The invention relates to a self-adaptive longitudinal vehicle speed
estimation method, which comprises the following steps of: establishing detailed vehicle
dynamic models, including a vehicle longitudinal dynamic model and a wheel speed dynamic model; a Sage-Husa self-adaptive
unscented Kalman filtering algorithm is combined with a
vehicle dynamics model, a state equation and an observation equation are designed, the longitudinal vehicle speed of a vehicle serves as a
state variable, and observation vectors are constructed through data collected by a
wheel speed sensor, an acceleration sensor and the like;
divergence calculation is introduced for detecting and correcting
divergence phenomena in the filtering process; and on the basis of a self-adaptive mechanism of a Sage-Husa
algorithm, the filtering
gain is adjusted in real time, and the vehicle speed
estimation process is optimized. The improved Sage-Husa
adaptive filter can dynamically adjust the filter parameters according to the change of the
system state and the difference of
noise, detects transient disturbance in combination with
divergence calculation, reduces the influence of the transient disturbance on longitudinal vehicle speed
estimation, and effectively improves the estimation precision.