System and method for state estimation in a noisy machine-learning environment

Abstract

A system and method for estimating a system state. The method includes constructing a first estimate of a system state at a first time including a first covariance matrix describing an accuracy of the first estimate. A second estimate of the state is constructed at a second time, after the first time, including a second covariance matrix. A value of a characteristic of the system state is measured at the second time and the second estimate of the system state and the second covariance matrix are adjusted based on the value of the characteristic. A third estimate of the system state is constructed at a third time, before the second time, including a third covariance matrix describing an accuracy of the third estimate. A fourth estimate of the system state is constructed at a fourth time being after the second time.

1 . A method, comprising:
 constructing a first estimate of a state of a system at a first time including a first covariance matrix describing an accuracy of said first estimate;   constructing a second estimate of said state of said system at a second time being after said first time including a second covariance matrix describing an accuracy of said second estimate employing a dynamic model of said state of said system;   measuring a value of a characteristic of said state of said system at said second time;   adjusting said second estimate of said state of said system and said second covariance matrix based on said value of said characteristic;   constructing a third estimate of said state of said system at a third time being before said second time including a third covariance matrix describing an accuracy of said third estimate employing said dynamic model of said state of said system; and,   constructing a fourth estimate of said state of said system at a fourth time being after said second time from said second estimate.   
     
     
         2 . The method recited in  claim 1 , further comprising altering said dynamic model in response to said value of said characteristic. 
     
     
         3 . The method recited in  claim 1 , further comprising reporting said state of said system based on said fourth estimate. 
     
     
         4 . The method recited in  claim 1 , further comprising constructing a fifth estimate of said state of said system at a fifth time being after said second time from said second estimate. 
     
     
         5 . The method recited in  claim 1 , wherein said dynamic model is a linear dynamic model with constant coefficients. 
     
     
         6 . The method recited in  claim 1 , wherein said constructing said first estimate and constructing said second estimate are performed by a Kalman filter. 
     
     
         7 . The method recited in  claim 1 , further comprising altering said state of said system based on said fourth estimate. 
     
     
         8 . The method recited in  claim 1 , wherein said measuring said value of said characteristic further comprising making a plurality of independent measurements characterized by a diagonal measurement covariance matrix. 
     
     
         9 . The method recited in  claim 1 , wherein said dynamic model comprises a matrix with coefficients that describes a temporal evolution of said state of said system. 
     
     
         10 . The method recited in  claim 1 , wherein said fourth time is on a different time scale from said first, second and third times. 
     
     
         11 . An apparatus operable to construct the state of a system in a noisy measurement environment, comprising:
 processing circuitry coupled to a memory, configured to:   construct a first estimate of a state of a system at a first time including a first covariance matrix describing an accuracy of said first estimate;   construct a second estimate of said state of said system at a second time being after said first time including a second covariance matrix describing an accuracy of said second estimate employing a dynamic model of said state of said system;   measure a value of a characteristic of said state of said system at said second time;   adjust said second estimate of said state of said system and said second covariance matrix based on said value of said characteristic;   construct a third estimate of said state of said system at a third time being before said second time including a third covariance matrix describing an accuracy of said third estimate employing said dynamic model of said state of said system; and   construct a fourth estimate of said state of said system at a fourth time being after said second time from said second estimate.   
     
     
         12 . The apparatus recited in  claim 11 , wherein said processing circuitry is further configured to alter said dynamic model in response to said value of said characteristic. 
     
     
         13 . The apparatus recited in  claim 11 , wherein said processing circuitry is further configured to report said state of said system based on said fourth estimate. 
     
     
         14 . The apparatus recited in  claim 11  wherein said processing circuitry is further configured to construct a fifth estimate of said state of said system at a fifth time being after said second time from said second estimate. 
     
     
         15 . The apparatus recited in  claim 11  wherein said dynamic model is a linear dynamic model with constant coefficients. 
     
     
         16 . The apparatus recited in  claim 11  wherein said constructing said first estimate and constructing said second estimate are performed by a Kalman filter. 
     
     
         17 . The apparatus recited in  claim 11  wherein said processing circuitry is further configured to alter said state of said system based on said fourth estimate. 
     
     
         18 . The apparatus recited in  claim 11  wherein said measuring said value of said characteristic further comprises making a plurality of independent measurements characterized by a diagonal measurement covariance matrix. 
     
     
         19 . The apparatus recited in  claim 11  wherein said dynamic model comprises a matrix with coefficients that describes a temporal evolution of said state of said system. 
     
     
         20 . The apparatus recited in  claim 11  wherein said fourth time is on a different time scale from said first, second and third times.