tilstandsnærmer

A tilstandsnærmer, often translated as state estimator or state observer, is a system used in control engineering to estimate the internal state of a dynamic system. Many real-world systems have states that cannot be directly measured due to sensor limitations or cost. The tilstandsnærmer uses a mathematical model of the system and available output measurements to infer the unmeasurable internal states.

The core principle behind a tilstandsnærmer is to combine the predictions of a system model with actual measurements. The model predicts how the system's states will evolve over time. When a measurement becomes available, it is compared to the model's prediction of that measurement. The difference, or error, is then used to correct the model's estimated states, bringing them closer to the true, but unmeasured, values.

Common types of tilstandsnærmere include the Luenberger observer and the Kalman filter. The Luenberger observer is a deterministic approach, while the Kalman filter is a stochastic approach that accounts for noise in both the system dynamics and the measurements. The choice of tilstandsnærmer depends on the characteristics of the system being controlled, including the presence and nature of noise.

Tilstandsnærmere are crucial components in many control systems, enabling feedback control when full state information is not directly accessible. They are used in a wide range of applications, from automotive systems and robotics to aerospace and process control, allowing for more robust and accurate control strategies.