Processing method for capturing movement of an articulated structure
A method of obtaining simulated parameters ( pos(t) , vit(t) , acc(t) , par(t) ) able to characterize the movement of an articulated structure provided with sensors, characterized in that the method comprises the following steps: calculating, from estimated movement state parameters of the structure, estimated measurement data ( H(t) , γ(t) ), each estimated measurement data item corresponding to a measurement delivered by a sensor, difference between the measurements delivered by the sensors and the estimated measurement data that correspond to them, global mathematical processing of the observer type of the data issuing from the difference in order to obtain at least one estimated difference for an estimated movement state parameter, and adding the estimated difference for the estimated movement state parameter and the estimated movement state parameter that corresponds to it in order to form a simulated parameter.
1. A method for reproducing a movement, in a reference frame, of an articulated structure consisting of a plurality of segments articulated with respect to each other, at least two of the segments each being provided with a measurement sensor, said method comprising:
reconstructing state parameters of said articulated structure, said state parameters describing the movement of said articulated structure;
using a prediction model to provide an estimation of the state parameters at time t as a function of state parameters at time t−1;
providing estimations of measurements of each measurement sensor using said estimation of the state parameters at time t and pre-stored sensor parameter data including drift, offset and sensitivity for each measurement sensor;
providing estimated differences of said state parameters, using a minimization algorithm, which minimizes an error between said estimations of the measurements generated using the estimation of the state parameters at time t and the pre-stored sensor parameter data and real measurement data; and
adding said estimated differences of the state parameters to said estimation of state parameters to provide updated state parameters at time t,
wherein estimations of measurements of at least one measurement sensor related to a segment include cross terms involving at least one state parameter of a segment different from the segment that includes said measurement sensor, said minimization algorithm processing simultaneously the state parameters and the cross terms.
2. The method according to claim 1 , wherein said estimations of measurements is carried out using a function h such that:
m k =h ( l k )+ t k ,
where:
k=index of iteration,
m k =set of estimated measurements,
l k =set of state parameters,
t k =noise term, and
said function h including said cross terms.
3. The method according to claim 2 , wherein said state parameters include position, velocity and acceleration of each segment.
4. The method according to claim 1 , wherein said minimization algorithm is extended Kalman filtering.