An ensemble of motion sensors is tested under known conditions to automatically ascertain instrument biases, which are modeled as autoregressive-moving-average (ARMA) processes in order to construct a Kalman filter. The calibration includes motion profiles, temperature profiles and vibration profiles that are operationally significant, i.e., designed by means of covariance analysis or other means to maximize, or at least improve, the observability of the calibration model's structure and coefficients relevant to the prospective application of each sensor.

An autonomous vehicle including a Kalman filter error recovery system is disclosed. The Kalman filter error recovery system includes at least one processor and at least one memory storing instructions, which, when executed by the at least one processor, cause the Kalman filter error recovery system to perform operations including increasing eigenvalues of a covariance matrix to adjust probability distribution of a state vector error due to unmodelled process noise in measurements from one or more position sensors, and returning the state covariance to a diagonal state to perform a dynamic covariance reset.

A method performed by at least one processor of an acoustic howling suppression (AHS) system includes receiving, from an input source device, an audio signal. The method includes refining one or more parameters of a Kalman filter based on one or more neural networks. The method includes filtering the audio signal using the Kalman filter with the one or more refined parameters of the Kalman filter to reduce acoustic howling included in the audio signal.

A traction battery is charged according to output, indicative of power capability of the traction battery or state of charge of the traction battery, from a filter that uses battery equivalent circuit model parameters that depend on a gain factor selected according to a learned capacity and an ohmic resistance of the traction battery.

An analysis device includes: an estimation unit that obtains a current state variable by state estimation based on a state variable for first sequence data generated by applying a first conversion to observation data, a state space model, and each element included in the first sequence data; an observation unit that generates second sequence data by applying an observation process to the current state variable; an elaboration unit that generates third sequence data by performing an elaboration process of correcting each element included in the second sequence data so that a value of each cell of the estimated observation data does not become a negative value and the number of cells having a value other than 0 included in the estimated observation data is suppressed; and a second conversion unit that generates the estimated observation data by applying a second conversion to the third sequence data.

A modified Kalman filter may include one or more neural networks to augment or replace components of the Kalman filter in such a way that the human interpretability of the filter's inner functions is preserved. The neural networks may include a neural network to account for bias in measurement data, a neural network to account for unknown controls in predicting a state of an object, a neural network ensemble that is trained differently based on different sensor data, a neural network for determining the Kalman gain, and/or a set of Kalman filters including various neural networks that determine independent estimated states, which may be fused using Bayesian fusion to determine a final estimated state.