It does sensor fusion. The google term you want is "optimal estimation".
A Kalman filter isnt really a filter. It keeps a state estimate of a system. The system state is updated on a regular basis using control inputs. That runs open loop, with a model of the drift gradually reducing your confidence in the open loop estimate. As sensor inputs arrive, they are incorporated in the state to refine the state estimate. The sensor noise model is your confidence in the measurement.
The Baysian arithmetic comes into play when incorporating measurements. The confidence in current estimate and confidence sensor are used to decide how much of each to average into the new estimate.
Ah, but it is a filter - it filters out high frequency noise. One interpretation of the Kalman filter is a variable cutoff low-pass filter. The cutoff is determined by the Kalman gain. Note the similarity between the Kalman state update equation and a first-order low-pass IIR filter:
Well, yeah. It has a smoothing effect on the sensor data. But you don't use it because you want a filter. You use it because you want to build a controller.
A Kalman filter isnt really a filter. It keeps a state estimate of a system. The system state is updated on a regular basis using control inputs. That runs open loop, with a model of the drift gradually reducing your confidence in the open loop estimate. As sensor inputs arrive, they are incorporated in the state to refine the state estimate. The sensor noise model is your confidence in the measurement.
The Baysian arithmetic comes into play when incorporating measurements. The confidence in current estimate and confidence sensor are used to decide how much of each to average into the new estimate.