Radar System Analysis, Design and Simulation

8.2: Derivation of Kalman Filter Equations

8.2 Derivation of Kalman Filter Equations

Refer to Figure 8.3. A passenger airliner has a smooth fight plan. The pilot plans to fly at a constant velocity at a constant altitude on a straight path. Due to non-constant engine thrust and nonuniform atmospheric conditions, the passenger airliner would traverse at varying velocities and an off-straight-line path.


Figure 8.3: Flight trajectory

The position of the aircraft can be expressed in the Cartesian coordinate system (CCS) as follows. The overhead single dot ( ?) is velocity and two dots ( ??) acceleration. T is the discrete time measurement interval in seconds.


Recast the above Newtonian motion equations into a more compact form with a x and a y, representing accelerations in x and y axis.


Equation (8.7) is still cumbersome. Can we write it the following vector-matrix form?


where


Equation (8.8) is the first of Kalman's seven equations and is called the state equation (the state of, or the position and the motion of, the passenger airliner).

The disturbance w k is assumed to be distributed as Gaussian with mean zero and variance ? 2. The covariance matrix Q k is given by,


We assumed that the accidental acceleration perturbations a x and a y at each measurement are completely uncorrelated and are small in magnitude, say 0.5g, uniformly distributed. (1g = 9.8 m/sec 2). (See Figure 8.4.) The expectation E{a xa y} is identically zero, for a x

UNLIMITED FREE
ACCESS
TO THE WORLD'S BEST IDEAS

SUBMIT
Already a GlobalSpec user? Log in.

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.

Customize Your GlobalSpec Experience

Category: IC Electronic Filters
Finish!
Privacy Policy

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.