Estimating the values of a variable, Kalman filter use case
Measuring the values of a variable using a sensor is not always possible, or the measurement lacks precision. To compensate for these values, state observers such as the Kalman filter are available.
The Kalman filter, used in a wide range of technological fields from radar to communications to robotics, is an infinite impulse response filter that estimates the states of a dynamic system from a series of incomplete or noisy measurements. In addition to estimating these values and providing the state of the system under consideration, the Kalman filter also provides the error in this estimate.
It is therefore more than worthwhile to subject the Kalman filter to a code optimization phase in order to estimate the potential gains. This is exactly the exercise that was used. We challenged the Matlab-generated Kalman filter in beLow-Explore to see if we could gain in performance.
You’ll find a use case on this infinite impulse response filter in our resource library. Spoiler alert: the application of 2 optimization techniques yielded interesting gains.