7

I can only speak for the industry I have worked in (heavy machinery). I have only seen Kalman filters used in practice as observers. Most of the data sources in heavy machinery tend to be quite noisy (pressure or accelerometer sensors). Kalman filters (as compared to simpler Luenberger observers) provide better resilience when faced with high noise levels. ...


6

I know that Luenberger and sliding-mode observers are used in field oriented control algorithms to estimate rotor position and speed of permanent magnet synchronous machines (PMSM). Here is a white paper by TI where they mention the sliding-mode observer. And here is a white paper by Freescale that mentions the Luenberger observer. Here is another from ...


3

They are both different forms of the same thing. I will try to clarify this a bit without a lot of buzzwords, which many people seem to do. In short: The Kalman filter is an optimal observer similar to how LQR is an optimal controller. Both together form an optimal state feedback controller. A bit longer: You already noted that an observer is nothing more ...


2

You can't just "average the two sensors" The acceleration readings are essentially meaningless unless you know their direction, you can't know their direction unless you have the platform orientation, and you can't know the platform orientation from the IMU (unless it has a reliable magnetometer). For the IMU-alone case, your system basically has the ...


2

In observer-based state estimation, the focus is on providing a mathematical proof that the error of state estimation goes to zero (exponentially or in a finite time). For that, the proof is usually based on characterizing of dynamic of error, and therefore the formulation is quite close to stability proof of control systems (with pole placement or through ...


1

First, start with a precise mathematical description of your system. Second, analyze it observability. If the system is observable, of course you can estimate the (augmented) state vector (along with the load as an unknown parameter). In practice, there are always some factors that are often a subject to experimental estimation rather than system ...


1

Using this reference on linear discrete Kalman Filters, it looks like you can apply a standard observability model. Namely, for a linear Kalman Filter system defined as $$ \begin{align*} x_{k+1} &= A x_k + B u_k \\ y_k &= C x_k + D u_k, \end{align*} $$ the system is observable if $M_{obs}$ is full rank, where $M_{obs}$ is defined as: $$ M_{obs} = ...


Only top voted, non community-wiki answers of a minimum length are eligible