Sensor fusion
Also: data fusion
Combining complementary measurements into one estimate while retaining uncertainty and disagreement.
Fusion requires compatible quantities, frames, timestamps, calibration, noise models, and correlation assumptions.
Example Encoders provide smooth short-term motion while camera fixes reduce long-term drift.
Common misconception Averaging more sensors does not guarantee better accuracy when errors are biased or correlated.