Here are some popular state estimation algorithms. These algorithms fuse accelerometer, gyroscope, and sometimes GPS data to determine position. Because of the nature of these algorithms in embedded systems, the quality of sensor data affects how well your system will perform. You typically need an unbiased measurement from the integral of a biased sensor to properly estimate its state. Therefore you can estimate accelerometer and gyroscope bias with GPS measurements of position and velocity.
[[Kalman Filter]]
[[Bias Instability]] – most of the algorithms depend on a model with a constant bias.
[[Extended Kalman Filter]]
[[Line Guidance]] – needs state estimation
[[State-Space model with Sensor Bias]] – you need an un-biased measurement of the integral of a biased sensor to properly estimate it
[[State-Space Estimators]]
[[Nonlinear State Estimators]]
[[H-Infinity]] – computes gains for estimator
- GyroNoiseAllan2021
- bevlyMECH4420Lectureb