Doppler frequencies outputted by the radar ( 2 ), and

a Kalman filter ( 3 ) intended to calculate a speed correction (ΔV) as a function of the frequencies estimated by the estimation module ( 5 ) and Doppler frequencies outputted by the Doppler radar.